Welcome back. Today’s topic is the Super Bowl favorite Los Angeles Rams. We’ll be discussing a particular piece of information shared in this clip:

Your ears may have perked up around the 1:42 mark. Tony Kornheiser states that in the last 30 years, 20 teams have been Super Bowl favorites with +500 odds or better (like the Rams this year). None of those 20 teams went on to actually win the Super Bowl. Not a single one.

Here’s the big question: Is this a coincidence, or does being a heavy Super Bowl favorite somehow have an adverse effect on a team’s ability to actually win the whole thing?

To answer this, it’s time to do some math. Let’s begin.

First of all, from what I’m seeing in the publicly available data at covers.com, there were actually 22 teams in the last 30 years with a league-leading +500 odds or better to win the Super Bowl. This discrepancy is because covers.com lists not one, not two, but three 2005 teams with the same +500 odds atop the NFL prior to Week 1: the Colts, the Patriots, and the Eagles. Not a big deal. I’m good with calling it 20, but for my analysis, I’m going to use 22. And yes, the factoid still holds; all 22 teams failed to win the Super Bowl.

Let’s crunch some numbers. +500 odds equates to a 1 in 6 chance of success. So, to flip that around, it implies a 5 in 6 chance of failure. In other words, an 83.3% chance of not winning the Super Bowl.

(5/6)^22 = 1.8%

That’s the probability of 22 Super Bowl favorites with +500 odds all coming up short due to random chance alone. Okay, that seems pretty surprising.

Here are two caveats to consider:

1. Some of these teams actually had better than +500 odds. The 2017 Patriots, for example, were +275. So, the true probability is actually less than 1.8%. If we go ahead and account for all the teams with better than +500 odds, the final figure comes out to a 0.5% chance.

2. Sportsbooks odds always sum to more than 100% across the market because of the built-in profit margin. The bookies guarantee themselves a profit so long as people are placing bets on all the possible outcomes. Because of this, the true Super Bowl odds for each of our 22 teams is actually a little worse than the odds themselves directly imply. This means that our calculated 0.5% probability is now an underestimate. The Super Bowl futures markets have a hefty overround, commonly cited in the 20-40% range. If we adjust by, say, 30% on average, we get a final probability that is back up to about 1.9% again.

The two caveats essentially cancel each other out when it comes to adjusting our calculations. However, these are still nuances to be mindful of. Accounting for them makes our conclusions stronger.

Most importantly, and I cannot stress this part enough, the use of +500 as the threshold for a heavy favorite and 30 years as the timeframe in question is completely arbitrary. In statistics, this is called p-value fishing.

If we adjust the odds slightly, we get a headline that reads:

In the last 30 years, 30 teams have been Super Bowl favorites with +600 odds or better (like the Rams this year). 4 of those teams went on to actually win the Super Bowl.

Or, if we adjust the timeframe slightly, we get a headline that reads:

In the last 33 years, 25 teams have been Super Bowl favorites with +500 odds or better (like the Rams this year). 2 of those teams went on to actually win the Super Bowl.

Hm, not as eye-popping.

In fact, running the numbers on these two hypothetical headlines generates probabilities of 52.3% and 16.8%, respectively. That is, there’s a decent chance in a single universe that only 2-4 teams are successful in each of these scenarios. It’s not that unusual.

Here’s the cool part: We can now adjust our probability for “0 of 22 teams with +500 odds in the last 30 years” based on the multiple alternative potential cutoffs that could have been used instead, using what is called a Bonferroni correction.

What the Bonferroni correction tells us is that there is not just a 1.9% chance, but actually a greater than 5% chance of having a run of zero successful Super Bowl favorites in the last X years when they had Y odds or better. This may not seem like a high probability, but in the world of statistics, it’s not low enough to claim with confidence that there is an adverse effect on a team’s ability to win a Super Bowl when they enter the season as heavy favorites.

Still not convinced? Try this thought experiment. Pick a random number from 1 to 20. Got it? Good.

Was it 17?

Okay, maybe it wasn’t. Or maybe it was! Either way, let’s pretend I guessed your number correctly. There’s only a 5% chance of that happening. Is that enough to convince you that I can read minds? Or rather, would you chalk it up to coincidence and want to see me do it again with a larger range of numbers (like 1 to 1000) in order to prove my powers?

To bring it full circle: The fact that 20 teams in the last 30 years were Super Bowl favorites with +500 or better odds and none of them finished as NFL champions is somewhat rare and surprising, but it is not unlikely enough to suggest that there is some kind of “curse” for the heavy Super Bowl favorites.

In other words, if you were hoping that this little factoid meant that the Los Angeles Rams were doomed this year, I’m sorry to disappoint. If this Rams superteam is going to be stopped, it will not be due to curses, fate, or statistical regression. It will have to be done by a team that rises to the moment and snatches victory for the underdogs, as has been done many times before.

EPA Per Play

As it turns out, efficient football teams are good football teams. 74% of Super Bowl teams ranked top 10 in offensive EPA per play, and 55% ranked top 10 in defensive EPA per play. 68% of teams were above average on both offense and defense, and 73% of teams with a below average unit were top 3 on the other side of the ball.

Why are these percentages important? Because they tell us that EPA per play is a great indicator of Super Bowl-caliber football teams. To prove this more robustly, we can employ some statistical tests. (Quick refresher: P-values range from 0 to 1. The smaller the p-value, the stronger the evidence of a significant difference. Generally, anything less than 0.05 is considered strong enough to accept as a “real” difference; that is, unlikely to occur by randomness alone.)

The combined rank of a playoff team’s offensive and defensive units is 11.1 on average; for a Super Bowl team, it’s 8.7 (p = 0.0006). A playoff team’s best unit averages a rank of 6.4; a Super Bowl team’s best unit is 4.0 (p = 0.0001). A playoff team’s worst unit averages a rank of 15.9; for a Super Bowl team it’s 13.5 (p = 0.0364). The numbers prove what already intuitively makes sense: Super Bowl teams are more efficient on a play-by-play basis.

Historical Super Bowl Teams

The graph below shows the offensive and defensive rank of playoff teams, with Super Bowl teams circled in red. As the data from the prior section would suggest, the Super Bowl teams cluster in the top right corner.

Now check this out. I’ve added two dashed lines to illustrate the pattern of how Super Bowl offensive units and defensive units relate to one another. The lighter dashed line represents the minimum threshold of a Super Bowl team: all but one team surpass this boundary. The darker dashed line represents the minimum threshold of a “good” Super Bowl team: just over 50% of Super Bowl teams surpass this boundary.

We can take the point-slope equations of these dashed lines to generate formulas that answer questions like, “If the offense ranks 8th, how good does the defense need to be?”

Minimum Criteria for a Super Bowl team

[Defensive Rank] = 44 + (-7/4 x [Offensive Rank])

[Offensive Rank] = -4/7 x ([Defensive Rank] - 44)

Criteria for a Good Super Bowl Team

[Defensive Rank] = 24 + (-7/4 x [Offensive Rank])

[Offensive Rank] = -4/7 x ([Defensive Rank] - 24)

For example, consider a team that ranks 7th in defensive EPA per play. We can plug that information into these equations to find out that to the offense needs to be ranked at least 21st to be in the Super Bowl conversation, or at least 10th to be considered a good Super Bowl team. If their offensive rank fell somewhere in between 10th and 21st, we would consider that team to have a shot at a Super Bowl appearance, but as one of the weaker teams in the modern era. If their offensive rank was better than 10th, we would consider that team to be among the favorites to win their conference.

Consider the 2025 Broncos and Rams. The Broncos were 8th in defense and 15th in offense. They were clearly good enough to make the Super Bowl, but they would certainly have been considered one of the weaker teams in recent memory to win a conference championship if they had done so. The Rams had a similar defensive rank. They were 7th in defense but 2nd in offense, and they were considered an obvious Super Bowl contender and one of the more dominant teams in recent memory to compete in a conference championship game.

The equations above are simply putting an objective process to that line of thinking.

Super Bowl DNA

Now that we have a framework for thinking about Super Bowl-caliber teams, we can ask some deeper questions. Does offense or defense matter more? Are teams with a larger gap between their offensive and defensive ranks less likely to win their conference?

