Welcome back to profspur analysis and a new Premier League season! As I did last year, I like to start things off with a brief accounting of what I got wrong in the prior season. This year, however, I’m also going to add some modest predictions for the 2026-27 season, which will give structure to the next version of this exercise and make things a bit more fun.

I guess I should confess from the outset that I think I got last season mostly right. That is, I was concerned from the second half of last year’s preseason match against PSG that our lack of interest in keeping the ball and controlling the match could be a major problem in the year to come. I also made no secret of the fact that I didn’t think Thomas Frank was the right choice, arguing as early as September 2025 that Frank actually had not made Spurs ‘more defensively solid’; in October, that we weren’t creating enough from open play to compete toward the top of the league; and in November, when we were riding high on early results, that Spurs were riding our luck and we should expect to plummet unless something significantly changes. Picking up on these kinds of things early on made for a depressingly accurate analysis of a season I’d love to forget.

But enough of that. As I said, I’m here to talk about what I got wrong.

1. The big one is that I expected Spurs to finish in the top 6-8 places, certainly not as low as 17th. I figured the summer business (Kudus and Xavi in particular) signaled some ambition and that we had the squad for a much higher finish. Though I was never a fan of Frank at Spurs, I never thought it would be as bad as it was.

2. I expected a lot more of Randal Kolo Muani, defending him throughout the season. In the end, he never really delivered, never found a goalscoring groove. Though he made a better effort toward the end of the season than he ever got credit for, chasing the ball down in the attacking third and causing problems for the opposition, especially under De Zerbi, it was nowhere near the performance I thought he was capable of.

3. I expected a lot more from our attack, believing that Frank’s Brentford had always found ways to score goals, even if not the way I like my club to score them. In particular I thought Spurs would be more effective on the counter and would find the likes of Kudus and Tel in behind much more than we did. Part of my failure here was not recognizing just how much the dearth of midfield passing would impact the squad. Losing both Maddison and Deki for the season was a major blow, as were the matches Cuti Romero missed. Without those three we had very little in the way of ball progression, a problem compounded by Frank’s lack of ideas for possession and progression, forcing everything predictably from Porro to Kudus along the right flank. Our xG shot up under De Zerbi with the same personnel simply because he has an idea of how to play with the ball and get it into dangerous areas.

4. On the whole I thought our summer business left us in decent shape when it came to attacking options. While we were able to play most of the season with players in their actual positions, it broke down in the end, revealing the extent to which we were left short in that department.

5. I thought in September that Brennan Johnson would have a big year for us, given how good he is at running in behind to receive direct, vertical passes. See item (3) above for a partial explanation of why that didn’t happen and we ended up moving him on.

Predictions for the 2026-27 Season

1. An unwritten story of last season was how many of the teams with European competition struggled in the league. I expect that trend to continue this season, which is to say I expect Spurs to capitalize on lack of European football as we did in Ange’s first season at the helm. After what’s already been the most ambitious and impressive transfer window for Spurs since I started following the club in 2007-08, we have depth of quality in defense and midfield (and appear to be working on the front line still). So I expect a 5th place finish at minimum.

2. I think Mathys Tel is going to have a breakout season.

3. I think Tonda Kinsky will also have a breakout season (a little softer as a prediction since it’s kind of already that since he’s been named the first-choice keeper, but I think he’s actually going to excel and become one of the better keepers in the league).

4. Given the transfer window is still open and Spurs are rumored to be in for Savinho and Marmoush, it’s hard to make predictions about too many of our attacking players. But if we end up relying on some combination of Richarlison and Solanke through the middle, I expect they’ll combine for more than 18 goals this season.

I’m just getting this in about an hour and fifteen minutes before opening kickoff. Come on you Spurs!

I grew up in the United States in a time before Major League Soccer, before YouTube, before streaming services enabled widespread fan access to the professional game. Like many other American kids, I started playing soccer from a young age; I was 5 years old when I played on my first organized team in a league.

Back then it was almost impossible to watch a professional soccer game on television in the US, save the World Cup every four years. There were no video tutorials on how to dribble or set your feet as a defender, no channels dedicated to drills, no highlight compilations of professional stars. I had a beaten-up soccer ball with the white and black pentagonal panels peeling off and I took it to the back of my school, where the outer wall of the gymnasium was broad and windowless, and kicked it against that wall and back for hours every day. That’s how I learned to do everything I know how to do with the ball: two-footedness, touch, passing weight, reaction time, controlling with different parts of the body, and striking technique.

Where I differed from the American norm, I suppose, was that I didn’t stop playing the game or taking it seriously as I grew up. Even in the 80s, I didn’t think of soccer as a kids’ game or a girls’ game, as were the national norms. It didn’t help that I grew up in a working-class city known for its steel production, its gruff idea of masculinity, and its (American) football tradition, so my continuing to play soccer seriously through high school was taken not merely as a difference, but as a defect. Where I’m from, men of my generation who played the game will immediately recognize epithets like ‘soccer f*g’ and derisive tropes about male soccer players, such was the US sports media’s penchant for focusing, during World Cup coverage, on European and Latin American players dramatically writhing on the ground after what was presumed light contact.

The prevailing attitude back then was also that if you’re a man, you only played soccer because you couldn’t make it in other American sports. That was never the case for me. While I was probably too slight for American football (if you’ll excuse my bragging a little bit to make the point), I was a gifted athlete, an all-star baseball player and an all-state track and field athlete in high school, in addition to starting as a midfielder in what was at the time one of the top-ranked large-school soccer teams in the US. For both soccer and track and field I was heavily recruited by Division I university programs, ultimately forced to give up soccer at university due to receiving a full athletic scholarship for track and field (and my coach’s insistence that if I risked injury playing soccer, he’d throw me off the track team and I’d lose my scholarship to a university that cost about $30,000 per year at the turn of the millennium. That was something my family and I could not afford).

I offer this detailed background in part to make the point that, if you can believe it, US soccer culture today is a lot better than what it used to be. We have domestic professional leagues that are relatively accessible to fans. We also have widespread coverage of the elite professional leagues, the Premier League in particular, which is so popular now that hundreds of thousands of Americans (myself among them when I’m residing in the US) now wake up at ungodly hours for ‘Premier League Mornings on NBC.’

The rise of on-demand streaming and video-based social media now mean that American kids can not only watch their favorite professional players, but also develop a sense from the pros of what kind of player they want to be. They can learn drills and skills online, pore over statistics and recruitment profiles, learn what pros eat and how they prepare on match day, and connect with fan communities from anywhere. If they happen to live in a place, like I did, that thinks soccer is for kids, girls, and f*gs (I don’t endorse these characterizations or this grouping of them; I’m just being candid about what many people in the US still really think and say about soccer) or that women and girls shouldn’t focus on sports (even soccer!) at all, they can connect with people who think otherwise, who share their interest in and passion for the game. Before even getting into domestic league attendance figures for women’s and men’s soccer, and the tremendous fanfare around the national squads (particularly the women, who are among the world’s best), this is all truly remarkable.

Yet there remains a stubborn streak in US soccer culture that isn’t reducible to the other kinds of reactionary politics (homophobia, misogyny, xenophobia) I’ve alluded to here, a stubborn streak that continues to stall progress. My focus here will be the shortcomings of men’s soccer in the US, given the success of women’s soccer, though it’s worth noting that I think the US women succeed despite, rather than because of the problems I’m about to name. The longstanding idea in the US that soccer is for women certainly helps in terms of recruitment and resources, but I think the US women remain undervalued precisely because the game at large is feminized in the US, thus considered less worthy of high wages or more general respect among sports fans.

US Soccer as US Exceptionalism

If you think about every practical barrier to success in US men’s football—socioeconomic chokepoints to a truly broad and representative recruitment pool; the relative popularity of US professional leagues such as the NFL and NBA as competition; and a fanbase with all the technological tools for access to and knowledge of the world’s game but often no clue about where to look in the first place—they all come down to one thing: US exceptionalism.

US exceptionalism—not to be confused with the actual state of being exceptional—is an ideology that posits the US as a transcendently unique country that, by virtue of its uniqueness, has the right to act and be treated according to its own set of rules. Many people misunderstand US exceptionalism as US exceptionality, but that’s a grave misunderstanding. The ideology of US exceptionalism doesn’t require the US to be exceptional in any way, only that we believe we are and can therefore point to that belief as justification for doing whatever we want.

The doctrine of US exceptionalism means we can bomb anyone to for any reason, whereas if you do it and we don’t like your reason, you’re a ‘rogue state.’ US exceptionalism means WE-BE-LIEVE-THAT-WE-WILL-WIN at soccer, even if we’re playing against France or Brazil, because we have a higher GDP. You get the idea. In fact, I think the outsized role of belief in US soccer culture and fandom is paradoxically toxic.

The core problem with US soccer culture is that it’s entirely constructed with the raw materials of US exceptionalism. Even the coinage ‘Major League Soccer,’ like ‘Major League Baseball,’ was designed to put an American spin on the world’s game. The shaping of US soccer culture by and of US exceptionalist ideas creates myopia, poor incentives, even delusions that keep the ‘USMNT’ (once again, a provincial, exceptionalist way of describing the team; it leaves out the name of the actual sport for which the team is assembled!) uncompetitive and the fan base unnecessarily ignorant of what the game looks like on the world’s stage.

Fans of the game outside of the US implicitly recognize as much when they complain about superficial components of US soccer. As a US national who lives part-time in England, I see that when fans of other clubs are taking a dig at me (or at Spurs via me, an American fan), they resort to mocking what they presume will be my ‘American’ traits, the fact that people in my country call penalties ‘PKs,’ or have stupid chants (like the aforementioned WE-BE-LIEVE chant). They’ll pronounce the word ‘soccer’ in an exaggerated, ‘American’ way—’SAHH-curr;’ ‘SAW-curr’—which, it’s worth noting, always has a deep-southern, rural Appalachian, or rural midwestern inflection, whether or not they know it, because what they mean to convey is ‘you’re a dumb yokel.’

Because US soccer culture is built on US exceptionalism, and we construct far-reaching marketing campaigns around the US national soccer teams depicting people arguing with the likes of David Beckham over whether it should be called ‘soccer’ or ‘football,’ we only invite such national-exceptionalist engagements with the game. In other words, I, too, find these things annoying, not because there’s anything inherently wrong with national lingo (I call them ‘penalties’ and am otherwise deeply acculturated by UK football fan culture, notwithstanding my nationality) but because this particular national lingo is, once again, just a reminder of the broader observation I’m making here about US exceptionalism. US fans have now had access to the world’s game for generations, but insist on framing everything about it in our own, provincial terms.

Further, the more serious forms of what I’m talking about have direct negative impact on US players and fans and our collective aspirations for soccer.

Pay to Play

Take, for example, the fact that football is a relatively inexpensive and socioeconomically accessible sport throughout the world. Many of the world’s greatest players grew up in slums. And many of the world’s most successful national soccer programs have built recruitment and development infrastructures around this fact. If you’re a talented Brazilian or French or English player, from a disadvantaged background, you have a much better chance at being noticed in the system than you do if you’re an American player in the US.

In the US, because the youth development setup is still heavily filtered through high school recruitment to Division I university competition—even with the MLS academy system in place—a lot of your opportunity to get plugged into a recruitment pipeline is dependent on your school district and region. In the US, the quality of even public schools, and the level of sporting and extracurricular opportunities they offer, is in turn heavily dependent on local property value, since local property taxes in the district are the primary funding source for schools.

This means, for example, that if you’re from a place like Medford, New Jersey—where Leeds attacker Brendan Aaronson is from—which has a median household income of around $160,000 per year, you’re going to have access to local private football academies, specialized coaching, and recruitment personnel who are already looking at your area. If you’re from a less affluent, less white area with less of a regional football reputation, it becomes very difficult to break through with the kind of opportunity Aaronson had when he joined the Philadelphia Union Youth Soccer Academy at age 18.

On top of this, because of the longstanding idea in the US that soccer is for kids, recruitment opportunities and access are highly driven by parents, especially affluent parents, instead of coordinated, national recruitment systems with built-in regional and socioeconomic status sensitivity. As with the ‘college prep’ industry in the US, which takes money from rich families to help coach their offspring on standardized tests and other elements of their university applications to increase their chances of getting into highly selective universities like Harvard or Stanford, US youth soccer has an extensive private recruitment industry in which parents pay recruiters to help place their aspirant soccer stars into MLS academy or university recruitment pipelines. To the side of recruitment costs, club soccer in the US is also very expensive, with annual costs ranging from $5000 - $25,000 per year, exclusive of out-of-pocket costs for travel to games and tournaments.

All of this, by the way, also impacts one’s chances of making it into a Division I university soccer squad. A glance at the top 25 US men’s Division I soccer programs shows a handful of them—UNC, Duke, Stanford, Georgetown, UVA, and Princeton—are highly selective, academically elite universities for which socioeconomic advantage is also a big admissions advantage. The school district funding dynamic I explained earlier also plays a significant role in access to elite higher education, with or without sports. Kids who grow up in rich school districts typically have more higher-level course options in school (AP and IB courses, for example), taught by better paid and more highly sought after teachers with better educational resources. US public education has extreme inequality by district and region, with far-reaching impact on both educational and sporting opportunities for local kids.

To summarize, one big reason US soccer culture is bad is that it’s simply not as socioeconomically and culturally diverse as it should be to attract the very best talent to high level competition, not just the richest or whitest talent. This is not just bad for performance on the pitch, given the way it artificially shrinks the talent pool, but also bad for the culture of the game in the US. It means most Americans continue to think of soccer as a kids’ game driven by helicopter parents—the ‘soccer mom’ trope is alive and well—and implicitly as a sport for rich kids. Another facet of this problem is that, compared with sports like American football and basketball, for which recruitment systems are well developed for finding the best players independent of family socioeconomic status, soccer comes to look amateurish and niche, despite being the most popular and important sport in the world.

A common, US exceptionalist justification for why the US men are not very good, relative to the population and GDP scales of the US, is that soccer is a ‘second choice’ sport, with our best athletes choosing American football, baseball, or basketball. But this rationalization for poor performance withers under the slightest scrutiny. The attributes that make one a candidate for the NBA or NFL have some overlap with elite professional soccer, but not enough to suggest with a straight face that Victor Wanyama and Victor Wembanyama could’ve swapped sports and reached the same heights in an alternate universe of Spurs and Spurs.

That is, the problem is not that the occasional universal athlete, like Bo Jackson or Allen Iverson, is not choosing soccer; elite soccer players would wipe the floor with these other athletes in a soccer game. The problem, rather, is that the soccer recruitment pool is artificially narrowed by financial gatekeeping and the development of soccer players is hampered by the very idea that it’s not a serious sport. US exceptionalist narrative steps in to naturalize such insufficient conditions, so we can all lie to each other about American sports and their athletes being so uniquely good that they’d also be better soccer players than elite soccer players (honestly I can barely type this with a straight face, let alone believe it), and that’s why the ‘USMNT’ sucks ass. Every time someone makes the specious argument that the US men are not good because our best athletes choose other sports—’American’ sports—they’re naturalizing, rationalizing, and apologizing for failure, all in the name of US exceptionalism.

Fight, Courage, Heart, Belief!

Two strands of US exceptionalist thinking come together to form a greater third, like a confluence of minor rivers. One is the myth of the ‘universal athlete’—that the athletes in traditional American sports are universally superior and would, had they chosen soccer, be better at soccer than professional soccer players—as a way of explaining the paradox of the US being an all-powerful country that’s the best at everything but also not even close to the best at soccer. The other is the provincialism of US soccer fan culture, which likewise tries to warp everything about the game at its highest level as played abroad to fit the norms of US professional sports and sporting cultures. From the confluence of these two things, you get a greater third thing, which is reinforced ignorance throughout the system of what it takes to be good at soccer. This manifests in many ways, including:

• The belief that raw athleticism trumps technique, skill, acumen, and tactics in the actual game of soccer, therefore the way you become good at soccer is by being more physical, more aggressive, having more ‘heart’ and ‘fight’ and ‘courage’ and ‘belief’ than the opponent, even if by comparison you are not actually very good. US culture at large values the rough-and-tumble underdog over the virtuoso, so our soccer culture doesn’t tolerate virtuosi. Instead, the mottos and cheers of US soccer are just warmed over versions of US exceptionalist cliché: hard work and graft (in the British sense, I promise!), rags to riches, belief you can overcome anything. I don’t think these are bad values, they’re just a bad formula for building a soccer program that’s internationally competitive at the highest level. They’re also what US fans clamor for and reward most with our support.

• The shaping of ‘Major League Soccer’ in terms of US franchising and professional sports league conventions, putting our premier league at odds with the literal season schedule of most of the world’s elite football leagues (mainly to avoid media competition with the NFL and NBA), and eschewing the tiered + relegation system practiced in the world’s best leagues, which adds jeopardy and competitive verve to every league season. MLS has since voted to align its schedule more closely with that of the top European leagues, a good move that also shows it was possible to do a lot sooner. So, in our domestic league system, we’ve done it all our own way, ignoring the things that make the world’s best leagues so successful, hampering our talent development in the process.

• A privileging of physicality over technique on the pitch and of patriotism over global soccer knowledge in the media, perhaps best exemplified by right-wing US soccer icons Alexi Lalas and Hope Solo, the latter of whom publicly criticized her teammate Megan Rapinoe for kneeling during the US national anthem. This is to say, the commentary around US soccer at the national level is not really much about soccer. It’s more about patriotism, fans and commentators wrapping themselves in the flag (literally and figuratively) and speaking of the MLS to the world as if it were a league the world takes seriously. ‘U-S-A!’ chants can be heard at soccer games as well as Trump rallies. Here again I don’t mean to say it’s inherently bad to support and take pride in your country in international competition, but with men’s soccer it feels more like trolling the world than supporting the team, since there’s such a massive gulf between the swagger of supporters and the quality of the team.

When all of these things combine, you get, in my opinion, a weak soccer culture. As an American who spends a lot of time living abroad, I’m usually quick to defend my people and my country, because I think a lot of ‘American’ stereotypes are pretty silly, and I have the vantage point to say as much as someone with experience living in multiple countries. But it’s really hard to defend the arrogance of US soccer culture, the reluctance to learn from the rest of the world about one of the things the rest of the world does a lot better than we do. Such brash ignorance is not just embarrassing; it actively prevents us from being as good at soccer as we could be, given the size and diversity of our population and the resources at our disposal for creating a top soccer program.

I’ll close with a few words from legendary Spurs coach, the one and only Mauricio Pochettino, who’s now, of course, the head coach for the US men’s soccer team. In a recent Spanish-language interview in which he was asked about US prospects for the upcoming World Cup, Poch lamented the American people’s ‘emotional relationship with the game.’ He went on to observe ‘the kids don’t develop until they are 11, 12, or 13’ and that ‘they take the ball with their hands, first thing,’ as opposed to his childhood in Argentina, where ‘you kick the ball with your feet.’ This is only the case, I think, because US soccer culture is still languishing, still a bit deprived, still proceeding with an inferiority complex in relation to other American sports. The reality is that many American kids do kick the ball with their feet from quite a young age; they just don’t keep doing it often and for long enough to become good, because the US system and US soccer culture aren’t designed for that. Nevertheless, in the same interview, Poch said:

When I met Mr President [Donald Trump] before the draw in Washington, he asked me, ‘Do you think, coach, that we can win?’ I said of course … Why not? Why not! It’s all about your belief. Look at Morocco in Qatar; no one believed. Look at South Korea in Japan. Semi-final, too. There’s plenty of examples: if you put a limit and say maybe, for us, the message is to reach the quarter-final. You don’t go through the group stage. All is possible in football.

