Command Deltas Primer

What is Command Delta?

Command Delta is a metric designed to measure pitch location consistency using pitch location dispersion.

Rather than evaluating whether a pitch is in the strike zone or not, the metric measures how tightly clustered a pitcher’s locations are for a given pitch type.

In other words:

How consistently can a pitcher locate a pitch?

How It Works

For each pitcher and pitch type, we measure the spread of pitch locations using the standard deviation of:

• Horizontal location (plate_x)

• Vertical location (plate_z)

We combine those into a single dispersion metric:

Command Delta = √(σx² + σz²)

Lower values indicate tighter pitch clusters and better command.

Converting to Scouting Grades

To make the metric easier to interpret, Command Delta is converted into a 20–80 scouting scale.

This is done by calculating a z-score within each pitch family and mapping it to:

• 80 – Elite command

• 70 – Plus-plus command

• 60 – Plus command

• 50 – MLB average

• 40 – Below average

• 30 – Poor command

• 20 – Very poor command

Why Pitch Families Are Used

Different pitch types naturally have different movement and command profiles.

To account for this, Command Delta compares pitchers within pitch families:

• Four-Seam Fastballs

• Sinkers

• Cutters

• Sliders

• Sweepers

• Curveballs

• Offspeed pitches

This ensures that pitchers are evaluated against comparable pitch types and their shapes.

Data Source

All pitch data comes from Statcast via the Python library:

pybaseball

The analysis currently includes the 2023–2026 MLB seasons.

Why This Metric Matters

Command is one of the most important and difficult pitching skills to measure.

Traditional statistics often capture results, not process.

Command Delta focuses specifically on the consistency of pitch location, providing a clearer look at a pitcher’s ability to execute.

Command Deltas should not be used in isolation to evaluate command, but rather in conjunction with things like strike and zone rates as well as heat maps.

A potential limitation of this metric is that it may assign a lower command grade to a pitch if there are multiple intended locations, so be mindful of that when using referring to Command Deltas.

Explore the Dashboard

Use the pages in the sidebar of the web app to:

• View command leaderboards

• Explore pitcher arsenals

1. What is a command delta?

   1. A command delta (CD) measures how tightly a pitcher clusters a specific pitch around an intended location.

   2. A lower CD suggests there’s a more consistent location while a higher CD suggests the location is less consistent.

2. How many pitches does it take to become reliable?

   1. Roughly 90 pitches to be considered 70% reliable and about 150 to be considered 80% reliable.

   2. I’ve set a minimum of 50 pitches to be included in the leaderboards. Anything less than 90 is relatively noisy, but I wanted to set a lower minimum to be able to look at early trends for some pitches.

3. How do I use command deltas in player analysis?

   1. In conjunction with other command specific metrics like strike rate, zone rate, heat maps, etc.

   2. Starters will generally have better CD grades than relievers.

Definition and Limitations

A CD is the sum of the standard deviations of a pitcher’s horizontal and vertical pitch locations for a given pitch type, capturing how consistently that pitch is located in two dimensions.

• Lower CD suggests tighter location dispersion and therefore better command.

• Higher CD suggests wider location dispersion and therefore worse command.

By summing horizontal and vertical variability, CD captures two-dimensional command consistency rather than isolating a single axis.

Using standard deviation here penalizes frequent misses more than rare outliers, it scales naturally on its own with volume, and I can calculate it to account for different pitch types. I chose to add together plate_x and plate_z numbers because pitchers miss locations laterally and vertically in different ways; slider and changeup locations might vary more horizontally while four-seam fastballs and curveballs may vary more vertically.

I then take the CD and assign it a z-core to account for how it compares to other pitches of the same type across the league.

CD measures pitch location repeatability, not target accuracy. It does not account for catcher intent or strategic location variation, so purposeful changes in pitch location can increase dispersion without indicating poor command.

The metric is also sample-size sensitive. Pitch types thrown infrequently may show unstable CD, making minimum pitch thresholds or stabilization methods important for reliable comparison. I calculated reliability at each target reliability below:

Substituting 0.7 for R, I found the stabilization to be roughly 90 pitches, and about 150 pitches for 0.8.

The stabilization point for fastball CDs is going to be on the higher end of this range because they have the broadest range of location diversity league-wide and therefore require a larger sample before we can be confident in an intended location. Breaking balls and offspeed pitches of the same type generally have similar shapes, so those stabilization points would on the lower end of our range.

CD weights horizontal and vertical variability equally, despite the fact that misses along each axis may differ in run-value impact depending on pitch type.

Finally, CD does not incorporate contextual factors such as count, batter handedness, or game situation, and cannot distinguish between execution error and intentional location diversity.

For these reasons, CD is best interpreted alongside pitch usage, location visuals, and complementary command metrics rather than in isolation.