To answer these questions, I ran three different importance methods (standardized logistic coefficients, permutation importance for both logistic and random forest models, and random forest impurity importance) on the dataset of all playoff teams since 2002. They all agreed on the following takeaways:

1. The “best unit rank” (the rank of a team’s stronger unit, whichever of offense/defense is better) was the most important feature of Super Bowl teams.

2. The “overall rank” (the average of a team’s offensive and defensive ranks) was more important than either offense or defense alone.

3. Offensive rank alone was slightly more important than defensive rank alone.

4. The “worst unit rank” and “rank gap” (the difference between a team’s offensive rank and defensive rank) hardly mattered at all; they had near-zero importance in the model.

So what does this tell us? For one, the most important element of a Super Bowl team is having one elite unit, on either side of the ball. A team with the #1 defense and the 11th best offense is better than a team with the 6th best offense and 6th best defense, despite both teams having the same average rank. There is no advantage to being the more “balanced” of the two teams; but rather the opposite. A team with the best unit on the field is more likely to win a championship game, even if they also have the worst unit on the field. Look back to our plot of playoff teams. Notice how the Super Bowl teams actually cluster around the top and right borders of the graph, with relatively fewer circled teams located in the space in between.

It also tells us that, while offense is slightly more important than defense, the average of the two units carries more weight than either unit alone. That is, a team with the 16th best offense and 10th best defense is more likely to make the Super Bowl than a team with the 10th best offense and 20th best defense. Yes, the latter team has a top 10 offense, but the fact that its average rank is 15th instead of 13th carries more water than the difference in the two teams’ offenses.

Takeaways

Does defense win championships? Well, sometimes. But not because defense is better than offense come December. It’s because the best unit on the field is the best advantage a team can have in the playoffs, and at least half of the time, that unit is the defense. When people use this antiquated cliché, it’s likely because they are watching the games with the expectation that better offense will win, not that the best unit will win. As such, they are specifically remembering all the times that the team with the worse offense but the better defense “upset” their expectations, and failing to recall as easily all the times that the better offense won out (because why would you have strong memories of the times your expectations were confirmed?)

This is called survivorship bias. Many of you will be familiar with this concept. For those that are not, it’s a logical error where you focus only on successful events that made it past a certain selection process, while ignoring the ones that failed.

The fact is, for every 2015 Broncos team, there is a 2006 Colts team. For every 2002 Buccaneers team, there is a 2017 Patriots team. For every 2025 Seahawks team, there is a 2020 Chiefs team.

Elite units win championships, whether those units play offense or defense.

Super Bowl Contenders Before Week 1

Here is the list of top 10 offenses in EPA per play last year:

1. Patriots

2. Rams

3. Bills

4. Packers

5. Cowboys

6. 49ers

7. Bears

8. Lions

9. Colts

10. Seahawks

And for defense:

1. Seahawks

2. Texans

3. Vikings

4. Browns

5. Chargers

6. Broncos

7. Rams

8. Jaguars

9. Eagles

10. Saints

Only the Seahawks and the Rams appear in both lists. As such, it only makes sense to rank them as the two most likely teams to make it to the Super Bowl this year.

Tier 1 - Great on both sides of the ball with one elite unit

Seahawks, Rams

Of the remaining teams, I ask myself two questions:

1. Which of these teams have a path to repeating as a top 10 unit, and have a sufficient floor on the other side of the ball?

2. For the teams without a real floor on the other side of the ball, does their better unit have a path to becoming elite (top 5)?

Here’s how I would categorize the remaining teams…

Tier 2a - Could have an elite offense with a decent defense

Bills, Ravens, Patriots, Lions

The Bills and Ravens have proven themselves capable of top 5 offensive production behind MVP-level quarterback play. Using the aforementioned equation of a good Super Bowl team, their defenses would need only rank 15th to contend. The Bills were 13th last year, and acquired Jim Leonhard, former defensive backs coach of the Broncos as their new defensive coordinator. The Ravens are getting back a healthy Lamar Jackson and added highly touted defensive mind Jesse Minter as head coach, who had the Chargers ranking 6th in defense last year. The Patriots added A.J. Brown to a team that had the #1 offense and MVP runner-up last year, while the Lions are loaded with talent at offensive skill positions.

Tier 2b - Could have an elite defense with a decent offense

Eagles, Texans, Vikings, Broncos

The Eagles, Texans, Vikings, and Broncos all had elite defenses last year and could easily repeat this year. Each has gone through their own set of changes on offense in the off-season that give reason to believe they can improve to an above average unit on that side of the ball this year as well.

Tier 3 - Could be top 10 in average rank of offense and defense

Jaguars, Packers, Chargers, Chiefs, Colts

The Jaguars, Packers, and Chargers are great examples of teams that seem to have decent floors on defense, but really need the offense to meet expectations and mature from great to elite to make a Super Bowl run in 2026. The Chiefs and Colts could easily be top 10 on both sides of the ball, but the path to either having an elite unit is difficult to see.

Tier 4 - Could be top 10 on one side of the ball and bottom 10 on the other

Bengals, Cowboys, 49ers, Bears

It’s easy to see Joe Burrow throwing to Ja’Marr Chase and Tee Higgins creating a great offense, and likewise for Dak Prescott throwing to CeeDee Lamb and George Pickens. However, their defenses have been so bad as of late that those offenses would not just have to be elite in 2026, they would have to be historic. If the defenses improve substantially, these teams have a shot to jump up to Tier 2. For the 49ers, it feels like their ceiling is a Tier 3 team, especially with all their injuries. History tells us Kyle Shanahan will find a way to produce a top 10 offense, but they simply lack the firepower to return to their glory days of elite offensive production. The defense has been a problem for several years now; they were 26th in 2024 and 24th in 2025, and the defensive roster is essentially the same heading into 2026. The Bears have the pieces (Odunze, Burden, Loveland, Swift, Williams) to produce a great offense, and the coach (Ben Johnson) to pull it off, but the defensive roster is so uninspiring that it’s hard to see them as any different than the rest of the teams in this tier.

Closing Thoughts

Super Bowl teams come in all shapes and sizes. Will this year’s victor be an offensive juggernaut with a decent defense? Or will the team hoisting the Lombardi trophy be a suffocating defense with a frisky offense? Elite units make all the difference, and nearly a dozen teams are in position to follow that exact path to a conference title this year.

As the season wears on, keep this fact in mind. It’s easy to fall for the well-rounded team with a good offense AND a good defense. However, history tells us that most Super Bowl teams are not just great, but ELITE on one side of the ball. Watch out for those teams, because their worse unit can always catch a hot streak at the right time in the playoffs and surprise everybody. I mean, just look at the 2025 Super Bowl Champion Seattle Seahawks.

Trust the process over the results.

If you don’t play fantasy football, no worries. I’m sure you already have your own preconceived notions about which wide receivers are underrated, which ones are overrated, and which veterans joining new teams are actually just washed. In that case, I’m about to provide you with some analytical arguments to use to your advantage in lighthearted-but-heated debates with friends and family.

Background

The goal here is to generate a ranking of wide receivers that is as close as possible to the final standings at the end of the season, as measured by total production. The easiest way to define “production” is total fantasy points. The good news is that there is a company called Fantasy Points Data which already looked into which metrics from the prior season best predict total fantasy points in the next season. The bad news is that their primary finding was that the best predictor is, well, total fantasy points of the prior season. No, it’s not yards per route run, or yards after the catch, or even first-read target share. Womp womp. So, I thought, why not improve on this idea that last year’s fantasy points predict next year’s fantasy points? Why just look at the total? What other ways can we summarize fantasy point data?

New Methodology

The first thing I did was gather the fantasy point scored each week by every wide receiver from 2016 to 2025. Then, for each player, I could calculate not only their total fantasy points for the season, but also their average points per game, their median points per game, and their standard deviation in points. (I should note here that my method of calculation was half-PPR, where each wide receiver gets a half a point for each reception in addition to their standard scoring for yards and touchdowns.)

To make things more interesting, I also counted up how many times a player scored more than 0 points, more than 5 points, more than 10 points, and so on all the way up to 30 points. To go one step further, I then calculated the percentage of their starts that resulted in each point bucket. (A player that scores 15 points every week they start but only started 6 weeks due to injury should project higher than a player who scored 15 points 6 times and 5 points 12 times but started every week. The percentage step accounts for this.)