Because even if you’re from the great footballing nation of Argentina, if you’re going to be the American coach, you have to look dead in the eye the subpar team you have out there (just like the shit president standing in front of you) and make it all about belief, WE-BE-LIEVE-THAT-WE-WILL-WIN. The trouble is that no one from Argentina, France, England, Germany, or Brazil would ever chant about belief, because no one who truly believes they can win ever feels the need to chant about it.

After a decade with an average Premier League table finish of 4.7—the 5th best in the league and virtually indistinguishable from Arsenal (4th — 4.6) and Manchester United (3rd — 4.5)—Spurs have now finished 17th two seasons in a row. The simple math suggests this is nothing less than a precipitous drop.

To give this a little more perspective, Spurs finished above this season’s champions, Arsenal, for 6 of the past 10 seasons, but 0 of the past 4. In the decade before Spurs plummeted to 17th last season, Spurs qualified for Europe 9 times, including 5 Champions League appearances, which is more than everyone but City, Liverpool, and Chelsea over the same period. Just 3 seasons ago, in the first season under Ange Postecoglou, Spurs were disappointed to have missed out on Champions League qualification by 2 points, finishing 5th. When Spurs sacked Antonio Conte in 2023, they were 7th in the table. Such were the standards for disappointment in the very recent past. Now we are elated to have narrowly escaped relegation.

How did this happen?

The Long Decline Theory is Wrong

One persistent theory I’ve heard a lot this season in football media and among fans—particularly in Spurs circles—is that the chickens are finally coming home to roost after a period of long decline. The rough shape of this argument is that it started with Levy’s failure to back Pochettino in the transfer market after Spurs accelerated to a couple of title-challenging seasons and a Champions League final; that the new stadium became a financial priority over the football side of the operation; that the decision to go after ‘win now’ managers in Mourinho and Conte was a mismatch for the culture of the club, further eroding player confidence and the club’s relationship with the fanbase while letting football operations off the hook for lack of squad investment; that lack of modern, professional recruitment and football operations structure left the club behind our competitors, making it harder to move players on at the right time to get value out of them and making our long-term transfer strategy incoherent; that our constant chopping and changing and sacking managers, often more than once in a season, has created a mismatch between manager play style and player suitability, thus a series of patchwork squads; and that our wage structure and tendency to haggle over relatively small sums in the transfer market has left the side uncompetitive for a while now.

There’s truth to all of these claims, but the problem with the Long Decline Theory as an explanation for the sorry state of Spurs at present is temporal. That is, Tottenham Hotspur Football Club have been making these errors for the better part of a decade and still qualifying for Champions League, finishing in the top 5, and sliding down to 7th or 8th in poor seasons. The Long Decline Theory makes sense, perhaps, as an explanation for why Spurs failed to kick on from regular top-5 finishes and European qualifications to seriously challenging for the Premier League title again like they did in 2016-17. But it makes very little sense as an explanation for why Spurs went from qualifying for Europe in 14 of 15 seasons from 2010-11 to nearly getting relegated in 2025-26.

Long Decline Theory is tempting, I think, because it permits an even, smooth, if false synthesis over a long period, gathering all of our gripes into a narrative of linear descent. But the reality here—by which I mean, again, the simple mathematical reality—is that Spurs’ descent has not been linear at all. It’s looked more like falling off a cliff in 2024-25 than a long, steady tumble down Cooper’s Hill.

I also think Long Decline Theory is tempting because it flatters a number of longstanding grievances against the club among fans. For example, I’d be the first to say Daniel Levy’s risk intolerance in the transfer market and long-term suppression of Spurs’ wage structure were probably the main issues holding Spurs back from taking the last and hardest step from competing for top-4 to competing for league titles. But many fans see in Levy a convenient scapegoat for everything that ails the club, so would like to make sure Levy gets the lion’s share of blame even for the first season after the board sacked him. If you really like blaming Levy for things, you need a Long Decline Theory to counterbalance the fact that, at least by the numbers, Levy’s chairmanship saw Spurs become more competitive on the pitch than in decades prior.

My goal here is not to litigate Levy’s legacy, however. The point here is not to quibble with why people are attracted to Long Decline Theory, only to show that it’s wrong.

It’s wrong, rather straightforwardly, because for the period during which we might compile about 80% of our observations of the ‘long decline,’ Spurs didn’t actually decline at all! In fact, they were remarkably, consistently around the top 5 in the Premier League. The concept of decline expresses a rate of change, typically a slow or steady rate of change. You would never say something like ‘WATCH OUT, YOU’RE ABOUT TO DECLINE OFF THAT CLIFF!!!’

The Cliff Theory of Spurs’ Decline

The challenge to proponents of Long Decline Theory, then, is how do you explain that under the same set of conditions (incoherent squad building, constant manager sacking, uncompetitive wage structure in comparison with the rest of the top quartile of the Premier League) there was no decline at all and then suddenly precipitous drop? I realize the question, framed that way, is kind of begging itself. But it strikes me as pretty difficult, if not impossible, to explain how this ‘long decline’ was basically biding its time, like a latent virus, for 10 years, and then suddenly struck in 2024-25, during which Spurs came down with a mild case of 17th in the league without relegation complications and a severe case of winning a European trophy.

For these reasons at least, we should reject Long Decline Theory on its face and instead try to explain the plain-as-day observation right in front of our faces, which is that the period of decline is two seasons—2024-25 and 2025-26—during which Spurs finished 17th after finishing 5th the season before that.

So here is the argument for the better theory of Spurs’ decline, the Cliff Theory:

1. In case it needs to be said, finishing 17th twice in a row after finishing 5th, 8th, 4th, 7th, 6th, 4th, 3rd, 2nd, 3rd, 5th in the previous 10 seasons is precipitous, not gradual. Observationally there is no decline there; there’s only an inflection point followed by a sharp plummet.

2. It’s the injuries! Spurs have had catastrophic injury crises for two consecutive seasons. I’m not talking about losing one or even two key players for a season, or having a period or two of missing 3-5 key first-11 players. No, I’m talking about all of that plus playing a whole 60-game season of domestic and European football with your starting CBs playing 20% of the time, while the rest of the time you’re playing an 18 year old central midfielder and Premier League debutante at CB alongside Radu Dragusin or Ben Davies. I’m talking about playing the final quarter of a season with only 3 fit forwards, all of whom (Tel, Richarlison, and Kolo Muani) are arguably strikers, such that when you’re not playing strikers as wingers, you’re having to rotate them with Conor Gallagher (midfielder) and Djed Spence (right back) on the wing. I’ve already written so many words on why these kinds of injury runs are abnormal and catastrophic, so I won’t rehash all of that here. It’s just not normal for a team with an average league finish of 4.7, which regularly competes in Europe and expects regular top-5 finishes, to leave the squad so short for two whole seasons. Here’s a short proof by negative example: Can you name any other Premier League club that’s attempted to play in the league and in Europe with a teenaged midfielder at center back for 25 games? That plays the the last quarter of the season with 0 fit wingers in the squad so has to play 20-year old league debutante strikers and fullbacks as wingers? How many minutes has Savinho played at CB for City? For how many games in the past two seasons has Areta been forced to use Jurrien Timber on the wing because Saka, Trossard, Martinelli, Madueke, Eze, Dowman, and Nwaneri were all injured at the same time?

3. Two consecutive years of catastrophic injury crisis is an effect of the club’s football operations executives leaving the squad short for two seasons in a row. A common observation that e.g. there is not enough passing range and quality in the team—which I largely agree with—is a luxury problem compared with not being able to consistently field 11 players playing in their actual positions as professional footballers. While I also agree that, as noted above, Spurs’ squad building and management have been poor for a long time now, there have been two acute acts of squad mismanagement over the past two seasons that were the difference not between finishing 5th and 8th, but between finishing 5th-8th and 17th. They both happened during the last two January transfer windows. In 2024-25, the players themselves were openly begging for squad reinforcements in their interviews and press conferences, such was the cycle of injury and overuse and re-injury. During that month alone—which happened to be a transfer window—Spurs played 7 matches without any squad reinforcements made available during the densest, most important run of games of the season to that point. The players they did bring in only came after. Fans will remember how difficult it was to watch during that period, not because of the style of football so much as how painfully obvious it was that the players were shattered and struggling. This past January, under Thomas Frank, with a continued injury crisis, Spurs signed just two players: the Brazilian teenager Souza, who could hardly be useful yet in top-flight competition, and Conor Gallagher, a midfielder who plays the same roles in the same positions with the same profiles as Spurs’ other midfielders—their deepest position in the squad—while Maddison and Kulusevski remained out for the season, while we sold Brennan Johnson, and while our front line was so depleted that by the end of this season we’d have just 3 fit senior attacking players. These failures to move in January only compounded the injuries that have transparently devastated the squad and Spurs’ league results. In either case, this is was not just another instance of a bad transfer window; these were inflection points that caused Spurs effetely to play with inferior squads for the majority of two seasons straight.

4. As I’ve noted, Spurs have been reckless with manager appointments and dismissals for a long time, but the choice to move consecutively from Postecoglou to Frank was uniquely stupid and catastrophic. No one reading this will be surprised that I think Ange was actually a good manager who was hamstrung (ha) by items (2) and (3) above in this list. But that’s not the point I’m making here. Whatever you think of Postecoglou, it’s a fact that he had a special bond with the players, not least of which because he capped the long struggle of his second season with the club with a Europa League victory. We don’t talk enough about the psychological message it sent to the players to sack him cold after arguably the club’s greatest achievement in decades. The message from the club to the players in that act was basically ‘have your little open-top bus parade, but this thing you’ve built together, through thick and thin? We don’t believe in it.’ To then turn around and hire someone, in Frank, whose attitude toward the position from day one was to lower expectations even further, to treat the players and the club like we’ve all gotten too far ahead ourselves and need a cold bucket of water to face, was clearly psychologically devastating to this group of players. And perhaps it would’ve been OK if Frank’s way of playing were enjoyable for them—it wasn’t—or, more importantly, if it were effective in securing results, which it wasn’t either. In pretty much any venture in life, to throw cold water on an emotional triumph and immediately replace it with dour ‘pragmatism’ is a recipe for disaster. It was no different for Tottenham. So here it’s worth saying that of course the long run of manager-merry-go-round has at least one major cost, which is that by constantly sacking and scapegoating managers, you necessarily deprive yourself of the opportunity to test whether the problem is the manger or the squad. But that doesn’t mean all such choices to move on from a manager are equal. I think you can explain poor league play under Postecoglou with recourse to things besides Postecoglou and his tactics; I think it’s harder to do so for Thomas Frank, who (unlike Tudor and de Zerbi this past season) wasn’t forced to play players out of position for long periods of time, for example, and was quickly out of non-league competitions. But the most devastating thing, in my view, was in the sequence of these two managers, the way they embodied not just tactical polarity, but completely different attitudes toward the club and the players and the prospects and ambitions of Tottenham Hotspur. If Spurs were already down after bad league campaign last season, we more or less asked for worse when we decided to follow Ange with Frank. The way the players responded (or not) to Frank this season reflects as much.

5. That leads to the a final point, which is that the many high-level internationals Spurs have in the team—players who are used to playing ambitious football at top clubs—do not want to be treated like they play for Brentford. I don’t at all mean this in a snide way. I mean when you start forcing this level of player to play reactive, timid football, you won’t get the best out of them. Even under Postecoglou, the squad was so thin due to injuries that Spurs had to play a lot of dour, reactive football. ‘Play styles’ are Platonic forms, but only flesh-and-blood players execute the football, and squads are only the players available to play the football, not the team sheet on Transfermarkt. For all the narrative about Ange, I find it absurd to think he simply abandoned his principles as an act of penitence or admission of guilt, as opposed to because his players were physically devastated for lack of cover. When you get used to playing reactive football instead of what I’ve been calling protagonist football—football that’s about controlling the game and forcing your opponents to react to you and not the other way round—it’s harder to get good results out of elite players. Frank and Tudor only compounded that problem by trying to force protagonists to become bit players every week in the opponent’s narrative. The re-animation of even a depleted squad under de Zerbi shows that control—whether through possession, territory, or other means—brings the best out of top players who are used to playing that kind of game. So I think the choice—and it was a choice!—for Spurs to look downward at Brentford, Brighton, Bournemouth, to want to trade places with those clubs (as we have), was translated into Spurs’ football this past season, via Thomas Frank, as a kind of Football Centrist commonsensical pragmatism. In that way Spurs’ football-side executives remind me of Democratic party operatives looking at polls and urging candidates to tack to the political center while their constituents are asking for boldness, resolve, and panache. Spurs became Football Centrists, not audere est facere, but contemptuous of possession and intensity. They started aping a version of a tactical meta that they never understood or planned for in the first place, and will likely have come and gone by the time they figure out how to operationalize it. This was, again, no standard misstep; it was a very recent choice leading directly to a precipitous drop not just in table, but in level, in class, in ambition.

When you put all of this together, it’s clear that we’re not talking about a decade of missteps, but an abrupt and penetrating shift in attitude and ambition within the club. The club’s new executive leadership structure thought they could make their bold mark in invisible ink. The contradictions of Postecoglou’s second season were more than their Football-Centrist brains could handle, so they turned the club’s values inside-out, suspicious of what might happen if Spurs were simply allowed to be Spurs. In practical terms, this meant, in Frank’s words, taking a risk without taking a risk. They sacked a manager who inspired the players; they appointed a manager who thought, as they did, that the club’s mentality was too big for its stature; and in any case they hung both Ange and Frank out to dry when both most needed club investment in the squad. All of this happened in the past 20 months. We should be so lucky to get back to the past 10 years.

When the final whistle blew in the ground yesterday, I felt the tightness in my shoulders loosen up, even as the sun beat down on my face and partially obscured my view of the celebrations on the pitch. But a part of me still wondered what I was in for on my trek back to Cambridge from Seven Sisters, clad in Spurs kit. I started steeling myself for the route through King’s Cross, where Arsenal, West Ham, and Spurs supporters, among others, were destined to converge at the end of an emotionally charged (and probably alcohol-fueled) day for all of us, in our different ways.

As I made way through bottlenecked crowds, occasionally checking in on the social media reaction to the day’s results, I was taken by surprise. What I saw—granted, purely from a sample of convenience—somehow made me feel slightly better about Spurs finishing 17th.

What I saw was Arsenal fans taking time away from celebrating their league victory to whine about Spurs fans celebrating avoiding relegation and about UK television coverage making Spurs the story of the day. I also saw no small amount of crowing and rending of garments and gnashing of teeth among not just West Ham supporters, but supporters of the likes of Everton, Liverpool, Villa, and others over the fact that Spurs managed to escape relegation. Everyone—even those with something to celebrate on the season’s final day—was mad that Spurs refused to slink out of the league to complete the narrative climax they all wanted and expected. It reminded me of the discourse of dissatisfaction over the final episode of Game of Thrones.

Like every other Spurs supporter, I’m more than capable of indulging the self-pitying idea that we are particularly hard done by in the media. Yet it still came as something of a shock to me just how powerful was the appetite around the league, among media and supporters alike, for Spurs to go down. I posted this shortly after the match:

Hatred Exemplified

Admittedly it’s a combination of amusing and infuriating to learn from yesterday the degree of intensity with which people who aren’t Spurs just hate Spurs. Only the day before, Times writer Alyson Rudd published one of the most openly hateful pieces about Spurs that I’ve ever read, shockingly and implausibly blaming the poor season on Spurs supporters.

Lest you think this was a one-off, be warned that it was only the latest in a series of Rudd’s unhinged flailings at Spurs. In February, after it was clear to everyone in football with a shred of integrity that Thomas Frank had woefully mismanaged Spurs into a dire situation toward the bottom of the table, she blamed Tottenham Hotspur for being too entitled to win under Frank:

And shortly before that, in January, she said the quiet part out loud: that Spurs are simply entitled for wanting to be back to competing for top-4 and playing a dominant style of football, and not languishing at the bottom of the table hoping to remain in the Premier League:

For those of us who attended most Spurs home matches during this extremely trying season, the idea that we are a ‘snide,’ ‘negative’ bunch who fail to get behind the team simply beggars belief. It is little more than a cruel and vindictive shot taken at a group of fans who’ve stuck by our football club through the kind of rollercoaster ride few have the privilege and misfortune of being strapped in for: Suffering an historically unprecedented run of injuries for more than two full seasons now after losing talisman (Kane) after talisman (Son) in quick succession; resting our few remaining first-11 players in the Premier League last season in a gamble to win our first trophy in 17 years; bringing in a manager in Thomas Frank who guaranteed nothing but that we would ‘lose games’ and are not good enough; fighting relegation, down to the last match day, just months after tasting European glory; pulling off one last home victory to stay up, in front of a roaring crowd full of resolve, relief, and, ultimately, belief in better days ahead.

Even Spurs’ own beat reporter, Jay Harris, got in on this ethereal hatred of Spurs, mere days before our decisive match against Everton. Furthering a long-running narrative that Spurs’ captain and best player, Cuti Romero, must also hate Spurs as much as everyone else seems to, he published this hit-piece declaring Romero a disloyal player and not a true captain. It was, of course, based on what turned out to be a lie, that Romero flew back to Argentina and would not be present for Spurs’ final game of the season.

In reality, Romero received permission from de Zerbi himself for his trip to Argentina, and was back in N17 for the match, celebrating with vigor on the touchline with the rest of his teammates. Here he is with his arm around Pedro Porro after the match, both of them looking as emotional as you would be after that match (photographic evidence, in case you’re still buying the conspiracy theory that our captain abandoned our team):

The nature of this kind of hatred has common threads. Spurs—not just fans, but the entire club—are made out to be, in particular:

1. entitled

2. snide

3. interlopers in the ‘big 6’

4. all about money, not character or resolve

Items 2, 3, 4 on that list at last—’snide,’ ‘interlopers,’ ‘all about money…’—feel like moderately strong antisemitic dog-whistles. Even massive organizations like Manchester United, Liverpool, and Arsenal can somehow skate by with a ‘salt of the earth’ image among football followers, whereas Spurs are plastic invaders of the ‘big 6’ who must have gotten there by some kind of financial trickery.

But I think it’s actually much deeper and more complicated, hence the following Class Theory of Spurs hatred:

The Class Theory of Spurs Hatred

In the past I’ve gestured toward a class theory of English football, but here I’d like to explore it more directly.

When I refer to ‘class’ I mean it in two, interlocking ways: (1) class is a metaphor for the rungs of Premier League club status, but also (2) how we think of such metaphorical football classes is directly impacted by wider English socioeconomic class politics.

To understand this, you have to start with the very idea of the ‘big 6.’ Historically, football fans might speak of something like the ‘big 5,’ which in the 1980s and 90s might have meant Arsenal, Liverpool, Manchester United, Everton, and Spurs. Into the 21st century, ‘big 5’ started to look more like Manchester City, Manchester United, Arsenal, Liverpool, and Chelsea. The ‘big 6’ only became a thing because Spurs became the only other club to consistently break up City, Manchester United, Arsenal, Liverpool, and Chelsea at the top of the Premier League table. Even if you define the ‘big 6’ by the attempted ‘Super League’ breakaway, Spurs were only part of that group because we were the only other club to break into consistent European qualification and top-5 table finishes. The data are pretty clear on this.

There are two major implications here for the formation of the idea of a ‘big 6.’ One is that historic clubs—for example, Everton, Aston Vila, Newcastle, and Nottingham Forest—see themselves on the outside of an arrangement that they—not Spurs!—believe themselves historically entitled to be part of, but are not part of because they haven’t come anywhere near Spurs’ footballing performance over the past 15 years. Two, the more recent ‘big 5’ clubs who’ve won the lions share of trophies and league titles over the same period—Manchester City, Manchester United, Liverpool, Chelsea, and Arsenal, some of which are also historical ‘big 5’ clubs—also see themselves as entitled to a degree of exclusivity over and above Spurs. What the combination of these two forms of entitlement means for Spurs—who are only part of any of these conversations based on merit of league table finishes and European qualifications—is that we are hated—seen as uppity interlopers—both from above and from below.