Application

George Kirby is someone who is known for his command, so he’s an excellent candidate to apply CDs to. Below is how he graded in 2024 along with his heatmaps courtesy of Savant:

And 2025:

We can see he has shown well above average command of his four-seamer in both seasons, primarily locating it up in and above the zone. His splitter was thrown significantly less in 2025, so the grade is likely not as accurate as it was in 2024. His knucklecurve locations in 2025 were a full grade more consistent than in 2024, but it looked like he was intending to throw it for a strike more in 2024 at times while it was primarily a put away pitch in 2025. I’d be inclined to say he had above average command of it in both years. His slider and two-seam locations in both years flirted with 55 territory, but just missed the numerical cut-off point once each. I’d still say his command of both is above average.

An example on the other end of the spectrum is Camilo Doval. He’s someone I’d consider to be effectively wild; he has an above average 83.8% save and hold success rate along with over 100 career saves so he is trusted in high leverage for good reason, but he does have a career walk rate north of 11%. Here are his CDs and heat maps for 2024:

And 2025:

In both years, he locates his slider down in and below the zone with relative consistency compared to the rest of his arsenal, but just average consistency league wide. The CD on his cutter is graded at just below average, but it’s clear based on the heat maps that he may have more than one intended target with his cutter, so it’s likely that he has average command of the pitch. His two-seamer is graded below average in both years, and the scattershot nature of each year’s heat map supports that conclusion.

Someone with knowledge of a player’s intent when executing certain pitches could make better use of CDs to evaluate command. Coaches and analysts within organizations who have knowledge of how a team likes to approach pitching would benefit from this data, and there are various coaching and player development implications too.

A couple take aways here in the application of CDs:

• Starters generally have better command grades than relievers.

  • This make sense; we know there’s a certain level of command required to be a major league starter.

• CDs are not a single, all-encompassing metric for command.

  • There is not an all-encompassing stat for command and there likely won’t be because of how nuanced is can be to evaluate command. I was able to create a stronger assessment of both players by including heat maps (I could use other data points too, but this evaluation was done as an example on how to use CDs so I wanted to be concise).

Conclusion

It was fun and challenging to create a statistically sound way to help measure command. I’ll continue to update my leaderboards throughout the 2026 season and beyond, and I’ll continue to look for ways to refine and improve the metric as a whole. Please reach out if you have any comments, questions, or feedback.

Humans have gone to great lengths to preserve art; creating exhibits across the world to serve as centralized hubs for art consumption while also providing security measures to ensure their preservation. However, one of these art forms I listed above is currently under threat of extinction as collateral damage in the pursuit of trying to be “right” (yes, I’m putting that in quotes… more on that later) about everything. I’m talking about catcher framing.

All baseball fans have experienced the pain of seeing an umpire call a batter out on strikes in a high leverage moment. Here’s an example from recent memory:

As a Rays fan with veins bulging from my neck and pulse escalating, I can see this pitch is clearly well outside. All I can do is watch helplessly as Shea Langeliers aggressively jerks the pitch back towards the plate in desperate attempt to trick the umpire. Unfortunately, Chris Conroy fell for it and Mason Miller was temporarily let off the hook - until Jose Siri took him deep in the following at-bat (because ball never lies, right?).

But to an Athletics fan? They see Langeliers gracefully letting his glove glide just a touch beyond where he anticipates that slider will cross the plate before subtly and smoothly catching and holding the ball as he moves everything back towards DeLuca in one fluid motion. The hand speed, wrist strength, and overall presentation of this pitch was compelling enough to convince Conroy to call it a strike with an equally elegant hand motion.

Who was “right” in their perception of that pitch? The red-faced Rays fan? Or the calm and collected Athletics fan? Like with many works of art, beauty is in the eye of the beholder. But framing is not just an art; it’s also a skill.

Framing strike rate refers to the rate at which a catcher earns a called strike on pitches in the “shadow zone” that aren’t swung at. While this also includes pitches that are just inside the zone, most baseball enjoyers think of framing as the act of stealing strikes. Because words matter, I’d like to present a more positive connotation of “presenting pitches as strikes.” Anyone can check out yearly framing strike rate leaders on Savant. You can toggle between overall rates for individual pitchers and catchers, both batting and fielding teams, and even individual opposing batters. Unfortunately, this feature on Savant could become less important when the league inevitably implements a challenge system to allow batters, pitchers, and catchers to challenge balls and called strikes in the shadow zone. In theory this seems like a fine innovation, and perhaps it’s a long time coming if you remember what Joe Nathan did to Ben Zobrist back in 2013. However, the challenge system and other forms of automated ball-strike systems (ABS) are unnecessary, and they could damage one of the longest running forms of art in American history.