With all of these metrics tabulated for every wide receiver of the last decade, I was able to build out a supervised machine learning algorithm (Ridge Regression, for those well-versed in the subject) and train it to predict next-season total fantasy points based on the various metrics from the previous season. For robustness, I also built a simple model that only looks at total points as a predictor of future points, so we have a reference to which we can compare the more complex algorithm’s performance.

The Model

Using a Lasso regression on the full set of 18 input variables with 5-fold cross validation, the best performance came from what I call the “Core-6” model, which keeps total, average, median, and standard deviation as primary metrics, plus two secondary metrics that were the most useful: percentage of starts with 15+ points scored and percentage of games with 30+ points scored.

As expected, total points as a metric does most of the predictive heavy lifting, but the other five features combined make up more than half of the feature importance in the model! Interestingly, median points matters far more than average points, which is another tip of the hat towards using medians instead of averages in football analytics (a hill I am willing to die on)!

The Core-6 model outperformed the simple model by every performance metric, especially in R-squared, which is the one that matters most for predicting the ranked order of receivers, as opposed to their exact point totals. The other number worth your attention is the mean absolute error (MAE). This tells us that on average, we expect our predicted point totals to be off by about 40 points in either direction. It’s something to keep in mind when you see the top 5 wide receivers and wonder why their predicted totals aren’t higher.

The Rankings

Here are the top 60 wide receivers in predicted total fantasy points in 2026. Puka Nacua, Jaxon Smith-Njigba, and Ja’Marr Chase all sit in the top five, which is a pretty solid sanity check. Some surprise names might be Amon-Ra St. Brown at 2, Chris Olave at 5, and George Pickens at 6. These are the three names I would keep an eye on to have a bigger year than the average person expects. (Note: You won’t see any rookies listed here because this model is trained on prior-year NFL data.)

Caveat 1 - Injuries

You may immediately notice some rankings that don’t make sense. Remember, the model has no knowledge of which players are on the same team and when injuries occurred. So when you see Michael Wilson ranked 10th and Marvin Harrison Jr. ranked 41st, you have to remember that there is some extra work to do. Let’s proceed with this example.

Cardinals Receivers

In the first 10 weeks of the 2025 season, Marvin Harrison Jr. had 525 yards and 4 touchdowns before suffering from appendicitis. From Week 11 on, Michael Wilson became the primary receiving target and had 775 yards and 6 touchdowns. Together, they produced 1300 yards and 10 touchdowns, but in separate halves of the season. In reality, I would swap their locations in the ranking. I may bump down Marvin Harrison Jr. a bit from 10th to, say, 15th, and Michael Wilson up from 41st to 30th, just because Michael Wilson did have six 15-point games and two 30-point games in 8 weeks, whereas Marvin Harrison Jr. had just three 15-point games and zero 30-point games in 10 weeks. It seems like Michael Wilson has earned his way into more targets and they are going to be more of a 1A-1B than a 1-2 in the pecking order going forward.

You can do the same exercise with players like Rome Odunze and Luther Burden and mentally adjust them in the rankings accordingly.

Caveat 2 - Team Changes

The other limitation of the model is that it does not factor in which players changed teams in the off-season. As such, Courtland Sutton ranked 16th and Jaylen Waddle ranked 20th is a bit logically inconsistent. It’s unlikely that they both have 1000-yard, 7-touchdown seasons in 2026 given that they now compete for targets.

Broncos Receivers

I’m actually not that bothered by the rankings themselves for Sutton and Waddle, if we are projecting their likely outcomes. The problem is that one of them is probably going to vastly underperform their ranking while the other meets or slightly exceeds it, just depending on who catches more passes, and especially, more touchdowns. At this point, picking one over the other seems impossible to predict. Sutton has more chemistry built with Nix, and was a reliable target over the past two seasons. In 2025, he had the second-most third down conversions among all wide receivers. On the other hand, the Broncos just invested significant capital to acquire Waddle, who is a more explosive playmaker who ranked 12th in yards per route run last year in a bad offense. If I had to guess Waddle vultures most of Sutton’s production in 2026 and finishes top 10, while Sutton lands outside the top 20, rather than both around 15.

This line of thinking also applies to Michael Pittman and D.K. Metcalf, A.J. Brown and DeVonta Smith, Mike Evans and Ricky Pearsall, etc.

49ers Receivers

Speaking of Mike Evans and Ricky Pearsall, I have to devote my own section to these two players. In 2026, I don’t think this is a case of one player outperforming another from a fantasy points perspective. I think that they will both have very productive seasons, but for different reasons. It’s my personal prediction that Mike Evans will be the red zone weapon for the 49ers, putting up only about 700 yards but scoring a dozen or more touchdowns, a la Davante Adams with the Rams in 2025. Ricky Pearsall, on the other hand, is going to get a ton of targets in my estimation, racking up plenty of receptions and yards but not scoring many touchdowns, a la CeeDee Lamb with the Cowboys in 2025.

Mike Evans’ route tree is mostly slants, outs, digs, and go balls. In limited action in 2025, his calling card was go balls; Evans had the highest average separation score of that route type across all wide receivers. That’s likely his main role on the 49ers now in 2026. Consider Ricky Pearsall. He’s the hitch, crosser, and corner route runner in Shanahan’s offense. If the 49ers are in the red zone often, I expect Mike Evans to be the go-to target based on his route tree. However, to hit explosive plays to move the ball down the field, I expect Ricky Pearsall to feast, especially with Mike Evans drawing safety attention on go routes and digs.

The hitch is a staple of Kyle Shanahan’s passing offense with the 49ers. In 2025, Kendrick Bourne had the highest average separation score on hitches. Yes, Kendrick Bourne, who just got paid $5 APY by the Cardinals. It was so effective that Bourne had 20 catches for 342 yards in Weeks 5-8 when Ricky Pearsall missed time. And in Weeks 1-4 when Ricky Pearsall was healthy and playing that role, he had 20 catches for 327 yards himself. That’s a 17-game pace of 90 catches for over 1300 yards. That’s 175 fantasy points right there. Here’s the thing: In the first 8 weeks, neither Bourne nor Pearsall scored a receiving touchdown. Jauan Jennings was the primary go ball and dig runner for the 49ers, scoring 9 touchdowns but just 643 yards in that role. He’s gone in 2026, and Mike Evans is in.

So, I’m expecting Ricky Pearsall to have easily over 1000 yards but probably very few touchdowns. Mike Evans is certainly a better player than Jauan Jennings, but in a similar role, probably only puts up 800 yards with quite a few touchdowns. I think both players will finish with about the same fantasy points in the end, but from two distinctly different sources of their production.

Blending Analytics with Context

Here’s the thing: Quantitative analytics never tell the full story, but sometimes reveal data that debunks our preconceived notions based on logical fallacies. Without it, we are prone to errors in subjectivity and bias. However, it’s all but impossible to capture the complete picture, and it’s important to use context when interpreting the output of a model on our screen. That being said, the following ranking is my attempt at taking an algorithm that is mathematically proven to be an excellent predictor of wide receiver production in the coming season, and using my football knowledge to identify outlier cases that are misplaced due to missing context. In January, you can all yell at me about how wrong I was about this player or that player, but I think blending multiple perspectives is a good thing, and this methodology does that to a tee.

Eclectic Wide Receiver Rankings

Here it is: My top 50 wide receivers for the 2026 season, starting with the machine learning model and finishing with custom tweaks to the projections based on injury context, team changes, and other miscellaneous factors not captured in the raw data.

Puka Nacua, Ja’Marr Chase, and Amon-Ra St. Brown round out the top 3, with some noteworthy risers and fallers in the top 25 relative to the Core-6 ranking.

Read more

A team’s “rest differential” for a particular week is the number of days since their last game, subtracted by the number of days since their opponent’s last game. For a whole season, their “net rest” is therefore their total rest differential across all 17 games (or 16, prior to 2021). If a team has a positive net rest, they have, on average, more days in between their games than the NFL median. Vice versa for negative net rest.

Schedule Variance

For each season dating back to 2002, I’ve calculated the variance in net rest among the 32 NFL teams. We only go back to 2002 because that was the first year post-realignment, when the Houston Texans were added and division lines were redrawn. If we include earlier years, we lose homogeneity in our data.

Net rest variance is simply a measure of how spread out the teams’ net rests are in a given year. If the 32 teams all had similar net rests, with values close to zero, the variance would be low. If the 32 teams had very different net rests, with high positive values and low negative values, the variance would be high. Mathematically, the variance is the square of the standard deviation. The higher the variance, the more inequity in rest advantage across the league.