Where this intersects with English class politics is that Spurs are a London-based club, carrying all the baggage (‘soft southern bastards’ and the rest of it) of justified socioeconomic grievance of the industrial north against the whiggish south. In this formulation, Spurs are not only metaphorical class traitors for taking a spot among the ‘big 6,’ but bourgeois upstarts in more ways than one. Just another big , leveraged London financial business (and here’s where some of the antisemitic dog-whistles come into play, I think) profiting off the backs of working people. I should probably say that, while I’m certainly not a Marxist (that’s a longer conversation for another time), Marx has quite a good explanation for the phenomenon by which the bourgeoisie mistakenly view themselves as class allies against the working class with the capital class that oppresses them both. But I think such class dynamics also explain why northern supporters of historical clubs such as Liverpool and Everton and Manchester United would ally with their real-life regional class compatriots against Spurs, even if in metaphorical football class terms it would be like the working classes allying with capital against whiggish bourgeois upstarts.

To put it more directly, I think widespread Spurs hatred from above and below comes precisely from these kinds of real-world and metaphorical class dynamics. There’s metaphorical class entitlement in Liverpool and Everton wanting Spurs out of the conversation of major English football clubs, the latter being jealous from below of Spurs’ recent class ascendancy, as well as real-world class resentment of Spurs as a London-based club that recently grew into a big global enterprise. The English football media—and even many Spurs supporters themselves—only magnify this dynamic by over-focusing on Spurs as a business enterprise (again, Daniel Levy’s chairmanship invited many dog-whistles along these lines) (and oh the stadium! Has anyone mentioned the Big Expensive World Class Stadium!?? That one that all those plastic fans and players don’t deserve???!!!) and dramatically downplaying how successful Spurs have been on the pitch—relative to the vast majority of the league—in the last 15 years. So much so that Spurs and football success have become a kind of joke pairing, even if it’s actually, demonstrably true that Spurs are far and away the 5th or 6th most successful English football club in recent Premier League history.

It is therefore helpful, I think, to understand this dynamic of Spurs hatred. This is the best explanation for why Spurs finishing 17th occasioned not the usual banter you might rationally expect, but something totally different: a league-wide media and fandom meltdown over Spurs failing to supply the schadenfreude enthusiast’s happy ending everyone else wanted: relegation for Spurs. Only this explains the oceans of piss boiled by one of our very worst seasons in the long history of the club.

COYS.

There remains a tendency in analysis of Spurs’ current situation to view this season and last as part of a failure continuum. The thinking goes like this:

Spurs currently sit 17th in the table with 7 games to go this season and finished 17th last season, therefore what we’re seeing now is simply a continuation of a process of rot that began last season. Since Spurs finished 17th last season, it’s not a surprise we’re in 17th this season.

I think this way of thinking is borderline pernicious and obscures the duration and depth of the problems that have Spurs facing the unthinkable, the prospect of relegation. And it’s fairly easy to demonstrate why.

Let’s start with this point last season. With 31 matches played last season, Spurs sat 14th in the table with 37 points, which was 17 points above relegation (Wolves sat 18th with 20 points).

With 31 matches played this season, Spurs sit 17 with 30 points, which is 1 point above relegation (West Ham sit 18th with 29 points).

Last season at this point, Spurs’ had a goal differential of +13, which was tied for 6th best in the league, despite sitting 14th. Spurs had scored 58 goals and conceded 45.

This season, we have a goal differential of -10, which is the 6th worst in the league, having scored 40 goals and conceded 50.

So right away we can observe some important differences between where we are now and where we were then, foremost among them being that this season’s West Ham are 9 points better than last season’s Wolves through 31 games, while this season’s Spurs are 7 points worse than last season’s Spurs. The combination of those two things is the difference between having a disappointing season and focusing on Europe with a 17-point cushion above the drop and being embroiled in what’s essentially a 4-way battle to avoid relegation.

Such differences are obvious enough to everyone who follows Spurs, but what hasn’t gotten due attention is how much worse the underlying numbers are this season than last season to this point. Our -10 goal difference right now is a net 23-goal drop-off from where we were at this point last season. I don’t know about you, but I think that simple statistic is simply astonishing. Likewise, for all the justified talk about Spurs’ defensive woes last season, we’ve actually conceded 5 more goals through 31 games this season, all while scoring a depressing 18 goals fewer.

These are not marginal differences signaling a slow and steady decline or, as I put it at the top, a continuum of failure from the start of last season to now. Nor am I playing cute with hyperspecialized metrics; we’re just talking about Spurs’ ability to pick up points, to score goals, and to try not to concede goals. A 23-goal swing, with 7 fewer points, from matchweek 31 last season to matchweek 31 this season is not a steady slope of decline; it’s falling off a cliff.

It’s worth digging into this a bit more in terms of player availability, considering Spurs have been totally injury ravaged in both seasons. The lack of goals in 2025-26 makes perfect sense in light of the lack of xG and the tactical reliance on set pieces for chance creation. But what’s telling is that even with greater depth at the CB positions, and greater availability of Romero and van de Ven this season than last, Spurs’ defending has been worse this season than last. And I think worse in substantial way, not in a ‘well this is just a continuation of last season’ kinda way.

As you can see below (I’ve used versions of this chart a bunch of times now, regularly updated), Spurs were among the best defenses in the Premier League last season with Romero and van de Ven in the team. It was, to put it mildly, a problem that they were in the team for just 15 games. But even with far greater availability this season—23 of 44 games in total so far, excluding where one of them played only a half due to a red card or tactical substitution—Spurs have struggled to defend.

If we zoom out from the CB situation and look at team metrics, it’s not surprising why the GD and points are so much worse this season.

Through 37 games in all competitions this season—the duration of Thomas Frank’s tenure—Spurs conceded 50 goals. Through 37 games in all competitions in 2024-25, Ange’s Spurs conceded 51 goals. The former was right on xGA (50.35), while the latter was a massive overperformance of xGA (63.03).

It’s worth pointing out, however, that through 37 games last season (chosen so the Frank and Ange samples are the same number), Romero and van de Ven played together just 8 times, whereas this season through 37 games under Frank they played together 21 times. Additionally, the 37 game sample from last season includes not only a big chunk of games without Romero and van de Ven, but a big chunk in which Spurs were playing a non-CB at CB alongside probably our worst CB in Dragusin:

Things haven’t gotten better under Igor Tudor so far (not that you needed me to say so). In 7 games so far under Tudor—with Romero and van de Ven paired at CB for 2, excluding a third in which van de Ven went off at halftime after playing a hybrid left back role—Spurs have an xGA / 90 of 1.76 and are surrendering an average of 3 big chances / 90. What’s more, our opponents are taking their big chances against us at a higher rate (57%) compared to 37 games under Frank (18%). Meanwhile, we’re taking more of our big chances now (50%) but creating fewer. The 7-game sample for Tudor is of course very small and very difficult to make much of, especially as it includes Arsenal, Liverpool, and Atletico Madrid twice:

From this minor update we can see that even though we’re sitting 17th now and finished 17th last season, we’re not witnessing a continuation of last season’s ‘form’ or problems. We’re witnessing a combination of (1) a much more dire situation with respect to relegation, since the bar is 9 points higher this season compared to last; and (2) a dramatic drop-off in goalscoring, chance creation, and defending. This season Spurs became a much worse side than even the one that finished 17th last season. To become the 17th place team we were last season, we’ll have to finish this one by creating a +9 goal differential in the final 7 games. I hold out hope that we can concede fewer than the 15 goals that separates where we are now to where we ended up last season; I’m less optimistic we can score the 3.43 goals per game it would take to catch last season’s goalscoring output. But if we’re going to stay up, we’d better find some balance between these objectives.

On more than one occasion I’ve sorted myself out on the post-match walk from the ground to Seven Sisters. Last night, it was more like the Slough of Despond.

I couldn’t shake the mental image of Vicario receiving the ball to feet, under no pressure, flanked by fullbacks and with midfielders checking—if you could call it that, more like shuffling—to the ball, also under no pressure. I watched Vicario push the ball out from his feet and shape into run-up to whack it long and said aloud to myself, up there in the Shelf Side where no player could hear, ‘Don’t do it!’

Vicario smashed it upfield, the ball bounced around a bit, Palace came away with possession, and Spurs went right back to defending.

This is the football we’ve played all season long. It’s not just lethargic and reactive, the antithesis of control; it’s also, in immediate, practical terms, what’s prevented Spurs from developing any semblance of ideas about what to do with the ball, such as how to play out from the back, how to work the ball through the midfield and into the attacking third, how to play combinations, how to maintain possession when the game state calls for calmness or change of pace, how to manufacture shots on target, how to pass the ball on the ground, and how to move off the ball while in possession in a direction other than forward and away, like a pack of NFL wide receivers in a trips formation sprinting to get underneath a hail mary in the end zone.

To put a fine point on it, Spurs don’t attack anymore because they’ve been coached not to. Spurs don’t play with intensity anymore because they’ve been coached with the mindset that intensity is naive.

The 9th richest club in world football should not be the 10th in the Premier League in possession, or the 14th in touches in the opposition box, or the 17th in xG. Spurs don’t have this kind of data profile because most of the same players who produced 64 goals last season forgot how to play, and so did their expensive, experienced reinforcements from Champions League sides. Spurs have this kind of data profile because senior leadership at the club hired Thomas Frank to put a stop to high-intensity, attacking football, and so he did.

Throughout the season, Spurs fans have mooted to what extent the players are to blame. While I agree with Jack Pitt-Brooke that responsibility for what comes next ultimately falls to the players, regardless of who deserves what share of blame, I think the view that Spurs’ players aren’t good enough is not a serious view and not worthy of serious engagement. Perhaps the depleted squad is not good enough to finish in the European places, but it’s more than a stretch to argue the players who’ve played for Spurs for the majority of this season—important internationals for England, the Netherlands, Argentina, France, Italy, Spain, and Portugal—aren’t good enough to beat Fulham or Wolves.

It’s comparably suspect to make assumptions about player attitude, to question their commitment, when in practice the game plan all season has been to surrender as much responsibility for the outcomes of matches as humanly possible to the opposition. In what sane world is doing nothing but challenging for aerial duels an occasion to show one’s full commitment to the club and the game? What opportunity is there to demonstrate the resolve to emerge from a relegation battle a Premier League side when every single one of these players has been deprived of so much of the actual, physical game of football, reduced to lifting it long and chasing it about like kindergarteners playing for the first time, or passing up quick-start opportunities in the final third to wait for someone to amble across the pitch to take a long throw once the opposition defense is settled in anticipation of exactly that?

In short, the football we play—by design—would be Valhalla with a squad of 24 Palhinhas, but that’s not what we are. We have a squad of expensive and dynamic players—Xavi, Kolo Muani, Solanke, Tel, Porro, Romero, van de Ven, etc.—who can do so much more than kick and chase. Arguably Spurs’ squad is weakest when asked to box defend and play for duels.

Another thing Jack Pitt-Brooke has been right about, I think, is that what’s ailing the players is foremost a crisis of confidence, not of ability. As I’ve explained elsewhere, there’s a psychological toll to what Spurs senior leadership have done by sacking Postecoglou and bringing in Frank. I’ve put it this way, by analogy:

Elsewhere I’ve suggested the analogy, for our players, of being an employee at a company that’s had a miserable year, with revenue running down and the looming threat of being out of business and out of a job, when suddenly the company wins a major award that brings business back to life. The CEO gathers a handful of creative employees and publicly praises them, attributes the award to their creativity and effort, and gives them a promotion to focus on their award-winning, creative work in the next quarter.

But then, suddenly, the shareholders sack the CEO. The new CEO comes in and makes it a point every week to let all the employees know that their little award is nice and all, but it doesn’t mean anything, because none of them are very good at their jobs and they’ve just had such a horrible year. The new CEO believes it was precisely the creative employees’ creativity that was dragging the company down, so he revokes their promotions and assigns them to manual data entry for the foreseeable future. For the talented, creative, award-winning employees who used to love coming to work, every day is now drudgery. All the while, the company’s revenue begins to tank again, worse than ever before, while the new CEO crows every day about how great things are going and how important it was to stabilize the company by refocusing on manual data entry instead of creativity.

If you were (or have been) in that kind of situation, you’d understand what I suspect our players have been going through this season, and your attitude toward work, no matter how professional and self-regulating you are, would necessarily suffer.

What I mean to illustrate with this analogy is not only that it’s a letdown after Europa to have your project and your accomplishments diminished in the way club executives have done, to the players, to all of us, but also that elite professionals want to play what I’ve called protagonist football. They don’t want to be told they’re not good enough to posses the ball and dictate the flow of the game. They don’t want to be told to sit and wait and react to what the opponent is doing, and build their entire game plan around that, especially if most of their opponents are not as good as they are. Especially those players who come from elite club backgrounds—I’m thinking of Xavi, Kolo Muani , and Tel in particular here—who’ve played technical, high-possession, high-expectations football their whole young lives, there’s no platform at Spurs right now to shine, no pathway to being the footballer they all want to be.

This is the source of the crisis of confidence that plagues the squad. It’s not that they don’t want to win, or work hard, or stay up; it’s that they’ve had it drilled into their minds for 80% of a season that they are not, nor should be, in control of football matches.

It’s not just about nice football that’s fun to watch.

Attacking is a strategy for winning football matches, not a ‘style of play.’ I want to be clear that when I say Spurs are bad because we no longer know how to attack, it’s not just a personal preference of mine, a reflection of the kind of football I like to watch. It’s because attacking teams are successful.

To put some substance to this claim, I looked into something relatively straightforward: Over the past 5 Premier League seasons (including this current season to date), how strong is the correlation between xG and points? As you can see here, it’s very strong!

Now, you might say, but what about the correlation between low xGA and points? And you’d be right to think it’s also very strong:

But here’s where it gets interesting, at least as I see it. In the graph below, I plotted the correlation of xG and points (green bars) for each of the last 5 seasons (including this one to date) alongside the inverse correlation of xGA to points. I’ve given you the correlation coefficients as the differences are extremely small (and a correlation coefficient of .7 or above typically indicates a strong correlation between two things). As you can see, in 2021-22, 2022-23, and 2023-24, creating a high xG correlated slightly more than giving up a low xGA with points. Last season and this season, however, it flips, and defending correlates slightly more with points:

The differences are so small that the ‘flip’ doesn’t mean much here, but foremost what this simple analysis shows is that good teams are generally good at both attacking and defending; both actions are more likely to get you more points. That’s obvious, I know, but I think it’s important to establish statistically because of what I want to say next.

It’s a timeless debate about whether attacking or defending is the more important thing to do to win. In reality, the best teams do both really well. But you could also ask whether they do both really well because one thing is the better enabler of the other.

In other words, which is more plausible:

1. Teams that focus on defending and defend well create high xG by virtue of defending well.

2. Teams that focus on attacking and attack well yield low xGA by virtue of attacking well.

My view is that the latter is easier to justify. When you have the ball far away from your own goal, it’s harder for your opponent to attack. When you’re creating a lot of chances, you’re forcing your opponent to focus on defending, making it harder for them to develop attacking moves and push players forward.

You could argue the reverse—that, in short, counterattacking from a successful defensive setup generates lots of chances it wouldn’t otherwise—but if you look at the top teams in the Premier League each season you don’t see counterattacking profiles; you see not only high xG and low xGA but also high possession, high territory, high touches in the opposition penalty area, and so on.

This doesn’t mean defense-first sides can’t be successful. They’re just much less likely to do so, and also more reliant, I think, on overperformance. Forest last season were a good case in point: they radically overperformed xGD, and now that they’re no longer overperforming at that rate, they’re in a relegation scrap. Sides that try to ‘first raise the defensive floor’ don’t exist, because attacking well is a big part of defending well.

Attacking is good, actually. We should start doing it again.

The particular hubris at work in Spurs’ situation is not that of a ‘big club’ like Spurs thinking we’re too big to go down. It’s the hubris of Premier League exceptionalism, the idea that the Premier League is so special that it defies the fundamental laws of football, therefore the only way to succeed in it is to chase the holy grail of ‘Premier League experience,’ even if it’s the experience of mediocrity and desperation, i.e. the experience of what’s widely called ‘pragmatism,’ which should just register as an epithet at this point. What matters about Ange, or Frank, or Tudor is not whether they have Premier League experience, but what they’re trying to achieve and how they’re trying to achieve it.

I don’t know what Tudor has in mind. It’s hard to draw conclusions so early, even if the reality is that unfortunately we don’t have much time to ponder. If he’s working toward some kind of attacking intent based in possession—at this point we don’t need to have City-level possession, just the intent to do more with the ball than smash it long—then Spurs will be OK. If not, the club should get someone in—anyone, really—who can set us up to play possession, attacking, territorial, control football. Protagonist football. Main character football. Not bit part football, cameo football, surrender football.

What stands between Spurs and safety isn’t player ability or attitude. It’s the impetus and structure to use the ball like a real football team for whom an 8th place League finish used to be a bad year, a team that’s been in more Champions League competitions in the last 12 seasons than Arsenal and Manchester United, a team that just brought home a European trophy a few months ago. That’s who we are. That’s still who we are. And even if club executives have forgotten that, even if the players are struggling to remember it, we the supporters have the luxury of enough critical distance to hold this in mind and never let it go.

This piece is less an analysis than a public service effort. Last Sunday’s loss to Arsenal was bad, though few expected anything different. It doesn’t follow, therefore, that Spurs are doomed. Consider the following:

1. Arsenal are probably one of the top 3 teams on the planet right now. They’re in the final form of a very expensive, multi-year project that has them in the position to bring on the likes Martinelli, Madueke, and Odegaard of the bench.

2. Spurs had a total of 14 (15 if you count the backup goalkeeper who’s played 0 minutes this season) fit senior players. You could build an 11 out of our injured senior players that, if fit, would probably defeat the 11 we had available to play.

3. The new manager, Igor Tudor, had a little more than a week to work with the team, so expecting any kind of substantial changes in how we play would have been expecting a lot. Expecting any changes at all to have actually been inculcated to the point of effectiveness would be expecting the impossible. Spurs did play with intent in the first half—even if they were outplayed in so doing—and that’s a baseline expectation. In short, it doesn’t make much sense to sulk over the fact that the performance didn’t match the good vibes in the stadium before kickoff, because it was never likely that Tudor’s presence could move the dial much in the first game. At the same time, the fact that we went into halftime even is a good sign of intent and something to take away, even if we were outplayed. Getting outplayed is a function of finer tuning than Tudor and the players had any time to execute in a week of training, though it’s possible to do a lot of that fine tuning with more than a quarter of the season left.

4. Had Kolo Muani’s would-be equalizer stood, it would’ve altered the game state significantly in Spurs’ favor. Arsenal were coming off a last-minute Wolves comeback in which the former dropped points, with ‘it happened again’ ringing in their ears. Gabriel’s school-play level acting job was an unalloyed embarrassment to the game, but it worked. Scenarios with the kind of hand positioning that Kolo Muani applied do get called sometimes, but scenarios with the degree of force that Kolo Muani applied almost never do. Particularly in light of Ekitike’s much more forceful push on Romero earlier in the season, which stood against us, you have to say it was a bad call that erased our second goal. Materially it means nothing to say this, but in terms of morale it’s worth bearing in mind. The game might have been more competitive than it was in the end had the game state that Spurs earned have been allowed to come into effect.

5. This is all to say: If you take the emotion out of losing to Arsenal, and the amplification of giving up a 4th goal in garbage time, this match was never the yardstick for what the team can accomplish in the final quarter of the season. It was never a ‘free hit,’ for obvious reasons, but that doesn’t mean it was representative of our task in the games that remain. It was not.