First of all, ABS is unnecessary. I’ll be looking at data on called strikes within the zone and outside of the zone to depict why. The data on framing strike rates has been pretty consistent year-to-year, but it only dates back to the public use of Statcast data in 2015 - which you’ll see is well past the golden age of framing we saw around the late 2000s and early 2010s. This era may have been more of a renaissance than the golden age, but we unfortunately don’t have publicly accessible data prior to 2008. Below is the percentage of pitches taken in the zone that were called strikes for each year:

Right off the bat, we can see 2008 was the wild west. In every 100 pitches taken that were in the zone, roughly 75 of them would be correctly called strikes. That’s really bad. There was clearly a push for better umpire accuracy, evident in the jumps to 81.9% by 2013, 86.2% in 2016, and now we’ve consistently seen just over 89% in each of the last five years. That’s much better, but still there are about 11 of every 100 pitches taken in the zone that are incorrectly called balls. How influential is that in a game? Last season, 113,531 pitches were takes in the zone. That equates to roughly 47 pitches per game that were takes in the zone. Of those 47, about five of them incorrectly called balls when they should have been called strikes. Are we really going to attack the beautiful art of framing over five pitches per game? Well there are also the pitches outside of the zone that are incorrectly called strikes when they should be balls, so let’s look at those year rates below:

Once again, we see that the umpire’s union finally negotiated for vision care to be included in their health insurance packages starting in 2013. They got even better as time went on, to where we see just 5.9% of pitches taken outside of the zone were called strikes last season. In 2024, there were about roughly 105 pitches taken out of the zone per game. Of those 105, about six of those are incorrectly labeled as strikes by umpires. Here’s some math for you:

5 missed calls in-zone per game + 6 missed calls out-of-zone per game = 11 missed calls per game

11 missed calls seems pretty significant, right? That’s double-digits! And some of those come in high leverage situations, and possibly leverage counts like 1-1. There were about 152 pitches taken per game last season in MLB, so this works out to 7.2% of pitches taken per game were called incorrectly. Umpires were getting the calls “right” 92.8% of the time. That’s really accurate, and frankly it’s good enough for me.

However, athletes pursue excellence in everything they do. It makes sense that players and their teams would want to see something closer to 100% accuracy - especially if there’s technology that could help make that possible. There’s nothing inherently wrong with chasing perfection or trying to win marginally more of those 11 calls than the opposing team. As a Rays fan, I absolutely understand the value of winning at the margins. However, in this quest for trying to be “right” all the time, it might hurt the value of catcher framing.

Below are the best framing strike rates by catcher since 2015 (min. 1000 framing opportunities):

On this list we have some serious star power. Buster Posey is one of the modern greats. Yasmani Grandal had a long and successful career as an above average player with a couple all-star seasons. And of course, we have the holy trinity of Patrick Bailey, Tyler Flowers, and Jeff Mathis. Nothing in the box scores would make these guys household names, but they are the social equivalent of the Beatles or Beyonce among their teams’ pitching staffs.

To check out some of the best framers since 2008 (as far back as our public Savant data goes) we have to change our parameters a bit. One way to do that is to look at called strike rate on pitches taken outside of the zone. Below are some real legends (min. 1000 pitches taken outside of the zone):

Where are my Yorvit Torrealba enjoyers at? We have some deep cuts on here: Gregg Zaun, Miguel Olivo, and of course, the legend himself, Jose Molina. Nobody can come close to Molina, and nobody ever will again when we consider how much better umpires have gotten over the years (and the fact the challenge system is on the way). At his age 35 season in 2010, Molina put up 3.1 fWAR in just 57 games while sporting just an 85 wRC+. In a full 162 game sample, Molina would’ve had 8.8 fWAR at that rate of production. For context, Shohei Ohtani has average 8.8 fWAR over the last four seasons.

The robo-umps are coming for not just our Jose Molinas, but they’re coming for our Jose Trevinos, Austin Hedgeses(?), and even our Kyle Teels. Well, Teel can hit relatively well for a catcher, so he should still be able to carve out a nice career for himself, but his value could be diminished because framing is one of his better tools. Are we comfortable with throwing our sub-80 wRC+, glove-first greats in the garbage over 11 pitches per game? The challenge system might change the way teams evaluate and develop catchers.

The most extreme outcomes in any scenario in life are usually the least likely. You’re not going to get in an accident most of the time when you drive a car, nor will you win the lottery every time you buy a ticket. Catcher framing hopefully probably isn’t going to go away completely; the new challenge system will come with a limit on the number of challenges each team gets per game. There will still be battles to win at the margins, some of those will happen in big moments and leverage counts.

At its, core, framing is an art, and art is a living thing that resides and grows wherever we choose to allow it. Are we willing to chip away at its roots, branches, and vines in our quest for some subjective form of truth? Should we push it to the brink of extinction because the third baseman on your fantasy team struck out looking on a pitch off the plate with the bases loaded? This could end up being a slippery slope and framing could eventually go away completely. Making things like this nothing but an echo from a distant past:

Data in this article comes from Savant and FanGraphs. The first video is from Savant while this last one is from YouTube (obviously).