The figure above shows the net rest variance for each NFL season. From 2002 to 2017, the variance fluctuated from year to year, but maintained a consistent rolling average. Since 2018, net rest variance has slowly increased from its lowest value since 2007 to its highest level of all-time in the upcoming 2026 season. This change can be seen visually, but it is also reflected in three separate statistical tests, which all reached a significant level. It's not important you are familiar with those tests, but for the curious, they are listed in the plot’s subtitle.

Why is this happening? One possible explanation is the increasing frequency of NFL seasons with a high density of primetime games with a small number of teams. In 2018, no team had more than 5 primetime games. In 2023, the NFL had four teams play 6 primetime games each, and another seven play 5 primetime games each. In 2024 it was three teams with 6 primetime games and eleven teams with 5 primetime games. In the upcoming 2026 season, not one but two teams have 7 primetime games, and another four with 6 primetime games. More primetime games means more “mini-bye” weeks where one team has extra rest before or after their scheduled primetime game. The less these primetime slots are spread out among all 32 teams, the more net rest inequity we are bound to see.

Another factor worth mentioning is the NFL’s increasing push for more international games. Long flights and different time zones make for a larger number of unique time slots for scheduling games, which means more variance in rest days and thus net rest as well.

Rest Advantage

With net rest imbalance becoming more and more relevant in the NFL, a league notorious for its parity between teams, the obvious next question to ask is: Does a season-long rest advantage equate to more wins? To do this, I paired each team’s net rest with their win percentage in that same season (using a percentage instead of total wins controls for the change from 16 games to 17 games in 2021). When we display all these data points on a plot and fit a trend line, we get the following picture:

In the figure above, you’ll see that there is no correlation between net rest advantage and wins. Teams further to the right are no more likely to be near the top of the graph than teams to the left. Similarly, teams further to the left are no more likely to be near the bottom of the graph than teams on the right. An R-squared value of 0.001 is extremely small; a perfect correlation would yield a value of 1, and perfect randomness would yield a value of 0. The respective statistical tests all say the same thing: there is no evidence of an effect on win percentage by net rest.

This may be surprising, but it is great news! The schedule is not unfair to certain teams on account of how clustered or dispersed their games are. Let’s explore some likely explanations. For one, the 2011 Collective Bargaining Agreement mandates players take four days off during a bye week. Because they spend less time on the field, the net rest "advantage" of a bye just transforms into extended time away from the facility, rather than extra physical preparation or recovery time.

Another factor worth considering is that NFL players and coaches often thrive on strict weekly routines. In this regard, extra rest can have a detrimental effect that cancels out its perceived benefits, causing a person or a whole team to lose their rhythm, overthink, and/or feel rusty from an extended time since their last competition. The biggest and most important takeaway however, is simply that there are so many factors that go into determining the outcome of an NFL game that outweigh any potential advantage from rest days. Weather, injury attrition, and coaching matchups are all highly variable and likely to play a bigger role in an NFL game than having slightly more time to prepare.

Closing Thoughts

If at any point this season, you hear someone complaining about rest differential, because their team “had to play X games in Y days” and that’s “so unfair” to the players, you know now that their argument holds no water. (Feel free to link them to this article!) The empirical data speaks for itself. Take this information and empower yourself with it. And if you were worried about how many primetime games your team has to play, don’t be! Enjoy the high-stakes competition for what it is, because the team with the shorter week has no real disadvantage, despite what you may hear from national pundits.

Trust the process over the results.

Filtering the Data

The data I’m using is play-by-play data from the nflfastR package in R. This is a massive data frame with hundreds of recorded variables in any given play, allowing use to filter by, let’s say, all the plays that were on 2nd down. Or 2nd downs in the 3rd quarter. Or 2nd downs in the 3rd quarter that were a rush attempt. Or 2nd downs in the 3rd quarter that were a rush attempt in the red zone. You get the idea.

The first thing I did was filter the data for regular season games only. Since we’re going to be comparing quarterbacks from different teams, and not every team plays in the playoffs, it only makes sense to exclude playoff games so that we maintain an apples-to-apples comparison. The next move was an obvious one: exclude any plays that were handoffs (minimal quarterback involvement). I should note that rush attempts were still included if it was the quarterback himself doing the rushing. In order to remove potential noise in the EPA and success rate data, I also excluded any plays that were a kneel or a spike, or had a penalty called.

My next choice was to exclude pass attempts with negative air yards, or in other words, passes that were thrown to players behind the line of scrimmage. This was my best approximation for removing screens and screen-like plays from the data, where again there is minimal quarterback involvement and the play is effectively an extension of the run game. I also excluded plays that occurred while the win probability of either team was more than 99%. This is what is commonly referred to as “garbage time” in NFL games. This removes situations from the data where, for example, a quarterback is down by multiple scores with under a minute left in the game. If he throws an interception on a desperation heave to the end zone, should that really factor into the analysis of his performance? I would think not.

The most pivotal filter I applied to the data was a strict inclusion criteria of greater than 60% probability of a pass pre-snap. This is commonly referred to as an “expected pass situation.” In NFL analytics, the use of 60% as the threshold percentage is standard. 55-60% is considered pass-leaning, and 60-65% or higher is considered pass-heavy. So, using 60% gives us a medium pass likelihood. I wanted to eliminate plays where the defense was keyed in on stopping a run and the offensive surprised them with a pass. To me, this is akin to playing on “easy mode” as a quarterback. To really evaluate their performance, we need to know how they fair when a defense is playing with at least a decent expectation of a pass play.

Yards After Catch

Yards after the catch (YAC) is often cited in reference to a receivers ability to catalyze a successful play into an explosive one. As such, it’s not so much a description of quarterback performance on a given play; by the time the ball is in his teammate’s hands, the quarterback’s job is done. For this reason I performed an adjustment calculation to EPA and success rate according to YAC. Each play has an expected YAC at the catch point, based on where on the field the ball was caught. It also then has an actual YAC that the receiver achieved, leading to the eventual success (or failure) and EPA of the play. Instead of using this actual YAC and its associated EPA, I used the expected YAC to get the theoretical EPA of each play, and the probability of a successful (positive EPA) play.

This achieves two things. First, it more accurately captures the quarterback-specific contribution to each play. If a receiver is expected to get 10 yards after the catch, but breaks a tackle or jukes a defender to get 25 yards after the catch, that should be a reflection of the receiver’s performance, not the quarterback’s. Secondly, it prevents a quarterback from being unfairly penalized for a fumble by the receiver after the catch (which would be negative EPA and an unsuccessful play). All of this helps to reduce noise in the data.

With that, let’s get into the results of our adjusted quarterback metrics in 2025.

Read more

There was a time, not too long ago, when tight ends were drafted out of college as a future investment. There was a recognition that the demands of the position at the NFL level were so uniquely challenging that their potential would not be fully actualized until their second, third, or fourth year with the team. Think back to David Njoku, Dallas Goedert, and T.J. Hockenson. None of these guys cleared 400 yards as a rookie. Even the All-Pro George Kittle had only 63 targets for 515 yards in his first year in the NFL, and now he’s an all-time great who we expect to have a 1000-yard season every year.

Traditionally speaking, the function of the tight end is to be a two-way player who can serve as an extra blocker, or an extra receiver, on any given play. They align “tight” to the “end” of the offensive line, posing a dual-threat to opposing defenses. Because they play multiple roles, learning the playbook and building their skill set takes that much longer. Or, at least, it used to.

Fast forward to 2026. Over the last several years, the NFL has seen Kyle Pitts, Sam LaPorta, Brock Bowers, and Tyler Warren all have Pro-Bowl seasons as rookies. Moreover, offensive usage rates of multiple tight end packages (12 and 13 personnel) have steadily climbed from 21% in 2022 to 28% in 2025. With tight ends becoming increasingly more relevant to NFL success, I decided to look at the skill sets of different players at the position and how it translates to their playing time on the field.