6. The fact is we still don’t know what Tudor will or won’t accomplish with this squad. Pundits are always going to say everyone else ‘shows more fight’ than Spurs, because that’s the stick they beat you with when you’re a big team underperforming. But I promise you none of these players want to be the ones who took Spurs down to the Championship, and no magic coaching is going to organize them (or anyone) in just a week’s time.

7. 11 games is a lot of games. 6 of those remaining 11 are against teams who are also in the bottom half of the table right now. 2 more of them (Fulham, Everton) are against teams just barely in the top half. Spurs are on a down-streak while some others (West Ham, Wolves) are on an up-streak, but rarely does anyone hold a steady trajectory either way for 11 games straight.

8. Pessimism is not inherently rational. Neither is optimism. It’s easy to feel pessimistic about recent events, but pessimism is not inherently better than optimism at predicting future events. I think we’ll have a better basis for predicting future events over the next few games than we do after an extremely unsatisfying defeat in the North London Derby, so again it makes little sense to assume that if we can’t beat Arsenal then we can’t beat Fulham or Palace or Wolves either.

When I sat down to write this I had in mind a statistical overview of Frank’s time at Spurs. I wrote about 500 words in that direction and inserted a few data tables before realizing that it would’ve been the wrong piece to write. Because as bad as the results were, as bad as the underlying numbers were, none of these things adequately captures why Frank didn’t work out at Spurs.

Indeed, the one thing that best captures the failure of Frank’s tenure—the thing smoldering under the surface for most of the season that would eventually erupt in boos and ‘sacked in the morning’ chants at the manager—still seems poorly understood among journalists, pundits, and the Spurs fanbase at large.

That’s not to say we haven’t all felt it. And when I name it, you’ll surely think ‘of course,’ because it will seem obvious. But still I haven’t seen many people put a finger on it.

I’m speaking of the fact that at first symbolically, and then in action, Frank embodied a total lack of ambition, which manifested in Frank himself perceiving and speaking of the club the way our most egregious banterers and detractors do.

He certainly meant no malice; he was just a true believer in the ‘Spursy’ narrative that Tottenham have no business thinking of ourselves as a big club with serious ambitions, no business being the protagonists of matches, no business occupying the tactical vanguard. He made continual reference to our finishing 17th in the league last season not because he saw it as the aberration it was, following seasons finishing 5th, 8th, and 4th, but because he truly believed the bottom of the table is where we deserve to be. Viewing Spurs that way, he came to see himself as a kind of savior and his role at Spurs as akin to his role at Brentford, which is not to compete for trophies and European qualification, but to remain solidly in the Premier League as a side character in the story of bigger clubs.

Frank’s appointment was, in and of itself, a sign that Spurs senior leadership were starting to look down the table with envy at the likes of Brentford and Brighton instead of up the table with ambition. I pointed this out back in August in my discussion of how we should set our expectations for this season:

But what I do know is Spurs can’t afford to keep slipping back into mid-table, by which I mean Spurs’ senior officers need to stop looking down the pecking order at Brentford and Brighton with envy and start looking up at Liverpool and Arsenal with purpose. That means as fans we should expect a top 4-5 finish in the league this season and every season. The fact that we finished 17th last season was an own-goal, a refusal to strengthen in January when we had the chance and a run of 14 crucial games in 7 weeks, among other things. It’s not our ‘natural’ table position going into this season.

And every day since Frank was appointed until he was sacked, he managed our roughly £800m squad like £100m squad, showing no belief in our players’ ability to possess the ball, develop attacking patterns, build out from the back, convert high turnovers into threat, or do virtually anything besides slamming the ball high and long into the channels, chasing it down, and throwing themselves into duels.

It comes as no surprise that our players neither respected nor enjoyed this approach to football. They found Frank’s tactics too passive, too reactive. They complained of spending too much time preparing for what opponents would do and not enough working out what Spurs would do. As the poor results came week in and week out, Frank spoke in press conferences as if the squad were overachieving, as if snatching a draw to Burnley were a significant accomplishment for Tottenham Hotspur, 2024-25 Europa League Champions.

As fans we found ourselves miffed and anxious over the results, baffled by the disconnect between the poor quality and lack of initiative on display for our eyes and the rosy picture Frank would paint in the post-game pressers, and frustrated by the gaffes (Arsenal cup, ‘true fans,’ etc.). But what unites all of these experiences, what accounts for the daylight between what fans have witnessed and how Frank spoke of the performances and the club, has gone largely unnamed. It is, again, the fact that Frank embodied such a complete lack of ambition, such a completely deflated view of the club and its place in the food chain, that Frank’s tenure always felt like a profound betrayal, like an affair with a rival taking place in our own house.

Understandably, then, the fans have been explicitly clapping the players and not the manager, because the players and fans both feel this betrayal, in our own separate ways.

Elsewhere I’ve suggested the analogy, for our players, of being an employee at a company that’s had a miserable year, with revenue running down and the looming threat of being out of business and out of a job, when suddenly the company wins a major award that brings business back to life. The CEO gathers a handful of creative employees and publicly praises them, attributes the award to their creativity and effort, and gives them a promotion to focus on their award-winning, creative work in the next quarter.

But then, suddenly, the shareholders sack the CEO. The new CEO comes in and makes it a point every week to let all the employees know that their little award is nice and all, but it doesn’t mean anything, because none of them are very good at their jobs and they’ve just had such a horrible year. The new CEO believes it was precisely the creative employees’ creativity that was dragging the company down, so he revokes their promotions and assigns them to manual data entry for the foreseeable future. For the talented, creative, award-winning employees who used to love coming to work, every day is now drudgery. All the while, the company’s revenue begins to tank again, worse than ever before, while the new CEO crows every day about how great things are going and how important it was to stabilize the company by refocusing on manual data entry instead of creativity.

If you were (or have been) in that kind of situation, you’d understand what I suspect our players have been going through this season, and your attitude toward work, no matter how professional and self-regulating you are, would necessarily suffer.

For all of these reasons, I think, players and fans alike turned on Frank by the end, tired of being told—in words and deeds alike—that we aren’t good enough, that we aren’t mighty Tottenham, that we aren’t the protagonists in this story. Such sentiments would be unacceptable, to me at least, even if were were soaring in the league, but to have our pride and expectations swept out from under us by the man who’s supposed to be our leader, our defender in the press, our visionary, while sinking into a relegation battle 26 games into the season is both inexcusable and intolerable. It has been, for me, the worse experience I’ve had with the club in my nearly 20 years as a supporter.

In Case You’re Interested in the Numbers

Thomas Frank managed Spurs for 37 matches across all competitions this season (excluding the Super Cup), ending his tenure with the lowest winning percentage of any manger in Spurs’ history.

In all competitions, Spurs won 13, lost 14, and drew 10 under Frank. In the Premier League, Frank’s Spurs won 7, lost 11, and drew 8. Here is a list of every team Spurs have defeated this season under Frank:

Burnley; Manchester City; West Ham; Villareal; Doncaster; Leeds; Everton; Copenhagen; Brentford; Slavia Prague; Palace; Dortmund; Frankfurt.

Of these, City (second game of the season) and Dortmund are probably the only ones you’d write home about (maybe a minor shout for Frankfurt).

Spurs exited both domestic cups early but took advantage of a relatively easy draw in the Champions League, placing 4th in the league phase and earning an automatic bid to the round of 16.

Struggles with Chance Creation

More than anything, Frank’s Spurs struggled with chance creation. Spurs had a higher xG than their opponents in just 15 of 37 games (green in the chart below indicates xG ‘win,’ while red indicates xG ‘loss’). Most of those 15 came against subpar opponents, such as Burnley (twice), West Ham, Doncaster, Copenhagen, Fulham, Brentford, Slavia Prague, and Sunderland. In the aggregate, Spurs have an xGD of -6.01, having created just 44.35 xG in all competitions. A massive chunk of that—22.95, or 52%—has come from games against a lower league opponent (Doncaster), a lackluster group of Champions League opponents, and Premier League teams that are either newly promoted this season or currently in the drop zone.

Spurs have also been quite fortunate in terms of big chances, both the ones we’ve had and the ones we’ve surrendered. When I looked at big chance conversation rates last season, I found that teams were missing about 64% of their big chances (league average at the time). Bearing in mind that 64% miss rate as a baseline, Spurs have slightly outperformed the baseline, missing 61% of our big chances. More significantly, however, our opponents this season have missed 72% of their big chances against us, meaning we have converted our big chances just a little better than the league average, but our opponents have missed theirs at much higher rate. In this sense, our defensive record should actually be even worse than it is.

A Poor Defensive Record

About that defensive record, by the way…

The narrative from the start was that Frank was supposed to come in and ‘make us more defensively solid,’ ‘raising our floor.’ As I wrote in September, I was never convinced of this. And indeed it never transpired.

Sure, Spurs are currently 13th in the Premier League in xGA (37.6 over 26 games, or 1.45 xGA / 90), which is only slightly better than 17th in this category last season (64.4 over 38 games, or 1.69 xGA / 90), but this is an apples to oranges comparison. For all the injury problems Frank has had this season, we’ve mostly had our goalkeeper and back line healthy, whereas last season we played 36 of 60 games, or 60% of all of our games, with at least one player out of position at center back. In fact, last season, Romero and van de Ven only played together for 15 of 60 games. If you take away the first-choice center back pairing for any team for 75% of a season, you can expect dramatically worse defensive performance (to say nothing of also taking away at least one of their starting fullbacks and their starting goalkeeper for 20+ games as well, and replacing the first-choice CBs with players who don’t play CB).

To provide more of an apples to apples comparison, I broke down our defensive numbers between all of last season under Postecoglou and all of this season to date under Frank according to center back pairings. As you can see, whereas Postecoglou had Romero and van de Ven together for just 15 games in all of last season, Frank had them together for 21 games already this season (playing one of them for another 13 games, meaning Frank has only had a non-CB at CB for 3 games, 2 of which by choice rather than availability issues). With Romero and van de Ven last season, Postecoglou’s Spurs conceded just 12 goals in 15 games or .8 goals / 90, with an xGA / 90 of just .86. In 21 games with the same pair in 2025-26, Frank’s Spurs have conceded 29 goals, or 1.38 goals / 90, with an xGA / 90 of 1.19. Postecoglou’s Spurs won 9 of 15 and lost 4 of 15 with Romero and van de Ven, while Frank’s Spurs won 8 of 21 and lost 7 of 21 with the same pair:

In short, if you account for player availability, Frank hasn’t improved Spurs’ defense; arguably he’s made it worse (certainly we’ve been worse defensively with our first-choice defenders in the team).

Even if you want to claim a marginal defensive improvement this season over last (again, I think that’s not well supported by evidence), the attacking returns under Frank have been so abysmal that the defensive improvement would have to have been far more dramatic to produce better results.

Over-Reliance on Set Pieces

One concern I had when Frank was appointed was the possibility of over-reliance on set pieces for attacking output. Last season, Frank’s Brentford were 4th in the Premier League in percentage of xG from set pieces, with 28%:

The season before that, Brentford were 4th with 27%:

And the season before that, Brentford were 2nd with 33%:

I haven’t yet calculated the league stats for the current season, but I do have Spurs’ data. As you can see below, Spurs under Frank this season have generated 30% of our already anemic xG (Spurs currently sit 17th in the Premier League with about 1 xG / 90) from set pieces. If this season’s league norms for set-piece xG are anything like the previous ones, 30% is likely toward the top of the league, suggesting my concern about over-reliance on set plays for attacking output was warranted:

Surrendering Leads

Finally, as I discussed way back in August after the Super Cup loss to PSG, a major concern I had about Frank from the start was that we had no plan for possessing the ball and otherwise controlling the game, especially when we have the lead.

Over 37 games under Frank, we took the lead 18 times and surrendered it 8 of those times, or 44% of the time. For 3 of those 8 times, we surrendered the lead in the 2nd half.

When we took the lead and kept it, our possession went down from the first half to the second by an average of 9.6%. When we surrendered the lead, our possession went down from the first half to the second by an average of 2.25%. I’m not sure there’s anything here beyond an observation, i.e. so much depends on game state:

However, here’s one way of looking this data. In the graph below, I’ve focused only on the games in which we took the lead, whether or not we surrendered it. I’ve plotted changes in possession from first half to second half (PStand, blue line) and changes in xGA from first half to second half (xGAStand, grey line) (because changes in possession percentages are much larger numbers than changes in xGA, I’ve normed the data using z-score standardization). As you can see here, when we’ve increased possession in the second half, in many cases, our xGA has decreased, meaning possession has tended to improve our defending in these scenarios:

It’s a shame, therefore, that across 37 games in all competitions, Frank showed so little faith in our ability to keep the ball, even when we had the advantage. I think that about sums things up.

In the 2022-23 season, Spurs finished 8th in the Premier League, their worst league finish since 2008-9. The club sacked Antonio Conte during the March international break, leaving the squad to his assistant, Christian Stellini, then to Ryan Mason for the final 6 games of the season after sacking Stellini. Prior to taking over full-time, Stellini covered for Conte (who was back in Italy undergoing gallbladder surgery), seeing Spurs out of the Champions League tie against AC Milan with a 1-0 loss and a 0-0 draw—two of the worst football matches any of us had ever watched—and crashing out of the FA Cup in the 5th round, losing to Sheffield United.

The period after Conte’s sacking stands out for me as one of the bleakest times at Spurs in recent memory, not because we were in any particular jeopardy, but because the club and the squad seemed rudderless, void of ideas about who we were and where we were heading, and in no position to do anything about it.

In other words, we were in a holding pattern, a joyless in-between during which there was still more than a quarter of the season left and we were scrambling under caretaker management every week. Regardless of your views on Conte—I liked him more than most—it was pretty clear back then that once the club sacked him, the remainder of the season was a write-off, a frustrating exercise in stasis.

Now 25 games into the 2025-26 Premier League season, I think Spurs are in another holding pattern and, as always, I think it’s useful to look to the past for instructive context.

With Conte still at the helm 25 games into the 2022-23 season—just 3 games before he was sacked—Spurs sat 4th in the league table on 45 points, having scored 45 goals and conceded 35, with a GD of +11:

That’s an average of 1.8 points per game. After Conte was sacked—which came after 28 Premier League games, the 28th being that horrific 3-3 draw to Southampton and Conte’s subsequent fireworks in the post-match presser—Spurs played 10 Premier League games to finish out the season. From those 10 games we took 3 wins, 5 losses, and 2 draws, for a total of 11 points, or 1.1 points per game. The downward shift even from a beleaguered side under Conte was unmistakable, as was the feeling among much of the fanbase that we just wanted the season to end.

That was still enough to finish 8th, in large part because Spurs had already accumulated enough points under Conte to keep us firmly in the top half of the table. Spurs in 2022-23 had more than 50% more points through 25 matches than we do now.

In fact, if we look at the first 25 Premier League matches in each season since 2022-23, our performance this season under Frank has been not only the worst in terms of points gained, league position, and goal difference, but also the most dangerous in terms of relegation proximity.

Goal difference is a particularly striking stat in this comparison, because even the struggling Conte and Postecolgou sides scored far more goals than they conceded through the first 25 games—GDs of +11 in Conte’s second season, +14 in Postecoglou’s first season, and +12 even in Postecoglou’s second season—whereas Frank’s Spurs this season, with a GD of 0, look like we actually deserve our league position:

Holding Pattern Characteristics

The reason I’m comparing the first 25 games of this season under Frank to the decline of the Conte era in 2022-23 and the caretaker spell of that season is these stretches of games have distinct holding pattern characteristics.

By ‘holding pattern characteristics’ I mean both the evidence and the sentiment around the club and performances indicate a lack of progress and a lack of prospects for improving under current conditions. Even with some variation in outcomes, like the odd win or snatched draw, it feels like Groundhog Day, only on a bad day.

We know this because, as was the case for the first 25 games in 2022-23, 2023-24, and even 2024-25, we were furnished with evidence that systems under Conte and Postecoglou yielded top-end results with healthy squads. In their downfalls, both managers dealt with injury crises the likes of which Frank is now also dealing with (Conte to a lesser extent, perhaps, than the other two managers, though his tenure was certainly when we started talking of injury crisis and overtraining), but what’s different for Frank this season is that we’ve seen neither success nor any semblance of progress even during the healthy periods. It would be fair to point out that Frank has never had the benefit of Son, Maddison, or Kulusevski, though in return I’d say Conte never had two wingbacks and Postecoglou never had Kane, nor any cover for injuries to Romero, van de Ven, Solanke, Odobert, etc. with the likes of Kudus, Xavi, Kolo Muani, and Palhinha.

Even the claim that Frank has stabilized Spurs’ defense, raised our floor, and made us more defensively solid has proven fatuous. The 35 goals we’ve conceded in the Premier League thus far is the same as Conte’s Spurs conceded 25 games into 2022-23, and just 2 fewer than Postecoglou’s Spurs conceded 25 games into last season. However, the anemic 35 goals Spurs have scored so far this season is dramatically lower than the 46 under Conte, much less the 52 and 49 under Postecoglou. Instead of raising the floor, we’ve lowered the ceiling.

We can see how wrong the ‘Frank has stabilized Spurs’ narrative is with more granularity by looking at our defensive performance this season with different center back combinations compared with last season. For example, Frank has had the benefit of the Romero and van de Ven pairing 21 times in all competitions this season, which is already 6 more times than Postecoglou had them all of last season. Yet the pair has performed worse under Frank this season than under Postecoglou last season, averaging 1.38 goals conceded per game (versus .8 last season); 1.19 xGA (versus .86); and securing just 8 wins in 21 games (versus 9 wins in 15 last season).

We can also see that although Frank has recently been forced to use Palhinha—a non-CB—at center back for one half of a football match so far this season, Postecoglou was forced into using at least one non-CB at CB 36 times last season (!), or 60% of the time. Whereas last season Postecoglou had the benefit of being able to use one of Romero or van de Ven just 7 times, Frank has already had the use of one or the other 12 times. While Frank has done a bit better with that situation, averaging 1.08 GA per 90 and 1.64 xGA per 90 (compared with Postecoglou’s 1.71 GA per 90 and 1.94 xGA per 90), it’s still not good, and it’s arguably with better defensive personnel (Danso instead of Dragusin, Vicario instead of Forster and Kinsky, Palhinha sitting in front of the CBs).

Zooming out, if you look at Spurs’ xG, xGA, and chance creation so far this season, you get a similar picture of decline from prior seasons, with nothing to suggest the likelihood of improvement. Spurs have had a superior goal difference against our opponents just 15 of 36 times this season in all competitions, and just 9 of 25 times in the Premier League. Meanwhile, in all competitions, we’ve slightly outperformed a league norm of 35% big chance conversion, while our opponents have taken just 27% of their big chances against us, meaning our chance creation and concession are both worse than the poor outcomes we’ve witnessed thus far.

More alarming still, of the 42.5 cumulative xG Spurs have generated this season in all 36 competitions (1.18 xG per 90), 15.6 xG has come from our 9 matches against non-Premier League competition (1.73 xG per 90). Taking away our domestic cup matches against Newcastle (.83 xG) and Villa (.76 xG) in addition to all the matches we’ve played against non-PL opponents, we’re left with a Premier League cumulative xG of just 25.31 over 25 games (1.01 xG / 90). That puts us 17th in the Premier League, about where we’ve been all season in that metric.

These are just additional ways of saying that under Frank, Spurs have been consistently poor in the Premier League. We’re defending as poorly as last season with fewer defensive injuries and more defensive cover; our attacking has been dramatically worse with comparable attacking injuries to last season but substantially more attacking cover; and our actual results through 25 Premier League games are worse than ever.