1. Get into the playoffs

   1. 90-95 wins is usually good enough to get in

   2. The postseason is a small sample; there will be lots of random variance and things beyond a team’s control

      1. Every team that gets into the postseason has a chance to win

      2. Sometimes the hottest team wins, sometimes the best team wins, and sometimes it’s a combination of both

   3. 96.7% of teams do not win a World Series each year

      1. Your season is not necessarily a failure if you don’t win the World Series

2. Get as many playoff appearances as possible

   1. Less than a third of organizations produce roughly two-thirds of teams that have played in a World Series since the 1994 playoff expansion

   2. This disparity has been even more drastic since the 2012 playoff expansion

Section Two: Things to remember

1. You’re never as good as your best third of performances, nor are you as bad as your worst third

2. Every team will usually win 50 game and lose 50 games

3. Every team will usually have a 5 game winning streak and a 5 game losing streak

4. Every team will have underperformers and overperformers

5. Every team will win games they should’ve lost and lose games they should’ve won

6. If you can’t control it, it’s not worth worrying about

7. The schedule will be tough sometimes

   1. Inconsistent start times can be tough; players might have a hard time settling into a routine

   2. Travel days with a game on that same day are rough, especially across time zones

   3. A day off between two series at home is priceless

8. You really only have to go at least .500 on the road, and try not to get swept

   1. You’re probably going to get swept on the road at least once

9. You should go at least .600 at home, and try not to get swept

   1. You’re probably going to get swept at home at least once

10. The training staff isn’t trying to do anything wrong nor are they actively trying to get players injured

11. You can never have too much pitching depth

12. You really only need 4-5 every-day type of hitters, the rest of the lineup can be filled based on matchups and/or glove-first players

13. You really only need a 3-man rotation in the playoffs, but a 5.5 man rotation in the regular season can increase your chances of getting there

    1. The “half” starter on day 6 could be a combination of bullpen and/or bulk pitchers

    2. SPs over the last 5 years have the lowest average wOBA against when pitching on the 6th day from their previous appearance

       1. 5.5 man rotation maximizes the number of starts SPs will make on 5 days rest

    3. The 3rd pitcher in this rotation needs to be able to provide the most volume to help manage bullpen workload

       1. Higher workloads for the bullpen would come day 1, day 2, day 4, and day 5 while days 3 and 6 would be lighter

14. Limiting free bases will win you games more often than not

    1. Walks, hit-by-pitches, stolen bases, errors, etc.

15. Your best RPs should pitch in the highest leverage situations

16. Extra innings games with the ghost runner rule are almost a coin toss and have essentially eliminated the home-field advantage that we used to see in extra innings games

17. The only “must-win” games are the ones where losing results in your playoff hopes ending

18. Rookies are almost always not very good

    1. They are almost always inconsistent

    2. If you’re going to rely on rookies, you’re going to have to live with the inevitable ups and downs

19. Prospects will break your heart

    1. Many factors can prevent most “top prospects” from reaching their full potential, and most won’t even become regulars on a ML roster

20. Team chemistry in some forms can be worth roughly 3 WAR in a season (see Dr. Bezrukova’s research on “fault lines”)

    1. Continuity in a clubhouse helps chemistry

    2. Every workplace needs at least one vibes employee, even if they are underqualified

21. A mediocre team usually doesn’t become a great team at the trade deadline

    1. Adding quality depth at the margins and preparing for the offseason 40-man roster crunch are important when approaching the deadline

The purpose of this is to:

1. Establish a framework for classifying pitches in a given arsenal

2. Analyze the implications this framework could have on talent identification and development

Classifying Pitches

In this framework, there are two types of pitches along with different subcategories within those types.

• Type I: in-zone pitch

  • Primary in-zone pitch (PiP)

    • Most used pitch to generate strikes in the zone

    • Generally higher velocity relative to rest of arsenal

    • Not as much swing and miss relative to rest of arsenal

  • Secondary in-zone pitch (SiP)

    • Second most used pitch to generate strikes in the zone as a change of pace from other in-zone pitches

    • Generally higher velocity relative to rest of arsenal, possibly similar velocity or slightly less than PiP

    • Possibly more swing and miss than primary in-zone pitch, but likely less than out-of-zone pitches

  • Tertiary in-zone pitch (TiP)

    • Third most used pitch to generate strikes in the zone as a change of pace from other in-zone pitches

    • Generally higher velocity relative to rest of arsenal, possibly similar velocity or slightly less than PiP

    • Possibly more swing and miss than primary in-zone pitch, but likely less than out-of-zone pitches

• Type II: out-of-zone pitches

  • Primary out-of-zone pitch (PoP)

    • Most used pitch to get hitters to expand the zone and/or disrupt timing out of zone

    • Generally lower velocity than in-zone pitches

    • More swing and miss than in-zone pitches

  • Secondary out-of-zone pitch (SoP)

    • Second most used pitch to get hitters to expand the zone and/or disrupt timing out of zone

    • Generally lower velocity than in-zone pitches

    • More swing and miss than in-zone pitches

  • Tertiary out-of-zone pitch (ToP)

    • Third most used pitch to get hitters to expand the zone and/or disrupt timing out of zone

    • Generally lower velocity than in-zone pitches

    • More swing and miss than in-zone pitches

Application and Examples

Pitchers without a reliable in-zone pitch tend to end up as relievers. Pitchers who have a reliable in-zone pitch but are without a quality out-of-zone pitch are limited to middle of the rotation upside. Pitchers with multiple in-zone shapes should be able to successfully navigate a lineup multiple times. Out-of-zone pitches are most successful when there is an in-zone pitch that it can tunnel well with.