Catching passes and blocking defenders

To examine tight end archetypes, I defined two main categories of their game: receiving passes and run blocking. Tight ends also pass block, but since it occurs the least often of the three, I decided to omit pass blocking as a data point to save ourselves from having to conceptualize three-dimensional plots. That is one limitation of this analysis, but it is one I am comfortable with. Secondly, I defined their receiving and blocking prowess by two factors: how often they do it, and how effective it is. For receiving, I looked at routes run per team pass snap and EPA per route run. Similarly, for blocking, I looked at run block assignments per team run snap and EPA per run block. Of course, this data is going to be dependent on more than just the skill of the tight end, as both usage rate and EPA per play depend on several other factors, like play calling, offensive line strength, quarterbacking, etc. But generally, this is a limitation of any player-specific analysis in a team sport, and it’s one that is almost impossible to work around. So we move forward, keeping this in the back of our mind and using our results to draw broad conclusions, not specific ones.

The last thing to note is that I applied a set of inclusion criteria to determine which tight ends would be included in this analysis. In the hopes of getting approximately one tight end per team, I only included tight ends who were top 40 in total receiving EPA in 2025, with at least 200 routes run and at least 100 run blocking snaps. This yielded a list of 33 tight ends, which I further narrowed down to 31 by manually excluding Michael Mayer and Jake Tonges, who had temporary playing time as backups for injured starters Brock Bowers and George Kittle last year.

Receiving

In the figure above you’ll see four quadrants: tight ends with high usage in the passing game, tight ends with high efficiency in the passing game, and those with both, or neither. George Kittle, Sam LaPorta, and Trey McBride lead the way in the “both” category. They are heavily utilized on passing downs and catch a lot of passes on those downs. Tucker Kraft only played half a season, but was highly efficient on route participation that was more or less equal to that of players like Hunter Henry and Dalton Schultz.

As a Seahawks fan, I’d like to point out the location of A.J. Barner on this graph. He ranks 21st in route participation, but 7th in EPA per route. On the select opportunities he was given, he was quite effective as a receiver. In this regard, he was kind of a Loveland-lite type of player in 2025.

Blocking

This next figure on run blocking follows the same structure as the previous plot on pass catching. Once again, NFC North tight ends Sam LaPorta and Tucker Kraft stand out in the high-volume, high-efficiency quadrant. Tyler Warren also makes an appearance here as a rookie. Brock Bowers did a decent amount of run blocking last year, but it didn’t translate to any improvement in the Raiders’ run game.

Note the similar usage of ex-Seahawk Colby Parkinson and current Seahawk A.J. Barner. In 2025, the Rams were much better than the Seahawks running the ball out of 12 and 13 personnel. I have to imagine that will be a focus for areas of improvement in Seattle’s offseason.

Receiving and Blocking

Putting these two plots together, we can make the following graph, which summarizes the archetypes of different tight ends around the NFL. Tucker Kraft, Sam LaPorta, and George Kittle are true two-way players that lift both the run game and the pass game of their respective offenses. Then you have the category of tight end whose bread is buttered in run blocking: Colby Parkinson, Dawson Knox, Darnell Washington, Cole Kmet. And lastly, players like Brock Bowers, Dalton Schultz, Colston Loveland, and Travis Kelce primarily serve as pass-catching options for their quarterbacks.

A.J. Barner lands pretty close the center of both axes on this figure. If we define a good tight end as one who is useful and efficient as both a blocker and receiver, then he ranks approximately 10th if we start counting at the top right corner and work our way toward the bottom left corner. Listing out the top 10:

1. Tucker Kraft

2. Sam LaPorta

3. George Kittle

4. Kyle Pitts

5. Trey McBride

6. Tyler Warren

7. Harold Fannin

8. Dallas Goedert

9. Hunter Henry

10. A.J. Barner

That feels pretty close to what we might expect if we just started naming great tight ends off the top of our head. An obvious caveat here is that Brock Bowers’ data point is probably getting unfairly weighted heavily in the run game by the Raiders’ offensive line struggles.

A.J. Barner

I think it’s time we start talking about A.J. Barner as one of the better tight ends in the NFL. In only his second season (and his first with a competent offensive coordinator), we see him in the company of players like Dallas Goedert, Travis Kelce, and Colston Loveland by a composite metric of pass catching and run blocking. There’s an obvious comparison to be made here between him and Trey McBride. Both were non-first round picks who more than doubled their production from their rookie season to their sophomore season. It was McBride’s third year in the NFL that was his big breakout, racking up 100+ receptions and 1000+ receiving yards, while leading the Cardinals to the 8th best rushing offense in the NFL by EPA per play. With George Kittle’s tight ends coach, Brian Fleury, coming from the 49ers to the Seahawks this off-season, could we see a similar leap from A.J. Barner in 2026?

As always, trust the process over the results.

The figure above describes the Seahawks schedule from Week 1 to Week 18, with taller bars representing tougher opponents. How opponent strength is calculated for each game is described in my previous article, which you can read here:

Week 1 - Home Game vs New England Patriots (W)

The Seahawks will play the very first game of the 2026 NFL season on a Wednesday primetime slot, hosting the Patriots in a Super Bowl rematch. The Seahawks easily beat the Patriots in the Super Bowl, and I don’t expect the results of this Week 1 matchup to be any different. Since that game, Patriots head coach Mike Vrabel has been the center of a scandal with NFL insider Dianna Russini in which the exact details of an apparent infidelity are yet to be revealed. This kind of off-season distraction can only worsen their preparedness for a game that will have tons of eyeballs watching it. Their off-season moves don’t move the needle much for me, and certainly not enough to outweigh the bad vibes in the building and hangover from the Super Bowl loss. This game feels destined to be a case of pouring salt in the wound, and nothing more.

Week 2 - Away Game at Arizona Cardinals (W)

To no one’s surprise, I believe the Seahawks start the season 2-0 here. The last time these two teams faced, the Seahawks ran up the score to 35-0 by the second quarter. The Cardinals quarterback situation is an uninspiring mix of Jacoby Brissett and Carson Beck, with third overall draft pick Jeremiyah Love being the most exciting addition to the roster. I don’t think running backs generally have much of an impact on team strength or win probability, and besides, the Seahawks had one of the best run defenses of all time last year and shut down Christian McCaffrey when they faced him. This should be an easy win.

Week 3 - Away Game at Washington Commanders (W)

The story here is going to be much of the same as Week 2. The Commanders were no match for the Seahawks last year, losing by 24 points in a primetime game in their own stadium. While they made a ton of free agent additions to the defense and added highly-touted rookie linebacker Sonny Styles, the improvements on that side of the ball can only take them so far when their offensive infrastructure includes the same five starters on the offensive line that ranked ~20th in blocking win rate. 30-year old Terry McLaurin poses the only real threat among their receivers, tight ends, and running backs, offering few options for Jayden Daniels to create explosive plays on a consistent basis. A 3-0 start for the Seahawks seems highly likely.

Week 4 - Home Game vs Los Angeles Chargers (W)

On paper, this game is the biggest test in the Seahawks 2026 season so far. The Chargers hired wide-zone play action offensive scheme guru Mike McDaniel to be their new offensive coordinator. They fortified the interior offensive line with free agent additions and draft picks, and hope to have a healthy Rashawn Slater and Joe Alt back in the starting lineup at offensive tackle. However, the picture on defense seems to be largely the same, which I find to be a bit of an overrated unit. Over the second half of the season last year, their defense was 17th in success rate allowed. The Chargers will be coming off of a Week 3 road game against the Bills, to play another road game against the Seahawks. Their front seven seems unequipped to handle consecutive battles against teams with good offensive lines that like to run the ball and assert trench dominance in their home stadium. For that reason, I have the Seahawks improving to 4-0 in an exciting, close game.

Week 5 - Home Game vs San Francisco 49ers (W)

Spoiler alert: You won’t catch me predicting a Seahawks loss to the 49ers any time soon. I believe the 49ers to be a team with dwindling talent, desperately clinging to a 2019-2023 championship window in which a Super Bowl victory evaded their grasp. Each off-season, they seem to make win-now moves that fail to set up the team for real, sustainable success. This year, it was signing 32-year old free agent Mike Evans, adding Christian Kirk (coming off a 238 yard season), blowing their early draft picks on players that had consensus Day 3 grades, and apparently trotting out the ghost of Trent Williams to play left tackle one last time. The Seahawks beat this team 41-6 the last time they faced. This should be another blowout.