This degree of consistency—of being consistently bad—is what makes this season feel like another holding pattern. We have 13 more league games (compared with the 10-game remainder after Conte was sacked in 2022-23) and appear wholly inert in the face of how poor we’ve been. The club did little to strengthen the injury-plagued squad in the January transfer window, just like last season. And the conversation among fans and pundits has shifted to whether we should consider ourselves a relegation risk, again without any indication from the club’s senior executives that they have any intention of making a managerial change.

Now, we wait. The time has long since passed to salvage much from this season in the league, at least according to the standards we’re accustomed to. Because Frank has less of a cushion above the relegation zone than Postecoglou had last season, he can’t afford to do what Postecoglou did and gamble all of his chips on the Champions League, resting key players in league matches for the final third of the season. He doesn’t seem the type to take that kind of risk anyway.

What this amounts to is very little prospect for change or adjustment between now and the end of the season, no obvious moves to change the situation, at least none that seem at all plausible at this point. Likewise, on the pitch, we have little evidence (if any) that the combination of Frank and this squad will be able to turn things around to the degree necessary to move into the European places. I’d love to be wrong about this, of course, but it’s hard to do much more than hope and wait. For me, it’s like circling Heathrow on the redeye, restless in my chair, eager for stimulus.

I had a particularly negative reaction to Spurs CEO Vinai Venkatesham’s recent message from the Board. A more positive reaction wouldn’t have been unreasonable, at least because the message was something rather than nothing after many years of uncommunicative boards. And as corporate pablum goes, the message was far from the worst of that genre.

What bothered me about it was less in its particulars than in its generality, a definitive trait of manicured corporate communication. What Venkatesham’s letter said to me was, essentially, ‘I know you’re frustrated with the state of the football, but don’t worry, we’re doing Best Practices here.’

The problem with that is I’m not sure corporate best practices are always aligned with football best practices, and I mean that well in excess of conventional tradeoffs between profit motive and footballing excellence, a stale conversation among Spurs fans in particular and sports fans at large. In this case, the corporate best practice I’m referring to is fast becoming a football best practice, and we’re seeing it not only in the leadership strategies of ‘moneyball’ clubs like Brentford and Brighton, but also, in different guises, at massive clubs like Manchester United. The best way to sum up the thing I’m trying to put a finger on here is the idea that you can ‘hack’ football, which is a version of the idea that you can ‘optimize’ an organization.

Fuck ‘Optimization’

If you’ll permit me a minor detour for a moment: I’ve always been a skeptic of the idea of ‘optimization’ in corporate culture. In its most innocent form it’s a platitude; it simply means ‘given your constraints, do the best you can.’ But in its more insidious form—which is also its most common form—’optimization’ is a kind of deception.

More often than not, ‘I can optimize this’ is a promise to transcend one’s hard constraints, to make possible the impossible. Discerning persons will recognize this is, well, not possible. It’s nonsense. The whole thing rests on sleight of hand. Look over there while I ‘restructure’ (eliminate) the Customer Service division, now look over here at the chatbot application I built for customer service. Does that chatbot perform what the Customer Service division did? Of course not, barely a fraction of it. But the chatbot’s line item is called the same thing (‘Customer Service’) and it costs 95% less. Optimization! The trick is to convince people that you’ve actually transcended a hard constraint (such as labor costs)—you’ve done the impossible—when what you’ve really done is hidden an opportunity cost by moving shit around.

Football’s version of ‘optimization’ is ‘do more with less.’

Like corporate optimization, it’s a form of deception, no matter which way you slice it. At Brentford and Brighton, it’s pouring analytics resources into recruitment, but the tradeoff is every year fans have to suffer the loss of their best players up the food chain, in essence supporting feeder operations for bigger clubs that will never, under that model, become the bigger clubs themselves. At Manchester United, it’s eliminating workers’ canteen privileges and leveraging the club’s historic and global commercial brand to manage PSR constraints, but also—opposite Brighton and Brentford—to sink all of that massive PSR overhead into expensive recruitment mistakes.

In either case, you can’t escape tradeoffs. Performances and outcomes against resources can always be a little better or a little worse based on hundreds of different factors, but ‘optimization’ itself is neither a coherent nor realistic instrument with which a football club can advance itself in the game of football. Whether statistically or verbally, it’s all rhetoric, all misdirection, all tradeoffs dressed up as intelligent leadership.

My concern with what’s happening at Spurs right now—indeed, with what’s been happening—is foremost that we have an overbalance of corporate-thinking people in club leadership, thus an underbalance of football-thinking people. Interestingly, I think Thomas Frank is really a combination of the two. Of course he’s football-thinking, but his idea of football is extremely ‘corporate’ in this sense: he believes in ‘optimization.’ He believes you can violate the second law of thermodynamics. He believes you can ‘hack’ football.

By stark contrast—for purely illustrative purposes!—his predecessor at Spurs was always clear-sighted and open about tradeoffs. ‘It’s just who we are, mate’ wasn’t only about identity and bravado. It was also, fundamentally, about understanding and accepting the reality of tradeoffs, that you can’t ‘fix the defence first and then work on the attack,’ that you can’t hack the game. I suspect this is one of the reasons our players appear to have loved playing for Postecoglou, even when things weren’t going well. As the primary practitioners of football, they understand better than anyone that you can’t be two places at once on the football pitch, you can only choose your tradeoffs and move forward from there.

Let me come now to the point of this discussion.

One of the tradeoffs that a corporate ‘optimizer’ approach to the game is poorly positioned to see is that whereas certain business decisions can be corrected, as can many football decisions, in the end you can’t rewire the psyches of the players.

In other words, if you start thinking about human footballers as assets, fungible tokens, or bundles of metrics, one thing you’re blinding yourself to is that they are also significant and complex agents in this whole network of club performance. They can’t be manipulated in straightforward or linear ways like other things in business, in part because they are highly non-fungible, in the sense that it’s a lot easier to find another office manager than it is to find another Micky van de Ven. You have before you a group of elite athletes who’ve gotten to where they are by learning and adapting to a concatenation of intense pressures. The group of players before us—like all such groups of top-flight players—are not necessarily all the strongest, or the fastest, or the most technically gifted (though many of them are also that); most importantly, they’re the ones among countless others who have the physical and technical gifts to play the game at the highest level who also possess the cognitive and psychological gifts to excel in elite sport.

Let me be blunt: Elite footballers want to play elite football. They do not want to be set up with the mentality that their opponents are superior and the best thing to do is sit back and let their opponents dictate the run of play, and then react to that. Cuti Romero won the World Cup with Messi, who calls him the world’s best center back. Xavi Simons came through La Masia and PSG. Mathys Tel was a prodigy at Bayern Munich. Randal Kolo Muani is a French international who’s gotten regular minutes in probably the most talented pool of attacking players on the planet. Whatever you think of their level right now at Spurs, they think it’s better than this, and they’re probably right.

When Spurs’ executive leadership make decisions—like the decision to appoint Thomas Frank—based on ‘optimization’ and the fantasy of ‘hacking’ football—the one big factor they’re failing to account for is that we have a roster of players who will absolutely see themselves as ready to play elite football, and with good reason: many of them already know what that’s like because they’ve done it before.

We are in dangerous territory now.

As a fan, I don’t think of our players as assets, but as components of building a successful football team. I think if we want Spurs to become a successful football team again—winning Europa was what success looks like; finishing regularly in the top 4 and challenging for the title under Poch was what success looks like—then we’re going to need to hold onto our very best players, many of whom are prime age. That point stands even if you only view them as assets.

Romero, van de Ven, Kulusevski, Udogie, Spence, Porro, Simons, Kolo Muani, and now Gallagher are top-end players with options, regardless of their contracts (unhappy players locked in contracts is not healthy for a football club or its performance on the pitch). They can all slot into any number of Champions League teams. Some of them have already had attention from the likes of Real Madrid, Barcelona, and Manchester City. Others, like Vuskovic, Gray, Bergvall, Sarr, Odobert, and Tel are not yet prime-age, but promising and increasingly experienced young talent in whom we’ve invested and whom we should want to nurture.

Again, people can quibble with how good I think these groups of players are, but most of them are coming from elite footballing backgrounds and have already proven themselves in elite club and international competition. The fact that they haven’t looked good this season under Frank—particularly that our attacking players have been systematically starved of opportunity—is, in my view, not surprising, and not a basis for negative judgment against them.

We’ve all watched many of these players show frustration this season, in some cases uncharacteristically. Micky van de Ven, for example, is not shy of a battle, but also not usually one to lash out like he has so often lately. Even Romero, despite his reputation, had settled in to a composed figure, with minimal cards, in the seasons prior to this one. The frustration on the faces of Spence and Tel every week, for example, has been concerning, as has the mis-usage of Tel, a player who pointedly came to Spurs from Bayern in search of more game time, and who’s seen his minutes stagnate even as he’s scored goals for us lately in his limited appearances.

I call this a dangerous situation because I fear that, results and performances aside for a moment, the club’s ‘more with less’ mentality, epitomized in the appointment of Frank and manifested in the unambitious football we’re playing, has begun to alienate our players. We speak often of the club’s relationship to the fans, of fan alienation from the club, but not enough about the perils of player alienation. I don’t say this out of concern for the players themselves (though I do care about them as human beings too, even as I understand they’re paid millions in part to figure out a way to show up to work and put their best foot forward even when they’re not feeling it), rather out of concern for the club and our prospects for success if we’re giving our best and most promising players incentives to start looking elsewhere to move their careers forward.

For this reason, I think, it’s imperative that we—as fans, and I hope as a club—can get behind our players right now and take the rest of this season to show them that Tottenham is their home, that this is the best club for them to achieve their footballing dreams. It’s hard to see it that way right now, I know, but we have to get ourselves there or we’re going to regret it.

In practice (a bit of burying the lede here), this means parting ways with Thomas Frank as soon as possible and bringing in someone—literally anyone, for now—with the directive to let this group of players play football: build-up, possession, control, attack. Let this group of players walk onto the pitch like the titans they are, rather than as pieces on a magnetic tactics board, an extension of some football ‘optimization’ scheme in which they’re not allowed to take the game to their opponents.

And for us, as fans, it means vocalizing our support for the players despite our frustrations, if we can manage it. I’m encouraged to have seen this already among our match-going fans and hope it continues. If there’s one thing this divided and beleaguered fanbase can unify behind, surely it’s remembering who we are.

Yesterday’s dour 0-0 draw at Brentford had something of the Spiderman meme about it, two sides mirroring each other in hoisting the ball aimlessly into the air and dropping back to defend, against what exactly we’ll never know.

The performance brought into relief an uncomfortable truth about the recent trajectory at Spurs: the club have been working hard to become Brentford, and might already have succeeded.

What do I mean when I say Spurs have ‘become Brentford’?

To understand this you have to go back to May of 2021, when the former chairman Daniel Levy released what’s become the infamous ‘Spurs DNA’ message. ‘As a Club we have been so focused on delivering the stadium and dealing with the impact of the pandemic,’ Levy confessed, ‘that I feel we lost sight of some key priorities and what’s truly in our DNA.’ Levy went on to describe the Spurs ‘DNA’ with language that haunts Spurs fans in our nightmares and dreams to this day: ‘We shall focus on the recruitment of a new Head Coach. We are acutely aware of the need to select someone whose values reflect those of our great Club and return to playing football with the style for which we are known – free-flowing, attacking and entertaining…’

One month later, the club announced the appointment of Nuno Espirito Santo as the new Head Coach, replacing Jose Mourinho. By November, Spurs brought in another icon of defensive football, Antonio Conte.

Audere est facere. Do dare îs to do. Spurs DNA. Free-flowing, attacking football. What are these markers of club identity supposed to mean?

It wasn’t until the appointment of Ange ‘It’s just who we are, mate’ Postecoglou in the summer of 2023 that Spurs executive leadership made good on the whole ‘Spurs DNA’ thing, two and a half years after Levy’s public reckoning about club values.

It’s important to bear in mind, however, that ‘free-flowing, attacking football’ is not just about entertainment, a point that Postecoglou found himself repeating ad nauseam across both of his seasons at the helm at Spurs. But you don’t have to take Ange’s word for it. If you cross-reference the top teams in the table with the top performers in possession, xG, and goals, you’re going to see mostly overlap. You’ll also see those top-ranking teams perform at the top in defensive metrics, too, but they do that in large part by having most of the ball against their opponents and keeping it in the attacking third, with some situational variation on the broader theme. Even Arne Slot’s ‘moderated’ Liverpool team that won the title last season did so with 57% of possession (2nd only to City) and the 3rd most touches in the attacking third.

High-End Pragmatism

‘Free-flowing, attacking football,’ then, is not just meant to be daring, dashing, and entertaining; it’s one of the most successful forms of tactical pragmatism. It’s just not what most people think of as ‘pragmatic’ football, because it’s the pragmatism of the top of the table, not the pragmatism of everyone else.

When you’re one of the richest clubs in world football, and you’re a regular qualifier for European competition—including a top-4 most qualifying Premier League team for the Champions League in the past 10 seasons—you can afford and attract a suitably deep, talented, and technical squad to play front-foot football. You don’t have to resort to the reactive pragmatism of mid-table sides, like playing wholly on the counter, or from set pieces, or overbalancing your focus on out-of-possession positioning and committing few players forward.

Put simply, there’s a high-end pragmatism of control, which mostly aligns with the spirit of the Tottenham ‘DNA,’ and then there’s mid-table pragmatism, which is more about letting your presumptively superior opponents dictate the course of the game and reacting in key moments to steal their thunder.

Becoming Brentford

What we’re seeing now at Spurs is a consequence of the club yet again contradicting its motto, just like it did when it hired Nuno in 2021. To dare is to do is not a motto about mid-table pragmatism. More importantly, it’s not a motto about fear of downsides. It’s a motto about audacity in the face of doubt. If you don’t dare, you can’t do.

In more concrete terms, Levy and the club built up a verbal head of steam in 2021 and eventually scrambled their way to something resembling audacity in 2023 when they appointed Postecoglou. But the moment for real daring wasn’t in the appointment itself, but the trying period last summer after we’d finished 17th in the league but won the Europa League trophy. That was the moment that required real courage, real daring, real conviction and clarity of purpose, because every self-styled ‘rational’ mind in English football was saying the ‘rational’ thing was to sack that manager for a horrendous league performance and find someone more ‘pragmatic.’

I’ve already written enough about the explanatory failure of that way of thinking, but here I’m calling attention to something different but equally important, if not more so. Sacking Ange when we did was not a rational act; it was an act of cowardice in direct contradiction of the whole idea of 'to dare is to do.’ It was an act of acquiescence to the conventional wisdom of English football that says Spurs are a banter club who must not aspire to loftier heights—we’ve already gotten too big for our breeches!—who must not dare to play high-end football with high-end objectives, who must not talk of winning things. No, no, no, Spurs must aspire to mid-table pragmatism, and anything greater is foolishness, emotional indulgence, delusion.

Overcome by cowardice and fear, our senior leadership faced just one test of their resolve—a wonky and aberrant 17th-place league finish off the back of the first trophy win in 17 years—and backed down. In that very moment—the decision to walk away not only from a manager who took the team from 8th to 5th in his first season and delivered an historic trophy in his second season, but perhaps more crucially, away from the commitment to playing ‘free-flowing, attacking football’ at the top of the table.

That was the moment the club announced—not in words, but in actions—that we would strive to become Brentford. Everything else that’s unfolded since was a predictable consequence of that failure of conviction, and when I say ‘predictable,’ I bring receipts.

Of course, ‘Brentford’ here is merely a convenient stand-in (all the more so because, well, we hired their manager, drew 0-0 against them yesterday, and are behind them in the table) for the type of club—Brighton and Bournemouth also fit the bill—that doesn’t regularly compete in Europe and must fight for mid-table positioning in the Premier League. The type of club that doesn’t have the budget to build a Champions League squad, so must work with extreme efficiency at the margins with smart, data-driven recruitment and a best-case scenario of reselling well-recruited talent to clubs like Spurs, or what Spurs used to be.

Becoming Brentford in this sense is not just about sitting 12th in the table with one of the worst xGD in the league, scrapping for 1-0 bailout wins against Palace or 0-0 draws against…Brentford, though such telltale signs of voluntary decline are a little too on the nose. Becoming Brentford also means adopting the conventional wisdom of mid-table pragmatic football while the top-end sides we used to compete with in the not so distant past—we finished above Arsenal for 6 of 7 seasons until a few seasons ago and defeated Manchester United in a European cup final last season—are right back up at the top of the table. Becoming Brentford means signaling to the football world that it’s a better play for the likes of Cunha and Mbeumo to head to the team we defeated 4 times last season than to consider coming to Spurs, better for Eze to hold out hope of joining Arsenal than joining Spurs. Without rattling on about this at too much length, becoming Brentford means being so scared of moving up in the football world that we’ve knowingly accepted—arguably we’ve intentionally effected—a status downgrade as a result.

For the past few months I’ve been saying we must not fall into this trap. In early August, I wrote ‘…Spurs can’t afford to keep slipping back into mid-table, by which I mean Spurs’ senior officers need to stop looking down the pecking order at Brentford and Brighton with envy and start looking up at Liverpool and Arsenal with purpose.’

Now, in January, the stakes of falling into that trap are becoming too clear for comfort. We killed the closest project we’ve had since Poch’s tenure to the spirit and the outcomes of audere est facere and replaced it with the spirit and the outcomes of mid-table pragmatism. We now risk losing the interest and loyalty of what top-end players we still have because our football is wholly unambitious. The gap between finishing 17th in the league last season and the spells of much better football we played along the way—including doing what we had to do to win the European League—is, at the very least, explicable. But you can’t divide 0 by 0.

The formerly Prussian city of Königsberg was situated across and between streams of the Pregel River, creating small islands within the city, requiring seven bridges to access various parts of the city, like so:

In 1736, the Swiss mathematician Leonhard Euler—widely credited today as the founder of graph theory—posed and resolved what’s known as the Seven Bridges of Königsberg problem. Euler wanted to find a way to walk through the city by crossing each bridge once and only once. In the end, he solved the problem by negative proof, demonstrating that the task was impossible.

What’s interesting about his solution, however, was the way he went about it. The special insight he had was that for the purpose of the problem, the actual geographic layout of the city didn’t really matter; all that mattered for the route one might take is the order in which one crossed the bridges. This insight led Euler to abstract the map of the city into a simplified form that would allow him to see more clearly whether such a route—under the conditions of the problem—was possible. Here’s a sketch of what he came up with:

This was foundational for the field of study we call graph theory today, because it introduced with impeccable clarity the idea that a graph is a model of the relations between real things. As such—thinking of a graph as a model—all graphs require a theory of what they represent, how, and thus which features of their target system (the real thing they’re representing) are kept in the graph and which are not. For Euler, it wasn’t necessary to model the seven-bridge route along the actual streets of Königsberg, so he left those out. Instead he included only the nodes (four green dots representing spaces separated by water) and edges (seven black lines representing the bridges) of the notional route, represented in the right-sided image above.

Euler’s insight turns out to be useful for how we interpret pass maps in football, as pass maps are just one kind of network graph, with nodes (each player on the pitch) and edges (each line representing a pass). Here are the PSG (left) and Spurs (right) pass maps for the 5-3 loss in the Champions League last Wednesday:

If we look at this standard format for football pass maps from a graph-theoretical perspective, a couple of things stand out, both of which football pundits and fans largely take for granted.

One is that producers of football pass maps tend to assume the actual space or geography of the pitch is important information to represent in the graph. In other words, most pass maps are overlaid on an illustration of a football pitch (as above), with the position of the nodes (players) roughly corresponding to the areas on the pitch where the passes were played. This helps us understand, for example, whether most passing took place in our own third or in the attacking third, whether we used both sides of the pitch, and which team was more spread out or more condensed in their passing network. All of this is useful information, but it also reflects a kind of pre-Eulerian idea about graphical representation, namely that the most important thing to represent in the network is the geographical space on the pitch where players receive the ball.