For the following examples, I’ll be classifying pitches in two different pitchers’ arsenals with usage%, zone%, chase%, and whiff% from 2022-2024 using Savant.

Tarik Skubal

Part of what makes Skubal so effective is his excellent feel for his SiP - as he can manipulate it to generate strikes in-zone or use it to generate chases out-of-zone. This hybrid type of pitch is an outlier and plays a significant role in making him one of the best starters in the league. He also has plus feel for his PiP. Skubal primarily uses it in-zone, but he can elicit swings out-of-zone by deploying above the zone as well as in on a hitter’s hands or away. The flexibility of these two offerings allows Skubal to disrupt timing and navigate a lineup multiple times.

Devin Williams

You’ll notice the usage rates don’t add-up to 100%, but that’s because I’m excluding the handful of pitches he has experimented with over the last few years. Williams has distinct examples of the two different types of pitches in this framework. His success primarily stems from the tunnel between his two pitches. The velocity and movement difference between the two pitches allows him to disrupt timing at a high rate - making Williams one of the most effective relievers in the game.

Implications on Talent Identification and Development

Players need some combination of in-zone and out-of-zone pitches to be successful in some capacity as they progress and face better competition. At the amateur level, a pitcher may be able to succeed with only one or two pitches. It is important to determine the capacity at which a pitcher could develop in-zone and/or out-of-zone pitches. This can be done by analyzing the underlying traits a pitcher possesses such as command and mechanics. Important questions to ask of amateur pitchers:

• How well do their land their in-zone pitch(es) in-zone? Could this be improved, and to what extent?

• What is their capacity to add one or more in-zone pitches to their arsenal, and what could that look like at the professional level?

• What is their capacity to add one or more out-of-zone pitches to their arsenal, and what could that look like at the professional level?

On the professional side of things, development and optimization could primarily be accomplished through a few avenues depending on the answers to some questions:

• Adjusting/optimizing usage rates

  • What could this look like in-game?

  • What could be used less, and in what context? What could be used more, and in what context?

• Adjusting/optimizing locations

  • Is this an optimal location for the pitch they are throwing, and where could it be thrown instead? What context would this location be effective?

• Adding new pitch(es) to their arsenal

  • Could the addition of an in-zone or out-of-zone pitch make this player more effective? What could this look like in-game?

  • How does this new pitch answer the questions above about usage and location?

Conclusion

I believe that less is more when it comes to classifying pitches. This framework allows for a more streamlined and universal type of communication among players and coaches. I’m sure that the more I continue to learn about baseball, the more I will adjust the ideology I’ve laid out here.

Listed at 6'5”, 191 pounds (but he currently appears to be heavier), Serwinowski has projection to add strength and possibly more velocity. He combines a 3/4 arm-slot and plus extension (6.7 feet) with a release height that’s low relative to his stature. He uses his body to hide the ball from the view of RHB to disrupt their timing, and he presents a challenging angle to LHB by releasing the ball in-line with the LH batters’ box.

Serwinowski features a four-seam fastball that sits in the 93-94 mph range and tops out at 97. It has plus ride (17-18 IVB), 5-7 inches of run, and roughly 2,500 rpm. His feel for the shape is relatively inconsistent despite its usage rate above 70% this past season. Baseball Savant will sometimes incorrectly label his fastballs with greater run as two-seam fastball. Regardless of the inconsistent shape, he has average to solid control of his four-seam. There’s projection to improve its command and effectiveness with a more consistent shape.

Additionally, Serwinowski has a low 80s slider that tops out at 86 mph. It has usually -1 to -2 IVB, about 7-9 inches of break, and nearly 2,600 rpm. He can control it at an average clip to keep hitters honest, but he primarily commands his slider just outside of the zone as a chase pitch (resulting in a chase rate near 40%). Surprisingly, his slider has been effective as a platoon-neutral offering. While not as drastic as his four-seam, the command and effectiveness of his slider has projection to improve with a more consistent shape.

In spite of him allowing free passes at a worse-than-average rate, Serwinowski has better control than expected, landing his pitches in-zone more than half the time. The discrepancy is due to where he commands his four-seam as he has a tendency to throw it middle or low. This has resulted in a trend; more advanced hitters can foul-off his four-seam and wait for a mistake or walk if they can lay-off the slider. His four-seam would play better near the top of the zone given its ride.

It’ll be difficult for Serwinowski to overpower hitters at higher levels with just deception and velocity while he develops as a starter. He’d benefit from adding a platoon-neutral 3rd pitch to help him present varying looks to batters as he navigates a lineup multiple times. He has the ability to locate both his pitches in-zone at an average rate so far, and he doesn’t usually struggle with count control. He needs more refinement with putting hitters away. As a result, an off-speed pitch could be a good option because he doesn’t necessarily need another pitch he can land in-zone. A cutter could also be a good mechanical fit as a moderately platoon-neutral offering to sometimes throw in-zone. His feel for his slider (and occasional accidental cutting four-seam) suggests it wouldn’t be too difficult to develop a cutter.