Week 6 - Away Game at Denver Broncos (W)

Up to this point, I’ve predicted the Seahawks to go 5-0 heading into a Thursday Night Football away game against a team that played in the AFC Championship last year. The Broncos are essentially running back the same roster, except for the notable addition of Jaylen Waddle to their receiver room. The problem is, I think this team outperformed their true talent in the 2025 season. Their offense ranked only 15th in DVOA last year. Their defense put up great numbers, but had a fairly easy schedule. Half of their wins were by 4 points or less. In two games against the Raiders, who finished with the worst record in the NFL, the Broncos outscored them 10-7 and 24-17. Some analysts will have this game as a loss, but I think the Seahawks expose the Broncos a bit in this nationally televised game. Seattle’s defensive line can make their offensive line look mortal, and Bo Nix is not the caliber of quarterback that can outwit Mike Macdonald’s defensive scheme post-snap.

Week 7 - Home Game vs Kansas City Chiefs (L)

The Seahawks have back-to-back primetime games here with a Sunday Night Football matchup versus the Chiefs. Generally speaking, I think the Seahawks have the more talented 53-man roster front to back, but this feels like a game where the three-time AFC champs from 2022 to 2024 prove they are back after a disappointing 2025 season. They’ve added three impactful rookies to the defense: cornerback Mansoor Delane, defensive tackle Peter Woods, and pass rusher R Mason Thomas. It’s easy to see how these players fit into a defense coached by Steve Spagnuolo, who is notorious for coming up with game plans that stress opposing quarterbacks in big games. I have a feeling that this game is hyped up similarly to the Week 11 match between the Rams and Seahawks last year, and has a similar result. A low-scoring, easily winnable game is lost by Seattle due to forced errors by Sam Darnold and conservative turnover-free football by Patrick Mahomes that gets the job done in lackluster fashion.

Week 8 - Home Game vs Chicago Bears (W)

All eyes will be on this game as the Seahawks play in their third of three consecutive primetime game, finishing the streak with a match against the visiting Bears in Monday Night Football. Chicago will be coming off of games against the Packers, Falcons, and Patriots. All three of those teams have decent offensive lines, and I would expect the Bears’ front seven to be gassed by the time they have to travel west to face the Seahawks. Their defensive line is a weakness of the team, and I view this game similarly to the Chargers matchup. Even if this game ends up being a shootout, I’d take Sam Darnold, Jadarian Price, and Jaxon Smith-Njigba over Caleb Williams, D’Andre Swift, and the Bears’ receivers. I think the Seahawks will be 7-1 through Week 8.

Week 9 - Home Game vs Arizona Cardinals (W)

What a nice opportunity for the Seahawks to catch their breath after three straight primetime battles. If there’s a trap game on the schedule, this is it. However, the Cardinals will have just faced the Lions, Rams, Broncos, and Cowboys leading up to this game. There’s no way that defense will have enough gas left in the tank to hold back Seattle, even if the Seahawks aren’t operating at full strength. I don’t see any reason why the Seahawks wouldn’t sweep Arizona again this season.

Week 10 - Away Game at Las Vegas Raiders (W)

Klint Kubiak versus Mikę Macdonald. That will be the headline for this game. What will get significantly less attention is the other side of the ball: Brian Fleury versus Rob Leonard. At this point, most casual football enjoyers are going, “Who?” and typing Wikipedia into the search bar. Brian Fleury had been the 49ers’ tight ends coach since 2022 and their run game coordinator in 2025. He’s now the offensive coordinator for the Seahawks. Rob Leonard had been the Raiders’ defensive line coach since 2023 and their run game coordinator in 2025. He’s now the defensive coordinator for the Raiders. So while the two Super Bowl winning play callers battle it out on one side of the ball, two first time coordinators will match wits on the other. This is a classic, vintage Shanahan run game versus Carroll run defense. Now, when the 49ers faced the Raiders in 2025, the Raiders built up a 13-3 lead, but the 49ers clawed their way back with field goals and ended up winning on a walk-off 59-yard field goal, 22-19. I have a sinking feeling that this is going to be an unexpected, tough, low-scoring game like the Colts-Seahawks game from last year. I have to think ultimately Seattle’s roster talent advantage wins out in the end, but I wouldn’t be surprised if this is the one head-scratching loss that no one predicted in their 2026 schedule.

Week 11 - BYE

At this point, I have the Seahawks sitting at a 10-1 record on their bye week. If they manage to secure 9 or 10 wins at this point in the season, they’ll set themselves up nicely for a chance to reclaim the number one seed in the NFC. I’m a big fan of having the bye in Week 11. It’s a little bit after the midpoint of the season, giving the team an opportunity to reset before the final push in which they face three NFC East teams, then the Rams twice in a three-week span to close out the regular season. If the Seahawks go on to play three playoff games, as one would hope, the Week 11 bye is actually perfect situated in the center of their full slate of games. That means they don’t have any stretches of 12+ games without a week off, where minor injuries can turn into major injuries. Staying healthy through the NFL season is paramount to making a deep postseason run.

Week 12 - Away Game at San Francisco 49ers (W)

For this game, the Seahawks are coming off their bye and the 49ers are playing their second straight home game after hosting the Vikings the week before. I have the Seahawks beating the 49ers in Week 5, and I don’t think any differently of this game. Home field advantage is dwindling in the modern NFL; the bigger factor here is the Seahawks having an extra week to prepare. Seattle is 3-0 coming off a bye under coach Macdonald, outscoring opponents 99-37 (and two of those games were against the 49ers!) while never allowing more than 17 points to their opponents.

Week 13 - Home Game vs Dallas Cowboys (L)

Having this game as a loss may surprise some people. In fact, I surprised myself a bit when thinking through this matchup. At first glance, it may be easy to look at the Cowboys’ 7-9 record last year and their lack of significant playoff success since… well, a long time, and dismiss this game as a win for Seattle. However, the NFL is a styles-makes-fights league, and the Cowboys’ style is a bad matchup for the Seahawks. Think about the team that gave the Seahawks the most trouble last year: the Rams. What were their defining traits? They had an offense built around the quarterback, who can be an assassin with his arm from a clean pocket, paired with two excellent complementary receivers to stress the secondary: an elite route runner and separator in the red zone, and a big-bodied contested catch threat down the sideline and up the middle of the field. The defense was good-not-great, but found ways to create opportunistic turnovers. Sounds a lot like the Cowboys to me, who have Dak Prescott, CeeDee Lamb, and George Pickens. In free agency, they added cornerback Cobie Durant, a key player in the Rams’ secondary that intercepted Sam Darnold 6 times last year. They also drafted versatile safety Caleb Downs in the first round, who some were talking about as a top 3 player in the draft if you set positional value aside. I think the Seahawks take their second loss of the season here and fall to 10-2.

Week 14 - Home Game vs New York Giants (W)

The Giants are an exciting young team for a New York fanbase that has little to cheer for lately, but I think this is a team that is still at least a year away from seriously competing. The receiver room has little behind Malik Nabers, and the roster overall has a serious shortage of star potential, especially after sending Dexter Lawrence to Cincinnati. A rebuilding New York franchise should have no problem taking a loss to the defending champs on the road.

Week 15 - Away Game at Philadelphia Eagles (W)

The Seahawks are 8-0 in their last eight games versus the Eagles. George W. Bush was still in office the last time the Eagles beat the Seahawks head-to-head. Normally, I would say this is the type of statistic that has no relevance to the particular match taking place in 2026. While still true to a large extent, I do think there is something to this winning streak. The Eagles have been managed by Howie Roseman for all eight of these games. Since taking over in 2010, he’s established a particular brand of football in Philadelphia: aggressively maximizing the championship opportunity of every season through a versatile and flexible roster that is constantly rebuilding. To his credit, this philosophy has made them consistently competitive and brought them two Super Bowl wins, one in 2017 and one in 2024. However, it has not made them successful against the Seahawks, who have established a very similar culture in the same time frame. Seattle’s general manager John Schneider, who also took over in 2010, prioritizes many of the same things and also won two Super Bowls implementing these methods. When the Eagles face the Seahawks, their primary structural advantage is neutralized. And for my money, Seattle does it a little bit differently that tips the scales in their favor each time they meet on the field; Seahawks rosters always feature an extremely tight brotherhood. On the flip side, the Eagles rosters are typically a hodgepodge of different personalities brought together by Roseman’s identification of their talent. Take Riq Woolen for example: a uniquely physically talented cornerback with a propensity for making boneheaded plays at the wrong place at the wrong time, who was at one point benched for a lack of effort on the field. The Seahawks moved on from him this off-season because he didn't fit their identity, while the Eagles happily scooped him up in free agency. Woolen is a microcosm of the way in which Seattle and Philadelphia are opposite sides of the same coin. I think the Seahawks beat the Eagles once again and improve to 12-2.