The other common feature of pass maps that reflects an assumption about which information is most important to represent is that pass maps tend to offer thicker or shaded lines to represent higher pass volume between players. In network analysis, the ‘intensity’ (for lack of a better way of putting it) of the edges between nodes is called ‘weight.’ In football terms, weight would be equal to the number of passes between players, where players are nodes and pass lines are edges. Again, this reflects an assumption that it’s particularly useful to see some representation in a pass map of the quantity of passes between player pairs. We like to know, for example, if there are 15 passes between Vicario and van de Ven but only 2 between Bentancur and Palhinha, and we take such data as a basis for explaining what we’re seeing on the pitch.

However, representing passing networks in the conventional style of pass maps is just one way of doing things, not necessarily the best or most informative way.

For example, here’s another way of representing Spurs’ pass map against PSG, a directional network graph based on the same underlying data as the conventional pass maps I’ve included above:

Here’s the same graph, but with a focus on Xavi’s connections (#7), notably a part of the network graph that’s not very dense with edges:

Here’s another, less pretty network graph for the same underlying data (thanks Gephi):

Notice that in each of these graphs, spatial location on the pitch is not indicated; the networks are not oriented on an illustration of a football pitch. In fact, the underlying data only includes a source of each pass, a recipient of each pass, and the number of passes between source and recipient (weight). These are not representations of passing distance, in other words, even if conventional pass maps frequently (sometimes misleadingly) imply as much. Instead, the three network graphs I’ve made reflect relation within the passing network itself, rather than to locations on the pitch. This is, if you like, a Eulerian approach to pass maps: let’s look at the network abstracted from the geography of the football pitch.

You might be wondering, then, why anyone would want this kind of pass map instead of one that indicates spatial position on the pitch. To answer that question, let’s look at the network’s underlying data.

As I mentioned, the only data I’ve used (the only data I’ve been able to download) for these graphs is player source, player recipient, and number of passes between each player pair. But we can also derive further network data from the underlying data by calculating various network behavior metrics, such as eigenvector centrality (but let’s be honest, I don’t do this by hand, I use network analysis software such as Gephi). Stick with me for a moment…

Eigenvector centrality (sometimes called eigencentrality or prestige score) is a way of measuring the importance or influence of a node within its network, which is to say deriving an eigenvector centrality score from a passing network can give you a measure of who the most influential players are in the passing network.

Keep in mind that ‘important’ or ‘influential’ here are not synonyms for ‘most passes.’ In other words, whereas conventional football pass maps tend to emphasize pass volume as an important metric—that is, they tend to visually emphasize pass volume in the way they represent the passing network—metrics like eigenvector centrality give us more than just who’s passing the most to whom. They give us a richer sense of how well a player enmeshes themselves in the passing network, or in football terms how how effectively they combine with other players who themselves are in a position to combine effectively with others, shaping the flow of the network. One way of interpreting this is that if you think of the players you most want on the ball making things happen, those are the ones you’d like to have the highest eigencentrality, as it reflects the extent to which they’re including themselves and being included in the network with other highly-involved players, imposing themselves on the in-possession game.

Accordingly, we can represent a pass map statistically, not just visually. Here’s what that looks like for Spurs against PSG recently (where the ‘ID’ column identifies the player by shirt number and the table is sorted by the ‘Eigenvector Centrality’ column to the right of the ‘ID’):

As you can see from this statistical representation of the pass map, focusing on eigencentrality, the most influential passer in the network against PSG was Vicario, the least influential was Xavi (came on as a substitute). Sarr, Porro, Romero, and Richarlison round out the top 5, but Kolo Muani and Bergvall—two from the starting 11 you’d very much want on the ball—were among the least influential.

Though our defending against PSG was woeful (which can happen against the best team in the world) and the effort amounted to a loss, the post-match commentary generally has been pretty favorable, finding much promise in our in-possession and attacking performances. Against PSG, Frank also elected not to use the now-infamous ‘Bentinha’ combination of Bentancur and Palhinha in the double pivot.

So let’s compare the PSG passing network to the passing network against Arsenal, which was arguably our worst performance of the season, and one in which ‘Bentinha’ featured prominently. Here are the conventional pass maps (Spurs on the right):

And here are my network graph and statistical breakdown, based on the same underlying data scraped from WhoScored:

Right away you can see the network graph looks a lot different—is distributed differently—from the graph from the PSG match. And if we look at the statistical table, again sorted by eigenvector centrality, we find the most influential passers against Arsenal were Palhinha, Udogie, Romero, van de Ven, and Vicario. We don’t need to see the pass locations on the pitch to understand from this data that our most important passing in the network happened exclusively among defensive players (though Xavi comes next, and that shows up far less prominently in the conventional pass map). Likewise it’s worth noting that the least influential passers were Kolo Muani (came on as a substitute), Odobert, Sarr, Kudus, and Bentancur, then Richarlison.

Just because this is data doesn’t mean interpretation is straightforward here. But it does illustrate another way of seeing passing networks, which emphasizes different information than we typically get from standard pass maps. You already know that against Arsenal we were stuck in our own half and never worked the ball forward, just as you know against PSG we did a bit better on that score, finding Richarlison and Kolo Muani in dangerous positions from time to time.

However, the interesting question about eigenvector centrality as a passing network measure is what it might explain. In other words, did our attacking players have less influence in the passing network against Arsenal because we couldn’t get the ball to them, or did we struggle to get the ball to them because they weren’t capable of getting integrated into the passing network?

Even if we can’t conclusively answer this question, just posing it brings into view the deep relationality of network and result. We often seek a causal arrow that points in one direction when in reality, when we watch a match, we’re watching a constant feedback loop. For example, failure to get Xavi involved in possession means failure to progress the ball means failure to get Xavi involved in possession, and so on.

In terms of concrete solutions for Spurs, the two sample matches I’ve looked at here aren’t enough to go on, but I do think that looking at passing networks and eigenvector centrality could be a good way of analyzing the extent to which Frank’s personnel choices—particularly ‘Bentinha’—are as influential as we think on our in-possession game.

Ten games into the 2025-26 Premier League season, Spurs sit 4th in the table, having scored the 4th (t) most goals, having conceded the 4th (t) fewest goals, with the 3rd best goal differential. If you showed me those figures in isolation, without context, I’d think we were having a pretty cracking season thus far.

But of course those of you who’ve watched Spurs this season probably don’t think we’ve been as good as those figures suggest, and I don’t either.

We could come up with all kinds of notions to explain the gap between such quantitative outcomes and qualitative performances—key players are ‘bedding in,’ others are injured, Frank needs time, we’re ‘focusing on defense first before focusing on attack,’ and so on. But I think the best way to describe Spurs so far is that we’ve been very lucky.

And of course I want to specify and quantify what I mean by ‘lucky.’

‘Lucky’ in Attack

Spurs’ 17 goals scored is right toward the top of the league, but it’s come from an xG of just 10.1. That means we’ve overperformed our xG by a stunning 70%. That’s how you can be both 15th in the league in xG and 4th (t) in the league in G.

We’ve also created 17 big chances in the Premier League and missed only 8 of them, or 47%. For reference, last season’s league average percentage of big chances missed, about halfway through the season, was 64%. So 47% is probably an overperformance, suggesting we’ve been especially clinical with our relatively few chances (Spurs’ 17 big chances created so far in the league is good for 12th).

‘Lucky in Defense’

We’ve also been especially lucky defensively. Spurs have the 5th highest xG conceded in the Premier League—14.6—while allowing the 4th (t) fewest goals—8—which is a wild defensive overperformance of 6.6 goals.

How do you end up conceding just 8 goals from 14.6 expected goals against? One way of answering that question is by looking at what our opponents have been doing with the big chances they’ve had against us, and I think this gives us a pretty solid answer.

Our opponents in the Premier League have mustered 21 big chances against us and missed 17, a miss rate of 81%. Assuming a league baseline miss rate (as above) of 64%, if our opponents performed on big chances against us this season at the league average—rather than underperforming it by 17%—that would account for 3.6 of the 6.6 goals by which we’re overperforming our xGA. Put simply, we’re not defending particularly well, conceding lots of chances, but our opponents are missing their chances at a much higher rate than is typical for the Premier League.

If you add to this the fact that Vicario leads the Premier League in goals prevented (5.2), by a margin of nearly 2 goals over 2nd-place Henderson—then you can account for the rest of the 6.6 goal defensive overperformance (and then some):

It’s pretty reasonable to conclude, therefore, that the reason we haven’t been punished in results for our poor defending is a combination of opponents missing their big chances against us at an unusually high rate and Vicario preventing a full 33% more expected goals than the next best keeper in the league.

‘Lucky’ Head-to-Head

To get a more precise handle on all of this, I’ve put together a table that tracks and aggregates how we’re performing head-to-head against each opponent across all competitions—a total of 15 games—across the metrics of xG, xGA, big chances created, and big chances conceded (also taking percentages of big chances missed for us and for our opponents). I’ve selected these metrics as proxies for how much we’re creating and conceding against each opponent and in the aggregate through what we can estimate as 25% of our season so far.

Some very interesting (and concerning) observations stand out.

For one, if you expand the scope of analysis for all competitions, you can see that our opponents are missing the big chances they create against us at an even higher rate than in the league alone, 84%. Again, as a benchmark, this is a 20% underperformance on last season’s Premier League average.

Two, our cumulative xGA of 22.11 is almost double our actual GA of 12, a massive defensive overperformance. Vicario has already prevented 1.89 goals in the Champions League from 3.89 total xGoT faced in 3 games. So again, if you take opponents’ 20% underperformance against us in big chances and assume a league average big chance conversion rate on the 31 big chances we’ve conceded across 15 games, that would account for 6.2 of the 10.1 goal difference between what we’ve conceded in xGA and what we’ve actually conceded.

Following the same method I applied above to just our Premier League games, adding Vicario’s 1.89 goals prevented in the UCL to his 5.2 goals prevented in the Premier League gives you a total of 7.09 total goals prevented by Vicario. Adding Vicario’s prevention total (7.09) to the 6.2 more goals an average opponent big chance performance would’ve scored against us, we can account for 13.29 goals, a little more than our 10.1-goal overperformance on xGA.

I know narrativizing calculations this way isn’t great, but the point here is, again, we can explain the ‘luck’ of conceding 83% fewer actual goals this season in all competitions than our xGA (12 GA out of 22.11 xGA) by our opponents scoring 6.2 fewer goals from big chances than they should have and by Vicario preventing 7.09 goals. Of course there’s some statistical overlap between opponents’ underperformance on big chances and Vicario’s prevented goals, but I think the combination of the two gives us a pretty good idea of what’s been happening, and matches the eye test pretty well too.

It’s also worth noting that we’re almost always ‘losing’ on xG this season. Across all 15 games, we’ve ‘won’ on xG just 5 times—against Burnley, West Ham, Villarreal (by just .06), Doncaster, and Aston Villa—while drawing just short of even with Brighton and Everton. When we’ve played against higher quality opponents—City, Bournemouth, Monaco, Newcastle, Chelsea—we’ve been particularly open defensively while creating very little. Our cumulative xGD across all 15 games is -6.68, a worrying sign. On average we’ve created 1.03 xG and surrendered 1.47 xGA per match, a losing formula. Our cumulative xGD drops to -7.57 if you exclude League One Doncaster, and -8.21 if you exclude Doncaster and relegation/promotion-zone opponents such as Burnley, West Ham, Wolves, and Leeds.

Finally, this is all happening while we overperform on our own big chance conversion, taking 15 of 26 of our big chances, or 58% against the 64% baseline I’ve been working with. We’ve scored a total of 23 goals from 15.4 xG, an overperformance of 7.6 goals.

When you add it all together on the defensive and offensive ends of the game over 15 matches, we have a +11 goal difference from an xGD of -6.68, a massive 17.68 goal ‘lucky’ swing. If you distribute those goals evenly over 15 matches, that’s 1.18 goals per match. Our average win margin is 2 (putting aside our 4 draws), so that gives you a rough idea of what our results would look like with a -1.18 goal swing applied to every match we’ve played.

I don’t want to overdo it here with micro-sampling, and I don’t want to be mistaken for ‘Frank Out,’ because that’s not what I’m arguing for. But I think, as I predicted, the narrative-driven rush to declare that Frank has made Spurs ‘more defensively solid’ was both wrong and has obscured people’s ability to analyze what’s really happening.

A quarter of a season is, in football terms, a fair enough chuck of games to develop useful intuitions about the underlying numbers, especially when the underlying numbers align pretty well with the eye test, as I think they do. My view remains that Spurs’ defensive performances are not well explained or predicted by who the manager is, but by the personnel available to play the matches, particularly the center backs. And that our inability to create, especially from open play, has meant that Spurs have been largely at the mercy of opponents just not taking their chances. We should not expect to keep pulling out results—even narrow losses—under these circumstances.

What we’re seeing so far, then, is virtually the opposite of what we saw last season under Postecoglou. Spurs’ underlying numbers last season were highly promising—far better than results (even as results were quite good through the first quarter of the season last season too). What changed the trajectory last season was a series of simultaneous, catastrophic injuries to key players, never relenting until the last quarter of the season, when Ange held those same key players out of the league to focus on winning the Europa League. With any such event or series of events, underlying numbers can take a turn. The question now is whether we have a positive event or series of events—whether tactical or otherwise—intervening and changing the direction of our underlying numbers. Because again, if these performances remain constant, results will not likely be good.

Only 6 teams in the Premier League have created fewer expected goals than the 8.3 Spurs have managed to scrape together so far this season: Fulham, Sunderland, West Ham, Wolves, Villa, and Burnley. Not great company on the whole.

But 8.3 is our total expected goals figure in the Premier League. When I talk with other Spurs fans, I hear emerging concern that Spurs under Thomas Frank aren’t creating enough from open play, that we’re too reliant on set pieces to generate what little goal threat we’ve produced. So I’ve taken a prompt from Milo of the excellent The Game is About Glory pod and had a closer look at how Spurs so far compare with Frank’s previous Brentford teams in terms of reliance on set pieces to create chances.

A Quick Note About Our Defending

One of the reasons it’s important to keep an eye on our expected goals production, even this early in the process, is that we’re not currently defending well enough to get by on limited chance creation.

Our expected goals against is actually looking quite good at this point after the Villa match. The 9.1 expected goals we’ve conceded is good for 5th in the table (behind Arsenal, Newcastle, City, and Leeds, which is pretty good company in this regard).

We look less defensively convincing, however, when you add in the Champions League performances, mainly the 2.52 xGA conceded against Bodo/Glimt and the 2.45 xGA conceded last night against Monaco. In the Premier League, Spurs average an xGA / 90 of 1.13, and in the Champions League an xGA / 90 of 1.82. The former is just 8 games and the latter merely 3, but eyes-wise it’s clear that opponents have been able to get into dangerous positions against us on a routine basis.

So before I dive into our meager chance creation numbers thus far, I think it’s important to point out some specifics about how we’re over-performing even our defensive underlying numbers. In the Premier League, Vicario sits 3rd among keepers with 1.8 goals prevented:

Additionally, his 1.89 goals prevented over just 3 games so far puts him in the 88th percentile among Champions League keepers. That’s nearly 4 goals prevented by Vicario alone over a total of 11 games in the Premier League and Champions League.

Perhaps even more telling is the fact that over 12 total games (PL, UCL, and EFL Cup), Spurs have conceded 20 big chances, of which opponents missed 17. That’s an 85% big chance miss rate among all of our opponents this season.

To put that into perspective, when I looked into big chance miss rates in the Premier League in January of last season, I calculated a league-wide miss rate of 64%, with the highest miss rate (Everton) at 80% and the lowest (Brentford and Spurs, curiously) at 56%. An 85% miss rate for our opponents is, by comparison, an extreme case, which tells us a good deal about why our results have been quite good this season but our performances and underlying numbers less so. The game against Monaco last night was a case in point, I think, with Vicario pulling out some incredibly saves and ex-Gooner Balogun pulling out some incredible misses.

Set-Play vs. Open-Play xG

Given our defensive vulnerability as indicated above, Spurs should not expect to keep getting results through ‘defensive solidity’ (such as it is!) alone. We’ll have to start creating chances at a much higher rate and in more diverse ways to have a shot at remaining in Europe next season and challenging for silverware this season, I think.

To help us understand more about what’s going on so far with Spurs’ attack, I’ve looked into how Thomas Frank’s Brentford performed over his last 3 seasons at the club. In particular, I’ve looked at the percentage of expected goals Frank’s teams have created that came from set pieces versus open play, and compared them to Frank’s Spurs so far. This will provide a basis for ongoing comparison as we get more data from this season. Some of what follows is as you’d expect, while some is perhaps a bit surprising.

First and foremost on this score, for the past 3 seasons, Frank’s Brentford teams were not only among the top in the Premier League in set-piece xG, but also had among the highest percentage of their total expected goals come from set pieces. In 2022-23, a full third of Brentford’s xG came from set pieces (only Everton had a higher percentage) (the 3 charts below are sorted by % set-play xG):

In 2023-24, that figure dropped to 27%, though Brentford still had the 4th highest percentage of xG from set pieces:

The following year, 2024-25, Frank’s Brentford once again had the 4th highest percentage of xG from set pieces, 28%:

Across 3 seasons before Frank moved to Spurs, his Brentford teams averaged 29% of their expected goals from set pieces. As you can see, I’ve provided standard deviation figures for each chart so you can see that although Frank’s teams aren’t alone in taking roughly a third of their xG from set pieces, they’re consistently among those who rely heavily on set-piece chance creation. With the exception of last season, in which looking for competitive advantage through set-piece ‘innovation’ is something a few top teams—mainly Arsenal—started doing, the teams in the top third in % set-piece xG tend to be lower- to mid-table teams. This is to suggest that set-piece xG percentages may be high in part because total xG is relatively low (I’ve given you all those figures above in case you’re curious and want to dig further).

The big question is how this compares with what Frank’s Spurs are looking like so far. And here’s where you might be surprised (though not in a good way, if you’re a Spurs fan). Spurs have created just 1.77 xG from set pieces, 15th in the league. And our percentage of xG from set pieces is similarly low, at least for a Frank side, at 21%, 13th in the league:

Of course, we’re still dealing with early-season small samples here. We can grab a few more games by including Spurs’ UCL and EFL Cup games, all 12 games we’ve played so far. When we do that, we see the percentage of set-piece xG increase to 33%, much more in line with Frank’s norm over the last 3 seasons at Brentford:

What to make of this?

If we widen the context of Frank’s Brentford teams in terms of xG creation, we can begin to assemble a few provisional conclusions, or more like a series of focused questions. In 2022-23, Brentford were 7th in the league in total xG (57.9); in 2023-24, they were 8th (59.8); and last season, they were 8th (60.2), averaging 59.3 xG.

Over that same period, Spurs were 8th (57.8); 6th (69); and 9th (59.7), averaging 62.2 xG. We can bring in all kinds of explanatory context here, but as you can see the figures are pretty similar.

On the whole, Frank’s Brentford teams generated plenty of xG (I’m sure if that weren’t the case then Spurs wouldn’t have hired him). Although, compared with Spurs—and with most of the best teams in the Premier League, save Arsenal—Frank has relied on set pieces to generate nearly a third of Brentford’s total xG. As the charts above show, in Frank’s last 3 seasons with Brentford, he consistently relied on set pieces for xG creation among the very most in the league.

Somewhat surprisingly, so far at Spurs, we’re not generating much xG either from open play or from set pieces. Which is to say set pieces aren’t as much of a strength right now in reality as people seem to think.