If he isn’t able to fully unlock his potential to be a significant contributor at the front of a rotation, Serwinowski could be a dominant high-leverage reliever with his fastball and slider combination. Being fast-tracked to the majors would be a less exciting outcome, but the potential to pitch in a rotation will always be a possibility for him as long as he’s still playing.

This simple statement holds a lot of truth. If we tell a 6th-grade student in Florida that the learning target for a math lesson is to determine the unknown decimal or fraction in an equation involving any of the four operations, relating three numbers, with the unknown in any position (MA.6.AR.2.4), why would we assess them on anything else? Assessing a learner based on the criteria outlined at the beginning of a lesson seems straightforward, yet it's not always the reality in classrooms.

Teachers must have a strong understanding of the curriculum and the standards they teach, as well as a clear sense of what mastery looks like for each standard. Thankfully, this understanding can be developed over time through professional development, intentional practice, reflection, and collaboration. Despite this, grades in many classrooms remain inaccurate.

Even when teachers are proficient in their curriculum and the standards they teach, bias and subjectivity still affect their assessments and grades. Factors like behavior or participation often influence student grades, distorting not only teachers’ perceptions of their learners' abilities but also the learners' understanding of their own capabilities.

The Foundation of an Equitable Grading Policy

Grades should reflect a learner’s performance in relation to the learning target or objective, which must align with the relevant standard. Factors like participation, conduct, and late assignments should not be part of grading, as they can distort the true academic picture. Behavior management and teaching life skills are important but should not affect academic grades. Grading learners on non-academic factors creates false impressions about their strengths and areas for growth.

Grades should be based on performance on assessments aligned with clear learning targets derived from standards. Adherence to rules and expectations should be addressed separately, but not in a way that skews a learner’s grade. Below are additional suggestions to ensure grades accurately reflect student performance on standards-aligned tasks:

• Grace for external circumstances: Students should not be penalized for late work. While teaching punctuality and organization is important, late work policies should not impact academic grading.

• Avoiding a 0% grade: A grade of 0% is not an accurate reflection of a learner's abilities and should not be part of a grading scale. A four-point scale aligns more equitably with traditional letter grades. Fs should be reserved for cases where a student consistently fails to complete assignments despite multiple interventions.

• No extra credit: Extra credit can distort grades. Assessments should focus on the criteria established at the start of the lesson. Teachers should not be expected to create extra assignments for students.

• Opportunities for resubmission: Students should be allowed to redo and resubmit assignments when appropriate within a given timeframe (but not across grading periods). This encourages mastery, especially when assessments are valid and reliable.

• Individual assessment: Grades should reflect individual performance. Group or partner work, while valuable for collaboration, should not be factored into grading, as accurate assessments require students to demonstrate their skills independently.

Introduction

Hitting is all about timing. Taking the inverse of that, we would see that the primary goal of a pitcher is to disrupt timing. Of course pitchers have the ability within their physical limitations to influence the shapes of their pitches, but those may have less of an impact on pitcher effectiveness than initially thought. Release characteristics matter far more than many “Stuff” models can even begin to quantify. Why? Because they struggle to fully account for how a batter perceives a pitch and subsequently tries to get on-time for it.

This is what led me to identify that pitch shapes and release characteristics work together to create the two of the three main areas of my focus. The first of which is the angle the pitcher creates when delivering the ball. VAA and HAA are nothing new, and even HRA and VRA are relatively common in most spheres. I believe these would be one of the optimal components for pitcher development and evaluation to focus on. The shape is less important to the batter than the angle it creates both out of the hand and in relation to the zone.

The second area of my focus in relation to shapes and release characteristics is timing. Pitchers can affect timing a number of ways in regards to their shapes and delivery. This includes things from velocity to occlusion and even having a hitch in their delivery.

The third area of my focus is somewhat separate from the first two: location. Location is comprised of control and command. While they are two different things, I’m including them in the same area because they are things a pitcher can heavily influence, and they are dependent on the purpose of a pitch. There isn’t a single “good” way to control or command all pitches across a given arsenal. The framework here is highly individualized, so there are only optimal rates for different pitches in an arsenal depending on the purpose of a given pitch and its context.

Lastly, I’ve made a couple adjustments to some labels to reflect the changes in my initial theory. I’ve changed the “FP” (fast pitch) label to “PP” for Primary Pitch. This refers to the most utilized in-zone pitch a pitcher throws regardless of velocity or shape.

The “SiP” designation simply refers to Secondary In-Zone Pitch now without consideration for movement or velocity. Similarly, “SoP” no longer takes into consideration movement or velocity. Both are primarily focused on usage, control, and command.