Week 16 - Home Game vs Los Angeles Rams (L)

The Seahawks had a road primetime game against the Eagles the week prior, in a battle of the trenches that likely takes a physical toll on the team. In Week 16, they have to make a quick heel turn to play the Rams in another primetime game on Christmas Day, with a short week of practice. The Rams will be hungry for revenge after getting knocked out of the NFC Championship in Seattle last year. I believe that Sean McVay and Mike Macdonald are the type of great play callers that will alternate getting the best of each other every time they face off. Mike Macdonald stopped the Rams’ offense in the second half of their last meeting, so I expect Sean McVay to have some answers in this game. This game has “emotional loss that brings the Seahawks back down to earth after a hot start” written all over it. I think Seattle will be 12-3 after Week 16.

Week 17 - Away Game at Carolina Panthers (W)

The Seahawks also played the Panthers in Week 17 last year, beating Carolina on their own turf in a 27-10 blowout win. Seattle should have no problem doing it once again in 2026, as the Panthers have one of the toughest schedules in the league this year and will be playing their 12th straight game when they face the Seahawks, due to a bye located in Week 5. The Panthers roster is largely the same as last year’s, which ranked 25th in DVOA. If the Seahawks are who we think they are, they’ll win this game convincingly.

Week 18 - Away Game at Los Angeles Rams (W)

The final game of the Seahawks’ regular season is a divisional battle against their main rival. It’s quite likely that this game decides the winner of the NFC West, and possibly the number one seed in the NFC. Fortunately for the Seahawks, they found themselves in this exact situation during last year’s Super Bowl run, where they had to defeat the 49ers in their home stadium in order to secure the NFC West division title and number one seed before going on to play the Super Bowl in that very same stadium. They say history doesn’t repeat itself, but it sure does rhyme; Los Angeles is hosting the next Super Bowl. The familiarity of the situation should give the Seahawks a tremendous advantage, who have everything to play for and everything to prove after what I am predicting as a loss in Week 16. As previously mentioned, I think that both teams’ head coaches are on track to alternate victories in the evolving chess match between their respective schemes that defines the modern NFL landscape. If Sean McVay ends up on top in Week 16, I have no doubt Mike Macdonald will have answers of his own in Week 18 to put his squad on top. I am predicting the Seahawks to win this game and finish the season 14-3 with a first round playoff bye for the second straight season.

Final Record Prediction

I think that the Seahawks will finish with a 14-3 record in 2026, first in the NFC West and first in the NFC. Repeating is never easy as defending Super Bowl champions, but I believe that Mike Macdonald is a head coach that has all the necessary ingredients to avoid common pitfalls and make it happen. Seattle’s roster replaced all of their free agency departures through the draft, adding running back Jadarian Price, safety Bud Clark, and cornerback Julian Neal, who have all been praised by analysts close to the team as excellent personality and scheme fits for the Seahawks. They also added Iowa offensive lineman Beau Stephens, who has experience playing both guard spots in a group that won the Joe Moore award in 2025 for the toughest, most physical, and cohesive offensive line unit in college football nationwide.

The 2013 and 2025 Super Bowl Seahawks teams share a common identity as a group of passionate individuals with a dominant defense and a chip on their shoulder from being underestimated. In this 2026 iteration of the team, I see another group capable of returning to the Super Bowl after winning the prior one. This team still has something to prove to the national media as the conversation this off-season continues to discredit their Super Bowl season as “a down year for the AFC” or “a fortuitous bust” in the NFC Championship. If destiny exists, this team’s is to right the wrongs of the Malcom Butler interception and win back-to-back Super Bowls, properly establishing the next NFL dynasty for years to come.

Trust the process over the results.

There are two main types of strength of schedule rankings you’ll find online. The first is probably the most widely recognized among casual football observers; it uses last year’s win percentage to determine the strength of each opponent. In fact, this is the methodology used by the official NFL account itself. Such an approach to strength of schedule modeling is easy to do and simple to understand, but for the fan who religiously follows the NFL hiring cycle, free agency, and draft (ahem, myself), a simple count of last year’s wins and losses falls short of expectations with respect to power ranking teams heading into the subsequent season.

Fortunately, there is another way of estimating strength of schedule that has recently gained steam: using each team’s predicted win total in the upcoming season from the lines set by Vegas. It’s easy to see how this would be a better approach. Why would we base a team’s strength off of their performance last year, when what we really want to know is how good of a team they are this year? This way, any roster or coaching changes in the off-season are accounted for in the model. (Note: There is another version of this methodology where we can take all of the opening spread lines from each game of the season and reverse engineer implied team strength, though this cannot be done until the exact schedule is released and bookmakers have set their spreads for all the games.) The system of using Vegas predicted win totals to calculate strength of schedule has been popularized by Warren Sharp this off-season, and looks to become the new standard going forward. I am eager to see if the NFL adopts it next year.

A New Approach

This brings us to a new strength of schedule model that I am going to propose. Why? Well, using predicted win totals is great if you believe that win totals are a great representation of team strength during the season. However, I’m not convinced that that is the case. There is plenty of publicly available research that shows that past wins are less predictive of future wins than advanced metrics like WEPA (weighted expected points added) and DVOA (defense-adjusted value over average). The table below is from an article by Robby Greer which shows exactly that.

Personally, I like total DVOA as team strength metric because it is the most stable from year to year and accounts for opponents faced, which seems like an important adjustment in strength of schedule modeling. I gathered DVOA rankings from 2022 to 2025 and evaluated their correlation from one season to the next, spanning three off-seasons (2022 to 2023, 2023 to 2024, and 2024 to 2025). Then I fit a linear trend line to the data and calculated its R-squared (a statistical measure of correlation ranging from 0 to 1). Just to be sure, I did the same exercise for some other advanced metrics and compared the results, which are summarized here:

As expected, DVOA was the most stable of the advanced metrics. However, somewhat surprisingly, the Vegas predicted win totals were just as correlated to the advanced metrics in the next season as DVOA. It was at this point that I decided a blend of Vegas win totals and DVOA would probably yield the highest correlation with next season’s DVOA. With a weighted average consisting of 60% Vegas predicted win totals and 40% prior season DVOA, the R-squared values of the correlations increased by about 10% relative to what you would get by using DVOA or Vegas win totals alone:

The improvement seen by using a weighted average makes sense because the two metrics capture different things, and by combining them, we get a more holistic picture of each team that is more stable. Vegas win totals capture a “wisdom of the crowds” perception of each team, influenced by many unique factors, like free agent additions, anticipated improvements by the starting quarterback, and/or scheme changes. DVOA on the other hand, adjusts for what the public eye may be missing: teams that over-performed or under-performed on the field relative to their situation: opponents faced, penalties, turnover luck, special teams, etc. This new metric—what should we call it, WAVES (Weighted Average of Vegas and Efficiency Statistics)?— provides a framework with which we can build out a new strength of schedule model.

Read more

Methodology

I’ve collected 16 different top 100 big boards for this analysis: 13 that were published and widely cited ahead of the draft, and 3 consensus rankings. I’ve limited these boards to just the top 100 players for a couple of reasons. For one, it’s the most common cutoff used, and thus maximizes homogeneity in our data. Secondly, it aligns nicely with the end of Day 2 of the draft, which is also when most people stop paying close attention. Don’t get me wrong, Day 3 is great for searching for diamonds in the rough, but it’s much harder to consistently rank players that are seen as “dart throws” with late-round picks.

As for evaluating these 16 big boards, I’ve used an approach that converts each pick number to its equivalent value in the Rich Hill Model, which assigns increasingly greater value to higher picks. This allows us to score big boards in a way that places greater emphasis on accuracy earlier in the draft. Think about it: if a player ranked 1st on a big board is actually drafted 5th overall, that discrepancy should be quantified as a greater difference than if a player ranked 91st on a big board were drafted 95th overall.

Then, each board was evaluated via two metrics. The first metric was generated by fitting a logarithmic curve to the Rich Hill value where each prospect was taken as a function of their 1 to 100 ranking and calculating the R-squared for the curve. All you need to know about the R-squared value is that it ranges from 0 to 1, with higher values (like 0.9) being a better fit. The second metric was generated by taking the implied Rich Hill value of each prospect’s rank and comparing it to their actual value as they were drafted. All of these differences were squared, summed up into a total, and square rooted. This is also known as the standard deviation, for which smaller numbers represent closer alignment.