Looking into all of this has left me with a few questions that I think are important to ponder as the season unfolds:

1. How much of Brentford’s open-play xG under Frank was a function of how teams approach a side of Brentford’s caliber, versus Spurs’ caliber, and to what extent is this transferrable to a Spurs context? In other words, can you manage the likes of Spurs the way you manage the likes of Brentford and still have attacking success?

2. Now that set piece micro-advantage-seeking is trendy in the Premier League, how much more competitive advantage can a set-piece reliant side squeeze out of this approach? In other words, if everyone is now re-attuned to long throws, rugby corners, and advanced set-piece routines, will we see diminishing returns on such strategies as defenses catch up to the trend?

3. To what extent can Frank adapt his in-possession approach to be less reliant on set pieces to generate xG?

Even with injuries to key attacking players such as Maddison, Deki, and Solanke, the summer window brought in plenty of elite-level attacking talent (Kudus, Xavi, Kolo Muani). This is a side that we should expect to put up top-5 attacking numbers.

The most recent episode of The Extra Inch podcast contains a good discussion of how we should interpret the fact that Spurs have significantly overperformed their underlying numbers so far this season. By this I (and presumably they) mean that through 7 games we’ve nearly doubled our expected goal figure of 7.4, scoring 13 goals, while we’ve conceded 3.7 fewer goals than our expected goals against figure of 8.7. In terms of expected goal difference—which is typically heavily weighted in expected points models—Spurs sit 13th in the table:

The entire episode of TEI is worth a listen, but the gist of where the discussion landed was well summed up by Nathan A. Clark, whom I’m paraphrasing here.

Nathan suggested that expected goals figures are best used to draw conclusions about large sample sizes, but when you’re looking at just a few games it’s tactical observations—not numbers—that tell the most reliable story.

While I agree with Nathan’s broader point that we should always be careful about drawing certain types of conclusions from small sample sizes, I would challenge the second part of his claim: that tactical observations are better than numbers for understanding what’s going on at this stage in the season. This isn’t a bone I’m picking with Nathan specifically or particularly—in fact I think Nathan is truly one of the best tactical analysts going and that we in the Spurs fanbase are fortunate that he happens to be a Spurs fan—because it’s a broader truism in football that you can dismiss the underlying numbers if you don’t have big data and that tactical observations by informed people—you might even say experts—are the gold standard.

But again, there’s an aspect of this way of thinking that I’d like to challenge.

Variance Exists Even If You’re Not Counting

It’s probably, on balance, a good thing if people are default skeptical of small-sample numerical analysis. At the same time, there’s a difference between descriptive statistics—summaries of a dataset’s key features, such as mean or standard deviation—and inferential statistics, which entail drawing inferences about things beyond the dataset. For both types, small samples are not as good as large samples. But when people use small samples descriptively in football, in effect to say ‘here’s what’s going on so far,’ they’re not doing anything more irresponsible than people making tactical observations, just because the former are putting observations in numerical form.

In other words, tactical observations constitute a sample size, even if tactics analysts aren’t counting or quantifying their observations in the way someone working with underlying numbers is doing. It follows that all the forms of variance and uncertainty you can get from a small-sample dataset—e.g. 7 games of xG—you can also get from making tactical observations about 7 games of football.

We can drill down a bit more to see how this logic works. When we say a small-sample dataset—let’s stick with xG here—is unreliable because it’s not broadly representative enough or could lead to false positives and negatives, what we’re really saying is just because stuff is happening on the football pitch right now doesn’t mean that stuff will keep happening that way in another 7 games, and then another 7 games after that, and so on. When we produce a figure like xG, we’re performing a kind of reduction. That is, we’re accounting for a bunch of stuff that happens on the pitch—shot location, shot angle, player positioning, etc.—and translating (or reducing) it into a number.

Likewise, when we’re making tactical observations, we’re accounting for a bunch of stuff that’s happening on the pitch, and reducing it to a verbal or visual representation of our observations instead of a number. Just as xG doesn’t account for every variable and sequence of actions on the pitch when recorded, a series of observations about pressing shape or buildup patterns doesn’t account for every variable, like whether the pitch is extra slick that day or whether one of the players was struggling with the pressing scheme, so his positioning is not what the tactical setup called for. You can get on-pitch weirdness from any sample of numerical or visual observations, and that’s the basis of wanting larger sample sizes regardless of analytical method.

In both cases—7 games of xG and 7 games of tactical observations—we’re talking about the same small sample size with the same on-pitch variance and the same epistemic pitfalls associated with small sample sizes. Just because statistical analysis involves counting observations and tactical analysis typically does not doesn’t mean we’re not still fundamentally dealing with sample sizes in both cases.

That’s to say when Nathan suggests tactical analysis is matter-of-factly better than statistical analysis for small samples, I take the critique of small sample sizes in statistical analysis, but we still need an explanation of what grounds tactical analysis, or why tactical analysis based on the same small sample is better than underlying numbers.

Threat of the Improbably Precise Figure

Tactical analysts are free to answer my challenge here—I mean it sincerely and with the hope of good-faith discussion—but in the interest of fairness and goodwill, let me lay out one point in their favor.

People working in the areas of scientific and information literacy—two areas that overlap with some of my work—have long and justifiably warned of the threat of the improbably precise figure. In plain terms, an improbably precise figure is figure given in a situation in which the person giving it can’t possibly have worked out their answer to that degree of precision. For example, if I asked you how happy you’re feeling today and you told me you’re feeling 56% happy, I’d think your answer a bit strange and maybe even unnerving, unless in your next comments you walked me through the detailed happiness formula you’ve been using for yourself, and the data you’ve collected to arrive at 56%.

The improbably precise figure can be dangerous when, for example, people use quantification rhetorically, to give the appearance of scientific merit when they haven’t actually done the work, or when marketers say things like ‘this supplement will give you 65% better sleep,’ or ‘our crisps have 30% more flavor than the competitors,’ or ‘our AI tool will make you 40% more productive.’ You can spot an improbably precise figure in such situations in which people are precisely quantifying a sentiment or notion without clearly defining it and showing how they’ve arrived at their figure. This is usually the case when someone is quantifying something with precision that makes no sense to quantify with that degree of precision in the first place (‘Mom, I love you 7.8% more today than yesterday because you helped me with my math homework.’).

So tactical analysis and football fans are generally correct to be wary of the improbably precise figure in football. I suspect that’s partly why there’s always pressure to justify sample size in football statistical analysis and stats-based claims, but virtually zero pressure to justify sample size in tactical observations. Even though, again, they both have the same small sample size, leading to the same potential variance, representativeness, and generalization issues.

The Case for the Specificity of Numbers

But there’s also a case for using numerical descriptions of what’s happening in football, like xG, even when sample sizes are small. People often say ‘you can argue anything with statistics,’ and that’s probably true, but you can also argue anything with words. If you’re going to dismiss the underlying numbers for that reason, it would be hypocritical—and not very smart—to let verbal and graphical descriptions of tactical observations off the hook.

The value of numerical representation lies in precision, though it’s important we take care not to be improbably or irresponsibly precise. The statistician David Spiegelhalter offers some great examples in his latest book of situations in which conveying vague notions in words instead of giving numerical probability led to real disasters. As Spiegelhalter notes, before the Bay of Pigs invasion—a debacle for the US military under President Kennedy in 1961—US military leaders had assessed the invasion’s chance of operational success to be 30%, and reported this to the president as ‘a fair chance’ of success. The president interpreted ‘a fair chance’ to mean ‘a pretty good chance,’ so ordered the invasion, in which most of the 1500 Cuban exiles invading Cuba on behalf of the US were either killed or captured.

Similarly, when we’re trying to describe what’s happening on the football pitch, there are advantages to using metrics such as xG. It’s one thing to say ‘our defensive structure looks solid in this sequence of screen caps’—and that might be true—but then we should still want to know why we’re conceding lots of chances and xG. The generality of ‘our defensive shape looks solid here’ clearly does not account for certain specific statistical observations even if it accounts well for other things. When there’s a clear discrepancy between what we’re observing visually and what we’re seeing in the numbers, the specificity of numbers can help just as much as the generality of visual observations verbally communicated. Remember, just as tactical observations are a basis for aggregated descriptions about what’s happening on the pitch, metics like field tilt or xG or big chances created (and conceded) are descriptive aggregations of what’s happening on the pitch. We’re just talking about two different ways aggregating descriptions from sample sizes.

My point, then, is not that numerical accounts of the game are necessarily superior, but that they’re not matter-of-factly inferior, even with small sample sizes. Likewise, all tactical observations are subject to the same sample size issues even if we’re not explicitly counting them. When we say ‘this team defends in a 5-4-1 shape,’ we’re implicitly saying ‘this team usually defends in a 5-4-1 shape based on x number of times I’ve observed them in this shape.’ There’s an implicit numerical claim in there and it still comes from a sample.

When Lucas Bergvall played Brennan Johnson through on goal against Doncaster on Wednesday, I felt pretty confident that Johnson would finish and wrap up the game. And so he did!

My guess is many of you felt a similar degree of confidence in that moment, because Johnson has looked cool and composed in front of goal for a while now, going all the way back to last season. His 1v1 goal against Doncaster was a thing of subtle beauty, a glance up to see the keeper’s position, a punctuated clip of the ball to lift it beyond the keeper, with a touch of bend to it, angling the shot into the side netting. The keeper’s positioning was fairly solid, leaving Johnson with plenty to do.

So I walked away from the game wondering just how good a finisher Johnson is on paper. Does his finishing profile match his demonstrated technique?

The answer to this question is pretty straightforward but also illuminating, at least in terms of how Johnson creates value on the pitch.

The first thing to point out is that Johnson hasn’t been a consistent finisher across his still early career, though his finishing picks up when he hits a groove and (presumably) gains some sense of comfort and familiarity with the level. As you can see in the table below, which breaks down Johnson’s xG over- and underperformance in all club competitions since 2021-22, his best career xG overperformance (+ 4.67) came with Forest in the Championship:

When he stepped up to the Premier League with Forest in 2022-23 he scored bang-on his expected goals, dropping the significant overperformance from the Championship season, but when he moved to Spurs the following season he dramatically underperformed his xG (- 6.46). In fact, in the Premier League during his first season with Spurs, his -5.3 non-penalty G - xG was the league’s 2nd worst:

In his second season with Spurs, however, Johnson adjusted, presumably both to the level at which Spurs were playing and the system he was playing in, and he leapt from the 579th ranked player in the Premier League in npG - xG to the 96th, turning a massive xG underperformance into a slight overperformance. If we add his finishing in the cups and in Europe last season, his total xG overperformance rose to 3.37.

So far this season, Johnson has 3 goals in all competitions on just 1.53 xG, scoring twice the number of goals as his expected goals. What we’re hopefully seeing is that Johnson’s 2023-24 xG underperformance was a blip—perhaps reflective of an adjustment and maturation period for the player—and he’s now taken a significant step forward in his finishing.

All he does is score goals?

Johnson undeniably scores a good amount of goals for a winger. He netted a career-high 18 goals in all competitions last season, and his 11 Premier League goals put him even with Antoine Semenyo for 18th most in the league, and 5th (t) among wingers. This season he’s off the mark with 2 goals in 5 games, 8th in the league.

But as we’ve seen, Johnson isn’t among the league’s best xG overperformers, at least not yet. Last season he was right around Gabriel Martinelli and a bunch of defenders in that category.

So where do Johnson’s goals come from? What is he doing that makes him a top-5 goalscoring winger in the league?

One thing he’s doing is getting a high number of touches in the attacking penalty area. Last season he was right in between Son and Kudus:

His 1.9 shot-creating actions per 90 last season was quite low—139th in the league—particularly compared to his alternative on the right wing, Dejan Kulusevski, whose 4.82 shot-creating actions per 90 was 7th in the league. This only backs up what we already know about Johnson, which is that he’s not the type of winger to create much for himself with the ball at his feet, but more likely to arrive in the box at the right time or make the right run in behind for a shot.

Another way of seeing Johnson’s value, then, is by measuring his ability to create goalscoring opportunities, even if not through dribbling or guile on the ball, in the ways that Kulusevski or Kudus do. That measurement is a simple one and we’ve already looked at it: expected goals.

In Johnson’s first season in the Premier League, with Forest, he put up .26 xG / 90, tied for 59th in the league with Wilfried Zaha, scoring 8 league goals. In the following season, his first with Spurs—the season in which he was the 2nd worst in the league in xG underperformance—he created .41 xG / 90, moving up to 30th in the league, but scored only 5 goals. Last season, he created .43 xG / 90, 22nd in the league, and scored 11 league goals.

Here we’re seeing steady progression in Johnson’s ability to create goal threat, and with it an increase in goals and perhaps a stabilization of xG overperformance. So far this season, Johnson has played on the left, where he’s less experienced, and shared minutes with Tel and Odobert in that role, with Kudus having established himself as Spurs’ undroppable right winger. Yet Johnson has still put up .34 xG / 90 so far in the league (34th) and is managing to find the back of the net as well.

What I take from all this is that Johnson is not an elite finisher, though he’s improving and showing signs that he very well could become one. The area of his game where he’s really pushing into the upper tier is his ability to get into positions where he can create goal threat, and his consistent ability to generate xG means he’ll likely continue to score lots of goals. It’s worth noting that Johnson had a higher per-90 xG last season than the likes of Noni Madueke, Joao Pedro, Chris Wood, Bukayo Saka, and Ebere Eze, which is to say his relative inability to beat players 1v1 or link play doesn’t mean he can’t find himself in dangerous situations as often as or more often than other very dangerous attacking players; it just means his creative ability lies off the ball rather than on it.

Spurs have kept 4 clean sheets in 5 games to start the season, leading many to the conclusion that Thomas Frank has already made Spurs more defensively solid than we were last season.

Looking at the scoresheets, that conclusion seems obvious. But obvious makes me curious, so let’s take a closer look.

Understanding the Baseline

I still think fans and pundits alike have mostly failed to appreciate why Spurs defended so poorly so often last season and ultimately why we finished 17th in the Premier League.

It’s easy to gesture at Postecoglou’s tactics, but from a defending standpoint, his tactics worked sometimes and didn’t work other times. The question is what was different between the times it worked and the times it didn’t?

It wasn’t the tactics, because the same tactics that won us the Europa League trophy, conceding just 2 goals in the 5 knockout matches, resulted in 20 goals conceded in our last 7 Premier League games over that same period of time, for 6 losses and 1 draw.

By that point in the season Spurs were no longer playing ‘Angeball’ as such, meaning we’d stopped pressing high, dropped our defensive line well back from what it was at the start of the season (from 3rd highest to 8th highest), and were attempting far more long, direct passes, both in Europa competition and in the league. Our average possession also dropped and became more erratic in both competitions (partly due to varying quality and play-style of opposition).

Once you realize that tactics can’t explain the radically different results last season in the Europa League and the Premier League, it makes sense to start looking at other variables.

The one variable that jumps out, screaming and raving, is the availability of our center backs.

First and foremost, I don’t think people have sufficiently processed the fact that last season, Spurs’ first-choice center backs, Romero and van de Ven, played just 15 games together. As Spurs played 60 games in total last season, that means we had our starting center backs for just 25% of the season (I haven’t counted 1 or 2 halves in which they played together before one or both got injured or subbed off due to injury concerns).

Moreover, those games in which Romero and van de Ven played together were clustered heavily around the very beginning and very end of the season, with the games at the very end almost exclusively Europa League games.

The last Premier League game they played together was against Chelsea on 3 April, thereafter playing zero Premier League games for the rest of the season, the last game of which was played on 25 May. Before playing together against Chelsea on 3 April, they’d last played together in the league against Chelsea on 8 December, about 4 months prior. In that game, Romero went off with an injury just 15 minutes in. Prior to that, they’d played together against Palace on 27 October.

All in all, Romero and van de Ven didn’t complete a Premier League match together between 27 October and 3 April, a roughly 6-month period.

I think you have to pause here and let that sink in.

Here’s what that looked like graphically and statistically. The green cells on the left indicate when Romero and van de Ven played together. The yellow cells on the left indicate when one of them played. And the red cells on the left indicate Spurs played someone out of position at center back, e.g. Archie Gray, Ben Davies, or (rarely) Djed Spence. On the right side of the table, I’ve highlighted the match outcomes using green for matches we won, red for matches we lost, and grey for matches we drew. This table (split into two images below, to be read continuously) covers all 60 games of last season:

As you can see in the table, the issue last season wasn’t only that we had our first-choice center backs available for just 15 of 60 games, or 25% of all games we played. It was also (drawing your attention to the red cells on the left side of the table) that we played at least 1 non-center back in one of the two center back positions in 36 of 60 games, or 60% of all games we played.

Let me make a new paragraph here in the hope that this will sink in: For 36 of 60 games last season—60% of our games—we played someone out of position at center back.

So it’s not like when Romero and van de Ven weren’t available we were just plugging in Dragusin (who suffered a major knee injury partway through the season) and Danso (who joined after the January window). We were not even playing with two recognized center backs in a back 4 for 60% of our games.

The results were both shocking and predictable.

‘Shocking’ in the sense of woeful, but I think both explicable and predictable. In the 15 games that Romero and van de Ven played together last season, Spurs conceded just 12 goals, an average of .8 goals per game. Maintaining that defensive record from 25% of our games across 38 games would mean conceding 30 goals, which would’ve been the best defense in the Premier League (Arsenal were best in the Premier League last season with 34 goals conceded).

Over those 15 games, Spurs’ record was 9 wins, 4 losses, and 2 draws. That record extrapolated over a 38-game season would be 74 points, which would’ve put Spurs in 2nd place last season, tied with Arsenal, not 17th. Obviously you can’t put much faith in such a crude extrapolation of just 15 games (not systematically accounting for level of opposition, etc.), but the point here is the level is dramatically higher with our starting CBs in action.

By considerable contrast, when neither Romero nor van de Ven were in the team last season—that was a total of 31 games, more than twice the number of games in which they featured together—Spurs conceded 51 goals, or 1.65 goals per game. Extrapolated over 38 games, that’s 63 goals conceded, 2 fewer than Spurs’ actual 65 goals conceded in the Premier League last season, good for the 16th best (or 5th worst) defensive record in the league, exactly where we ended up by season’s end. Spurs’ record over those 31 games without both Romero and van de Ven was 11 wins, 15 losses, and 5 draws, good for about 48 points extrapolated over 38 games, which would put Spurs in 13th place last season, tied with Everton.

In case you’re curious, over the 36 games in which Spurs played with a non-center back at center back, we conceded 59 goals, an average of 1.63 goals per game, with a record of 14 wins, 17 losses, and 5 draws. That record over 38 games would be good for 50 points or 13th in last season’s table.

The expected goals against figures are well in line with the actual defensive records above. In the 15 games that Romero and van de Ven played together, Spurs conceded an average xGA of just .86, right around the .8 goals per game they actually allowed. Without either of them, the average xGA jumped up to a dreadful 1.97 (and a roughly as dreadful 1.85 when playing with only one recognized center back in the two center back positions).

What you should take away from this is that Spurs’ defensive record so far is NOT well explained by who the manager is, NOT well explained by what we’re doing tactically, and VERY well explained by whether Cuti Romero and Micky van de Ven are on the pitch together.

It would be a lot of work (and more time than I have to spare) to do a similar analysis to what I’ve done for Spurs here for every other Premier League team, particularly those who’ve had similar injuries to their center backs. Then again, I don’t think there’s any other team who’ve had such an awful injury record among their first-choice center backs, nor one that left the squad so short as to be forced to play non-center backs at center back for 60% of their games. If you find anything like that, though, do let me know!