I have created a visual to help conceptualize the relationships between angles, timing, and control and command.

Analysis in this framework

Before we take a look at a couple of examples, there are three questions to ask about a pitcher when analyzing them under these new parameters:

1. What difficult angles can a pitcher create?

2. How does a pitcher disrupt timing?

3. Can a pitcher locate their pitches in optimal spots?

And these are the three categorizations of pitches with different designations:

• aPP or gPP: an arm-side or glove-side primary pitch utilized in-zone

• aSiP or gSiP: an arm-side or glove-side moving pitch utilized in-zone less than the PP

• aSoP or gSoP: an arm-side or glove-side moving pitch utilized out of the zone

Yu Darvish

A good pitcher to test any pitching theory or model on is Yu Darvish because he throws so many distinct shapes from what’s really four different pitches.

Primary Pitch (Slider and Sweeper)

Darvish uses a gPP that he has advanced feel for to attack the zone at an above average 60.4%. He almost triple it’s breaking movement at the expense of 3-4mph when needed, so overall hitters need to be prepared to see it in the 82-86mph range. His lower-than-average release height coupled with solid extension and a longer-than-average horizontal release creates difficult angles even when his locations are sub-optimal. Both gPP variations play well in-zone primarily due to the horizontal angles he creates and how they pair with his aSiP shapes.

Part of what makes Darvish so unique is that his gPP shape operates between his aSiP and gSoP shapes both in terms of movement and velocity. Most pitches have a PP shape that requires SiP shapes to blend and connect to their SoP shapes. PPs with less total movement are generally easier to command, and those tend to be the higher velocity shapes. Darvish is an outlier because his gPP variations have quite a bit of movement. This speaks to his advanced feel for control and command of that pitch.

Secondary In-Zone Pitch (Two-Seam, Four-Seam, and Cutter)

Darvish deploys a trio of aSiP shapes ranging from 91-94mph at an above average rate of 57.8%. The aSiP variation with the most vertical movement plays best at the top of the zone and above because of the uncomfortable vertical angle it creates coming from a relatively low release height. Darvish has comfortably plus feel for his more runny aSiP shape and uses it primarily to keep hitters from trying to time up his more vertical aSiP. His cutting (but on paper neutral) aSiP variation blends well with his gPP shapes - especially the tighter one. This combination also does well to disrupt timing because of the three velocity bands.

Secondary Out-of-Zone Pitches (Curveball, Knuckle Curve, and Splitter)

We’ll begin with his gSoP shape and its variations. Ranging from 73-78mph, Darvish lands it in-zone at a modest 45.2%. This is much lower than his gPP and aSiP shapes, but that’s understandable because it’s not a pitch he uses to get strikes. SoP shapes are primarily used to get hitters to chase and Darvish is no exception here. There really isn’t much of a significant or functional difference between the two variations of his gSoP aside from the velocity, but they pair well with his gPP (especially the lower velocity variation) to generate chases. So despite their average and somewhat redundant characteristics in relation to his arsenal, they still serve a purpose.

Darvish’s aSoP is quite interesting because it’s his only pitch that doesn’t work off of his gPP variations. It pairs best with the more vertical aSiP shape and somewhat with the running variation. Its primary function is to keep hitters from sitting on the more vertical aSiP and is most effective when located below the zone. This makes sense considering its 43.8% in-zone rate.

Garrett Crochet

Crochet serves as a more practical pitcher to analyze in this framework because he has a more standard arsenal.

Primary Pitch (4-Seam Fastball)

Crochet’s primary in-zone weapon is an aPP with pretty standard movement and a respectable 59.3% in-zone rate. It stands out for it’s plus-plus velocity and near elite extension to create a challenging angle for hitters. Crochet leverages most of his 6’6 frame into creating extension and a horizontal release point which further add to his ability to disrupt timing. Hitters are not used to seeing a LHP with that kind of delivery and velocity. The aPP is a relatively platoon-neutral weapon that helps him leverage count control against RHB, but the emergence of his gSiP as a viable in-zone weapon has propelled him to becoming one of the best pitchers in the game today.

Secondary In-Zone Pitch (Cutter)

This gSiP has decent velocity and vertical separation from his aPP, and shapes with that kind of velocity and tight break tend to play platoon neutral. Being able to land it in-zone over 50% of the time is what has allowed him to effectively navigate lineups multiple times even when they are loaded with RHB. This shape also blends well with his gSoP which further allows for better sequencing as a starter versus when he used to be a limited to a relief role. Pitchers like earlier versions of Crochet who try to blend a PP with a SoP tend to struggle with navigating a lineup multiple times because hitters can more easily anticipate which pitch they will see in a given count.

Secondary Out-of-Zone Pitches (Sweeper and Changeup)

His gSoP is his primary chase pitch as he lands it in-zone 41.3% of the time. It has optimal velocity separation from his gSiP, and has a unique amount of break considering its above average velocity on that sort of shape. His ability to consistently locate it down and glove side allows him to sneak it under the bats of RHB while LHB simply have almost no chance of making contact, let alone putting it in play.