The last thing I need to mention about this analysis is that a specifically omitted Jermod McCoy from all of the boards. He was unanimously a first-round talent and accordingly ranked as such across the board (pun intended). Due to injury concerns apparently more closely understood by NFL teams than these board makers, he fell outside the top 100 on draft day. For this reason, he’s been excluded from the calculations to avoid unnecessary noise; it seems unfair to apply a greater penalty to a big board for ranking him 12th than to another big board for ranking him 14th.

Results

Here are the results of the evaluation process. There was a strong correlation between R-squared values and standard deviations, as you might expect. In only a few cases did a board score markedly better in one metric than the other. Generally speaking, the consensus boards were more predictive than the individual big board creators. The top spot goes to the Mock Draft Database, though they were essentially tied by The Huddle Report, which has been one of the best top 100 big boards in recent years (namely 2023 and 2024) and has apparently reclaimed the throne here in 2026.

Both Daniel Jeremiah and Arif Hasan seen near the top of the list should come us no surprise, as the widely known NFL Network analyst and infamous consensus board apologist have perfected their respective crafts over the years. I’d also like to shoutout James Foster for his ever so detailed big board that he puts out for free each year that was actually one of the higher ranking boards on this list. Lastly, I’m pleasantly surprised to see my own consensus board cracking the top 7. This was just my second year making such a board, and my first focusing on the top 100.

Closing Thoughts

It’s always important to focus exactly on what the data is telling us and nothing more. By no means is this a ranking of the “best” big boards (whatever “best” means), nor is it a ranking of which board will be most predictive next year. It’s simply telling us which boards were most similar to the collective opinion of the prospects from NFL teams who ended up actually doing the drafting. While it is likely that these same boards are also better at estimating the future success of prospects (NFL scouting departments have access to far more information on college players than the rest of us), it is entirely possible that any one of the boards in the bottom half of the table is actually doing something uniquely better than everyone else and will be proven right in the coming years.

For me, this yearly exercise is just a useful tool for checking which big boards I should be paying attention to when I see a prospect I like and want to know where they are likely to be drafted.

Trust the process over the results.

First of all, I need to define my grading scale. Each pick earns a letter grade determined by a point total. Points are earned in three categories: Steal/Reach, Positional Priority, and Board Status. These criteria are described in the following table:

The “highest expert rank” refers to the eight top 100 big boards I’ve collected from industry experts on the NFL draft. The eight boards are Dane Brugler, Daniel Jeremiah, Kyle Crabbs, Bleacher Report, PFF, Tankathon, The Huddle Report, and Draft Board Guru, which historically have some of the most accurate big boards publicly available. The maximum possible points for a pick is 4, for which the letter grade would be an A+. The fewest possible points is -4, which is an F. The point-to-grade conversion is described here:

One thing I should point out is that some grades are more likely than others. For the point totals near zero, there are more combinations of criteria that could lead to those grades. For example, a pick could earn 1 point for being scored as Consensus Rank, High Priority, Neutral Board, or for being scored as Steal +, Medium Priority, Favorable Board. For the point totals far from zero, there are fewer possible combinations that yield those grades. This gives us a normal distribution, or bell curve, for our grading system:

This should make intuitive sense. We want A+ and F grades to be relatively rare and saved for picks that are exceptionally great or awful. What good does a letter grade do to inform us of the quality of a pick if all the grades were to be equally likely?

So, to recap: Each pick is evaluated in three categories, earning points along the way. All the points are added together, and the total points corresponds to a letter grade. Alright. Here we go.

Round 1, Pick 32: RB Jadarian Price (B+)

The Jadarian Price pick gets a B+ grade. Let’s walk through the evaluation process. Some may consider this selection a reach, but I would argue that Jadarian Price would have been drafted within the first five picks on Day 2 if Seattle had passed on him. Daniel Jeremiah had him ranked 34th and Bleacher Report had him ranked 32nd. He is described by Jeremiah as a “patient runner with terrific vision and balance.” Bleacher Report graded him as a better prospect than Jahmyr Gibbs in 2023. Running back was a high positional priority for the Seahawks. With Ken Walker departing in free agency, Seattle’s only other back with starting NFL experience is Zach Charbonnet, who has an ACL recovery timeline that points to a possible return around early November. The rest of the depth chart is yet to see significant NFL snaps in a starting role, and as such poses a substantial unknown at the position. Lastly, the board status at pick 32 was unfavorable. GM John Schneider said the Seahawks’ offers to trade down fell apart at the last minute. The two obvious choices at pick 32, safety Dillon Thieneman and cornerback Chris Johnson, were taken at picks 25 and 27 respectively. Moreover, Seattle needed a rookie running back who could start right away. Jadarian Price was the last available one with this type of projection, and he was almost certainly not going to be available at Seattle’s next pick (pick 64).

The B+ grade feels about right. While it’s not an exciting, high ranking player that fell to them at pick 32, Jadarian Price was the most logical choice for the state of the roster and the picks they possessed in the draft. If they passed on him, they would, in all likelihood, be without a starting caliber NFL running back for most of the 2026 regular season. Their other high priority positions of need, safety and cornerback, had their best players taken just before pick 32, and the next best choices were also second round talents. The running back talent cliff in this class was much earlier than that of the defensive backs, so it made sense to address this need first.

Round 2, Pick 64: S Bud Clark (B)

The Bud Clark pick gets a B grade. Daniel Jeremiah and The Huddle Report both ranked him in their top 70, and he was drafted 64th overall. To me, this is another consensus rank, neither a steal nor a reach. Safety is another position of high priority on Seattle’s roster, with Coby Bryant joining the Bears in free agency and little depth behind Julian Love and Ty Okada, the latter of which only achieved playing time as a backup in 2025. Bud Clark projects best as a free safety, which was precisely the role the Seahawks needed to fill with rookie talent. I considered the board status at this pick to be neutral because there were other safeties with similar expert ranks (A.J. Haulcy, Jalon Kilgore, Zakee Wheatley) still available and a chance that one of them lasted until their next pick at 96, but likely not the one they wanted most.

This B grade also seems appropriate. Seattle could have instead gone with a cornerback at pick 64 and a safety and pick 96 and likely come out fine. However, the cornerback class was deeper than the safety class and it was probably wise to take the position with the earlier talent cliff first. They could have gotten a steal in offensive guard Emmanuel Pregnon with this pick, who had expert ranks in the late 30’s and early 40’s, but that position was only a medium priority, whereas safety was more critical to address with higher draft capital. That seems deserving of a solid, above average grade, but nothing too fancy.

Round 3: Pick 99: CB Julian Neal (A)

The Julian Neal pick gets the highest grade of the first three rounds, earning an A at pick 99. He was ranked 86th by Daniel Jeremiah and 87th by Bleacher Report, yet Seattle was able to trade back from pick 96 and still get him at pick 99. To me, that’s the definition of a steal. Cornerback was the third of three high priority positions that needed to be addressed with early draft picks, and at this point, the Seahawks already picked a running back and a safety, so this pick needed to be a corner. I scored the board status as Unfavorable because Seattle lacked 4th and 5th round picks this year. Without trading up, they would not be on the clock again until much later in the draft, when all the corners with starting upside would be almost certainly gone.

The A grade for Neal is high, but rightly so. He has great size, length, strength, and explosiveness. He excels at shedding perimeter blocks in run support and making strong open-field tackles, which is something Seattle’s defense values greatly in their secondary players. His stickiness in man coverage at all three levels of the field combined with his long arms to disrupt the catch point make him a near-ideal candidate to fill the role vacated by Riq Woolen’s departure in free agency.

Closing Thoughts

The Seahawks made three top 100 selections and received three grades no lower than a B. That’s good work. And it’s worth noting that these grades are evaluation tools for the process of the picks, not the results of the picks. If Bud Clark becomes an All-Pro safety in his sophomore season, that doesn’t make the B grade wrong. If Julian Neal doesn't make the 53-man roster, that doesn’t make the A grade wrong. It would just mean that, in a specific case, the results did not match the process. We must not be fooled by myopic perspectives. If the process is good, the results will eventually follow as the sample size increases.

Trust the process over the results.