Defensive Record So Far Under Frank

If we look at the 5 games on either side of the black bar below (which marks the end of last season under Postecoglou and the beginning of this season under Frank), we can already see the kinds of things we should keep an eye on as Frank’s project at Spurs progresses:

One important thing that’s clear enough so far is that Frank has had both Romero and van de Ven healthy and playing together virtually every minute of every match this season. And when Frank chose to take one of them off the pitch (Danso’s 10 minutes for van de Ven at the end of the West Ham game), it’s not Davies or Gray entering the fray in center back cosplay, but a recognized top-flight CB (and quite a good one at that).

Thus far, Spurs under Frank have conceded just 1 goal in 5 games, or .2 goals per game, with an average xGA of 1.03. That’s a touch higher than the average xGA last season under Postecoglou with Romero and van de Ven playing together, .86. But it would be foolish to quibble over such a small difference over such a minuscule sample size on both ends of the chart, given how few games Romero and van de Ven actually played for Postecoglou.

What I think so far is Spurs aren’t obviously better or worse defensively this season due to manager or tactics. We’re just performing right around the level we do with Romero and van de Ven in the team.

To know anything more—to say Spurs are suddenly fixed defensively by tactics—would require both a lot more data and something to control for, e.g. a record of performance without our first-choice CBs, or a run of games when we’re using non-CBs at CB (of course, I hope we don’t have to see much, if any, of the latter, though healthy CB rotation with a deeper squad in that position is something I’d very much like to see this season).

This, by the way, matches the eye test for me. Keep in mind Frank plays a pretty high line, and we’ve already seen opponents play long over the top on us, or break fast on the counter, and get into some dangerous positions. Our Premier League xGA and xGD thus far are mid-table even as we’ve kept 4 clean sheets in 5 games, and my sense eyes-wise of what’s making the difference (as against City and Bournemouth, both of whom should’ve scored more than they did) is stellar defensive play at the individual level of Romero’s all-around excellence, van de Ven’s physicality and recovery pace, and Spence’s emergent 1v1 defending prowess (to say nothing of Porro’s unrelenting peskiness), and Palhinha’s self-assured aggression. We now have a fair number of really good defenders in the team, so we should expect good defending, even when we’re open at times.

‘Has Daniel Levy been good for Spurs?’ is one of those arguments I just can’t get into and have never been particularly interested in. The rough consensus seems to be that he was good for Spurs on the business side but not so good on the football side. The Tottenham Hotspur Supporters Trust statement on Levy’s resignation captures as much in predictable fashion.

However, I think this binary ‘good for finances, bad for football’ way of looking at Levy’s tenure is a misleading way of telling his story at Spurs.

So I’m going to approach Levy’s tenure at the helm of Tottenham Hotspur from a strictly numerical standpoint, which I think is quite revealing. Hat tip is due to iain.coys.social over at BlueSky for suggesting this idea. I had a vague notion of what this investigation would look like, but wasn’t fully prepared for what I’d find.

Boring Financial Stuff

Let’s get the obvious out of the way before we get to the good stuff. I think it’s relatively uncontroversial to say that Levy’s chairmanship was a resounding financial success for the club. I don’t think Spurs offer annual financial reports before 2006, which makes a pre- and post-Levy comparison impossible with the data I could find. But the general direction not only of the chart below, but of the commercialization of the Premier League during Levy’s tenure (2001-02 to 2024-25) should be enough for us to accept that Levy brought the club to a new level financially:

The only other thing I’ll say about this is we all know about the new stadium and the go-karts and the Skywalk (TM), which our club and fans have been ceaselessly bantered over, but the fact is, under the various financial stability regulations now in place, Spurs will almost certainly have the last laugh. Levy has made Spurs consistently one of the 10 richest clubs in the world (and growing), while the likes of Aston Villa and Newcastle, for all of their owners’ riches, remain stuck in PSR jail.

Yeah Yeah Whatever, What About the Football?

Here are a few data points worth mentioning:

In the 24 seasons prior to Levy’s assumption of the club’s day-to-day operations in 2001-02, Spurs’ average league position was 8.8. In the 24 seasons of Levy’s tenure, it improved significantly to 6.6. (N.B. The English first division had 22 teams before the 1995-96 introduction of the 20-team format. Spurs average league position in the 22-team format was 8.2, then 10.8 under the 20-team format until Levy took over.)

In the 24 seasons prior to Levy, Spurs won 5 major trophies: 3 FA Cups, 1 EFL Cup, and 1 UEFA Cup. During Levy’s tenure, Spurs won 2 major trophies: 1 EFL Cup and 1 Europa League title.

In the 24 seasons prior to Levy, Spurs qualified for European competition 6 times. This figure is partially deflated by the fact that English sides were banned from European competition from 1985-1990, during which Spurs’ league finishes would’ve meant qualifying for the UEFA Cup 4 more times. Nevertheless, during Levy’s tenure, Spurs qualified for Europe a remarkable 19 times. That’s 3 UEFAs, 7 Champions Leagues, 8 Europa Leagues, and 1 Europa Conference League appearance.

Looking at the raw numbers of this track record, it’s very difficult to say Levy’s leadership at Spurs hasn’t been successful on the pitch as well as off it. The rise to regular European competition under Levy is I think especially noteworthy, bringing Spurs into the upper echelon of the Premier League in a way that no other Premier League club outside of the ‘big 5’ has done in modern football.

Yes, Spurs managed 3 fewer trophies under Levy than in the previous 24-year period, but as I’ve broken down, knockout competitions are extremely low-odds. The difference between 5 trophies in 24 years and 2 trophies in 24 years is not causally interesting, in my view. In other words, you could end up on either side of that situation just by luck or variance.

For example, in the 24 years before Levy, Spurs made it into 2 additional cup finals that they didn’t win (1 FA and 1 EFL), a total of 7 opportunities, 5 of which taken. During Levy’s tenure, Spurs made it to 5 cup finals that they didn’t win, including a Champions League final, and excluding a 2nd-place league finish in 2016-17.

You could argue, therefore, that Levy’s leadership gave Spurs the same number of opportunities to win trophies than in the previous 24 years—more if you count the 2nd-place league finish—and losing 5 cup finals over that time was perhaps an anomaly.

A fair counterpoint might be that comparing 1977-78 Spurs to post-2000 Spurs is apples and oranges, given that the club became much richer during the latter period, hence expectations should be higher for Levy than pre-Levy footballing outcomes. Of course, if you agree with this point, then it loops you right back to the causal question: How did Spurs become so much richer and more competitive during Levy’s leadership? The question practically answers itself.One Unforgivable Thing

For all the statistical positives I’ve pointed out above, there’s a statistical negative that I’d be remiss to leave out.

Spurs had 15 different managerial tenures from 1976 to 2001. Thereafter, under Levy’s chairmanship, Spurs had 19 different managerial tenures in 24 years, 8 of which were packed into the last 6 years, 7 of which packed into the last 4 years, averaging more than one manager per year since 2019.

If there’s one thing responsible for the squad planning, recruitment, and tactical continuity and cohesion issues Spurs have suffered in recent memory, it’s Levy’s penchant for scapegoating the manager when things get rough.

I might even go so far as to say the difference between the level that Levy has brought Spurs to over his 24-year tenure and the next and final level up—competing at the very top of the table for league titles—is the inability of Spurs’ executive leadership to develop and stick to a long-term project, like Pep’s City, Klopp’s Liverpool, or even Arteta’s Arsenal. We tend to fixate on spending and transfers, but I think continuity is equally important for taking the next and final step to the top.

I don’t believe in transfer window ‘ratings,’ but I do think it’s a valuable exercise to check in on where we ended up in light of where I thought we’d need strengthening back in July. Here’s a quick summary of what I think have been our most pressing personnel needs:

• left-sided defender

• wider range of midfield passing profiles

And here are a few more needs that have arisen, in my view, since I wrote my last transfer-window check-in post in July:

• depth in the creative / attacking midfield role (since we’ve now lost both Maddison and Deki long-term to injury)

• goals and chance creation from wide forward roles (since we’ve lost Son to LAFC)

• depth at striker (since Solanke’s ankle injury appears to be more persistent than initially thought)

Defensive Needs

Although we’ve brought in two young center backs in Vuskokvic and Takai, we’ve loaned out the former while the latter has been injured. Dragusin remains out with a long-term knee injury, leaving Spurs looking worryingly light in defensive personnel.

Given Spence’s emergence and versatility, the fullback trio of Spence, Porro, and Udogie is a solid base depth-wise and one of the most exciting groups of fullbacks in the Premier League. Spurs could weather an injury to any one of these three players and still have outstanding fullback options, but what concerns me is the prospect of any two of our fullbacks getting injured. I’m reasonably comfortable with Gray deputizing as a right-sided fullback and Davies as a left-sided fullback for a short period, but not more.

Similarly, the center back trio of Romero, van de Ven, and Danso inspires confidence and, when fit, is also one of the league’s best core CB groups. But should any two of those three suffer simultaneous injury, we’re currently without any further CB cover. Davies might be OK as a left-sided CB in a back 3, though, as I wrote about extensively last season, I don’t think he’s either a true CB or up to the level Spurs require in a hyper-competitive Premier League and Champions League.

Last season, Spurs were among the best 5 defenses in the Premier League with their starting back line, and among the worst 5 with Gray, Davies, and Dragusin slotted in for long stretches of the season. I shouldn’t have to keep making this point, but losing our starting defenders for all but roughly 10 out of 60 games last season and replacing our CBs with non-CBs playing out of position is the reason we finished 17th last season.

It would extremely foolish—particularly as we’re in the Champions League now—to leave open the possibility of that happening again. Likewise, I would not bet on the manager change having any impact on how many injuries we sustain, especially as so far Spurs have already played a high line under Frank that has us sprinting back to chase balls played over the top.

In terms of defensive cover, then, I can understand why we haven’t brought in a left-sided defender, perhaps someone who could play as a CB and FB in a pinch, given that we have Takai and Dragusin on the roster, though injured. Should those two recover well and before we sustain any further injuries to our defensive players, we could be OK. But this is one area of the squad that still leaves me feeling nervous.

Midfield Needs

With Maddison likely out for most if not all of this season, and Deki also out long-term through autumn, Spurs were pretty desperate for an attacking midfielder who could create multiple ways, through passing as well as ball-carrying, 1v1 dribbling, and goal threat. Kudus can do some of that and could certainly function as a 10 as needed, but I think is strongest on the wing. For this reason, signing Simons was a major victory. It’s always tough to know how a player from Europe will adjust to the particular brand of pace and physicality in the Premier League, and I’d be lying if I said I didn’t have some reservations about Simons’ tendency to agitate for a move rather quickly wherever he goes. But with his ability to dribble, pass, and finish, and his inventiveness and daring, I think he’ll be a phenomenal player for Spurs.

With the addition of Simons, then, Spurs are suddenly looking good in the advanced midfield role. Once Deki and eventually Maddison come back, Simons can also play on the left wing, so the possibilities are pretty tantalizing.

But Spurs still lack a bit of passing in the midfield squad, particularly long-range passing from deeper positions. The ball-carrying and dribbling profiles we’ve been stocking up on will help a bit with that issue—there’s more than one way to progress the ball, but especially if we part ways with Bissouma (who could still go to Galatasaray, with still more time left in the Turkish transfer window), our midfield looks a little flat in terms of ball progression. The upside is that Bergvall may just continue to improve and acclimate, and his ball-progression abilities are there to see.

Many have said that what Spurs really needed in the midfield was strong defensive cover. I didn’t see that as quite as high a priority as passing and ball-progression. The addition of Palhinha is certainly good for the former, though not so good for the latter (as we saw last weekend against Bournemouth). That said, Palhinha is among the best at the particular thing he does, and if doing that thing (ball-winning, breaking up play, aerial duels, being a menace) frees up Simons, Maddison, Deki, etc. to advance, that’s a good thing. I’m all for a well-rounded midfield squad with lots of options.

Forward Needs

As mentioned, the addition of Kudus is crucial and adds ball-retention, hold-up play, 1v1 dribbling, chance creation and—I hope—goal threat on an order we didn’t have before on the right side. And with Son departing for LA, Spurs will certainly need to recoup his 15-20 goal contributions (minimum) per season.

Accordingly, I hadn’t identified the left wing as an area of particular need when Son was still in the squad, but it’s a bit more pressing now. That’s not to say Odobert and Tel can’t step up, or that Johnson can’t be valuable on the left—indeed, this is a great opportunity for all of them—but if we don’t have consistent goal contributions and threat from the left wing by January, we should be looking for it then.

That leaves only the need for cover in the striker role, given that Richarlison—our only healthy out-and-out striker so far this season—has a rough track record of injuries, and that Solanke has been out injured all season so far. We know that Tel can play through the middle, though we lose crucial stickiness and hold-up play without Solanke and to a lesser extent Richarlison in the team. For this reason, and others, I’m very excited that Spurs have brought in Kolo Muani, who maybe suffers from a case of inflated expectations from earlier in his career, but who’s now prime age and can really do it all. He’s strong, physical, fast, technically gifted, great 1v1, and an excellent finisher (despite that one, er, high-profile miss).

Last season for Juventus, Kolo Muani played in 16 games, started 13, and scored 8 goals (with 1 assist). We could wax nostalgic about his world-beating performances at Eintracht Frankfurt in 2022-23 (26 goals in 50 appearances), but it’s worth pointing out that he’s scored 10 in 22 appearances for Juve most recently, about the same ratio as the season that made him PSG’s record signing at the time.

Conclusions

In terms of our attacking and midfield needs, this has been a very strong window. Spurs have continued to build on a good foundation in these two areas of the pitch, only to be throttled by serious injuries to two of our best players in Maddison and Deki and the loss of our talisman, Son, to LAFC. Still, bringing in the likes of Kudus and Simons is not only, as they say, excellent business, but will go a long way to giving Spurs the creative dynamism they need to stay competitive.

That said, the relative lack of depth in the back line—put more precisely, the level of drop-off after our best 3 CBs and FBs—risks a repeat of last season should any two CBs or FBs suffer simultaneous, long-term injury. Maybe we’ll be lucky this year—we certainly deserve it after last year—and it won’t be a problem. But for me that’s where we’ve fallen short in the summer window.

There’s a kind of ‘football centrism’ in Premier League coverage and fandom, akin to prominent voices in US and UK politics for whom certain issues, phrases, and framing are 'distractions,’ even when they name something real or true. Some of the core tenets of football centrism are that the old explanations are always the best ones; that you shouldn’t look too closely at something if it’s not obviously broken; and that you shouldn’t sweat the small stuff.

I don’t like football centrism, because—like its political counterpart—it often and implicitly disfavors new ways of understanding something, or less popular angles on a question or an issue, out of fear of being perceived too radical, too ideological, or too ‘biased.’

To give a concrete example, a canonical ‘football centrist’ view is that ‘pragmatism’ equals a deeper-sitting, primarily defensive tactical approach to the game. Challenges to that view then become ‘ideological’ or ‘naive,’ even if and when such challenges can be demonstrated true.

Accordingly, especially as Thomas Frank’s tenure gets underway at Spurs—with much success so far!—you’re going to watch, hear, and read a lot of football centrist takes on Spurs. It’s baked into the scenario in which the previous manager, Ange Postecoglou, was painted as an antithesis to the ways of the Premier League and ‘proper’ English football (sometimes fairly, many times unfairly), while the new manager, Thomas Frank, has long been touted as the league’s premier Pragmatist.

Football centrists are going to tell you that, for example, Spurs have started the 2025-26 season with two victories and back-to-back clean sheets because Frank’s defensive setup is peak pragmatism and thus the polar opposite of Postecoglou’s ‘naive high line.’ ‘He’s already sured up the leaky defense,’ etc. What you won’t hear from football centrists is the fact that last season, Postecoglou’s Spurs and Frank’s Brentford played an almost identical average defensive line height. Spurs played the 8th highest line in the Premier League (45.89) and Brentford played the 9th highest (45.34), a difference in average defensive line height of about a half of a meter:

If you look closely at Spurs’ impressive win against City last weekend—the numbers as well as the performance itself—a couple of things stand out.

One is that Spurs pressed City high and aggressively, pushing our back line forward, especially with Porro and Spence ready to spring on the counter. On a few occasions, City fashioned decent chances by playing over our press and into our relatively high defensive line, winning the ball and driving on from there. The football centrist view is that Frank has immediately made Spurs more defensively organized and eliminated the vulnerability of the high line, but I don’t think either is quite the case.

City generated the better xG against us and missed both of their big chances:

This doesn’t mean we weren’t good defensively, however. I think Spurs defending very well throughout the match, and there’s only so much you can do against the kind of lineup that City throw at you.

Rather, I think it’s both too early and to football-centrist to uncritically accept the idea that a new manager and new tactics have already made Spurs more ‘defensively solid’ (another big football-centrist phrase). A better explanation, at this point, is still that Spurs are more ‘defensively solid’ when we have our best defenders in the team. I’m still hearing commentary that’s straightforwardly assuming Postecoglou and Frank have played in the Premier League with the same group of defenders, which is straightforwardly untrue. You can’t compare the effect of different tactical approaches (between Postecoglou and Frank, in this case) without holding the personnel variable constant. That is, comparing manager / tactics A with Dragusin and Gray to manager / tactics B with Romero and van de Ven, for example, is an inherently faulty comparison.

That said, the incorporation of Palhinha playing in a double pivot with Bentancur is certainly a tactical adjustment that Frank has added that makes us more solid; but it’s also, like all tactical adjustments, inextricable from the personnel that Frank has at his disposal now.

Likewise, Frank’s Spurs have already gotten lots of credit for playing long from the back, especially for Vicario’s long passes, one of which was the impetus for Spurs’ first goal against City. As I’ve shown, Postecoglou’s Spurs also shifted to more long passing on the back end of last season, sometimes to good effect but mostly not, both in the league and in Europe.

There’s a tendency to want to explain all of this in purely tactical terms, but again it matters who’s on the pitch and who’s not. When Spurs played long last season, we often lacked the personnel up front to bring the ball down, hold it up, keep possession, and reliably win second balls. This meant that the ball regularly came back on us, causing havoc for our makeshift back line and midfield. Now, however, we have healthy center forwards, the unplayable Kudus (whom every ball seems to stick to), and much more ball-winning in the midfield with Palhinha and Sarr. Which is to say what we’re seeing now is not pure tactical change or ‘pragmatism’ in football centrist terms, but better alignment of managerial aims and the kinds of players needed to implement them.

There is, however, one striking tactical change we’re seeing so far with Frank’s Spurs that’s, in my view, a marked improvement upon Ange’s Spurs: the structure of our pressing.

Whereas a weakness of Postecoglou’s press was a tendency to overreact and overcommit in the press, Frank’s pressing scheme thus far looks simultaneously more conservative in the midfield and in wide areas and more intensive in the front line. Both Postecoglou’s and Frank’s versions of Spurs appear to rely on pressing to generate chances from high turnovers—like we saw with Sarr’s tackle that led to the second goal against City on the weekend—a type of ‘creativity’ that easily flies under the radar.

Looking ahead, it will be fun (and I think useful) to keep an eye on all of these developments: how high of a line we play; how vulnerable we are to long passes and counters; how well we hold the ball up on long outlets of our own; and for how long we can maintain enough pressing intensity, as matches come thick and fast, to generate chances from high turnovers in the absence of strong creative midfielder. Football centrists might accuse me of not letting go of some of my least favorite Ange myths, or not giving Frank due credit, or nitpicking at something that, so far, is working really well.

But what matters to me is getting things right, not following the dubious ‘common sense’ of football centrism. Even if it’s the little things. Because down the line—when we have more matches, when injuries start piling up, when it’s harder to prepare custom tactical plans for individual opponents, i.e. when the core principles Frank instills have to function on auto-mode—I’d rather be able to explain what’s going on at a high level of detail and with ample context than to fall back on truisms.