Crochet does have an aSoP, but it is rarely thrown so it’s not something most hitters need to worry about. However, it does blend surprisingly well with his aPP. So when he does sprinkle it in, it is rarely a pitch hitters can do damage on.

Practical Applications of this Framework

A simplified language applicable to all pitchers regardless of the shapes they do or don’t throw allows for a more consistent, uniform type of analysis. It also allows pitchers to be evaluated for what they are with respect to their individuality. Everyone can throw what they want in different ways to achieve the same common purpose. It also allows us to look at the whole pitcher; how they go about disrupting timing and creating difficult angles without favoring different archetypes or guys who pop on “Stuff” models. It’s clearly more complex to analyze pitchers and evaluate performance in this paradigm, but it’s more more holistic and comprehensive so it will yield more thorough and thoughtful results.

Common language is also beneficial in coaching and development because it allows for players to work towards a common goal to build an arsenal that fits the pitcher rather than chasing shapes or doing something that doesn’t make sense for their body. If every pitcher had a 95mph fastball with very little run and 20 inches if IVB, hitters would get more comfortable with those pitches over time and they would be less effective. It’s important to preserve individuality in pitching out of respect for the uniqueness of all pitchers as well as the practical, in-game applications that uniqueness can lead to. Common language can clarify and simplify concepts before they lead to misunderstanding and inconsistencies in messaging.

This common language also inherently values control and command - two things a large faction of modern pitching development and analysis has gotten away from. As stated before, a more holistic view is more complicated and messy, but it is more thorough and thoughtful than simply trying to put labels and numbers on things. That’s not to say numbers and metrics aren’t useful in some capacity, but there’s so much nuance that gets lost when we disregard the context.

I think it starts with the body’s natural inclination towards being able to handle the volume of a larger workload like starters and bulk pitchers, or being better suited for shorter outings like relief pitchers.

For larger volume pitchers, they typically need three distinct pitch shapes to be able to effectively navigate a batting order at least two times. Most arsenals look like this:

• Highest speed, low/medium movement pitch (the fast pitch - FP)

• Lower speed, low/medium movement pitch (the secondary in-zone pitch - SiP)

• Lower speed, medium/high movement pitch (the secondary out-of-zone pitch - SoP)

A few things to note here: Smaller volume pitchers can typically survive with just two of those types of pitches. Also, both speed and movement are relative to the individual and the rest of their arsenal. High speed fastballs tend to allow for more wiggle room and “arsenal protection” regardless of shape, but velocity isn’t everything.

Pitches with a lower amount of movement are easier to command, so those tend to be higher usage pitches because throwing in the zone is much more valuable than throwing outside of the zone. However, its lack of movement tends to make it easier for hitters to make contact with.

FPs that are platoon neutral tend to be more successful for high volume pitchers. This is their primary in-zone pitch that they can command well to hitters on both sides of the plate. It is best used to generate called strikes and foul balls, and to generate less than ideal contact if the ball is put in play.

An FP with more velocity allows a pitcher to be less fine with their command because it is usually difficult to get an ideal swing on a pitch like that. This also “protects” the other pitches they throw; if a hitter has to be ready for high velocity, they’re less likely to sit on the SiP or SoP.

The ideal FP shape depends on the pitcher and what the rest of their arsenal looks like. I think it is best to lean into the strengths of an athlete rather than trying to make their bodies do something they’re uncomfortable with, and a pitcher could have multiple FP shapes. Some pitchers have an inclination to throw a cutting or rising FP (gFP: glove-side moving FP) — these tend to be platoon neutral or even better against opposite handed batters. Running or sinking FPs (aFP: arm-side moving FP) tend to be better against same handed batters.

SiPs would be a little less easier to command than an FP, but still able to get a strike when needed. They also might get more swing and miss than FPs. Glove-side moving SiPs (gSiP) work best against same handed batters while arm-side moving SiPs (aSiP) work best against opposite handed batters. Again, there is no ideal shape here and a pitcher could have more than one SiP shape.

SoPs are the most difficult to command, but tend to get the most swing and miss. They are often difficult to get a good swing on due to their movement and usually don’t end up in the zone. There is no one ideal shape other than the one that works best for the individual. Glove-side moving SoPs (gSoP) work best against same handed batters while arm-side moving SoPs (aSoP) work best against opposite handed batters. However, gSoPs can still be highly effective versus opposite handed batters and aSoPs versus same handed batters when their primary movement pattern involves more sink than run or sweep.

Constructing an Arsenal

The athlete should always lean into what they do best. Their best pitch is the one that should be built around. It helps if their best pitch has “outlier” traits different from what other pitchers do with similar pitches — comparing FPs with FPs, SiPs with SiPs, and SoPs with SoPs.

There’s no one way to build an arsenal, but for larger volume pitchers, having an FP and at least one SiP that move in opposite directions of each other is important. Without at least one platoon-neutral in-zone pitch, a pitcher would almost certainly struggle in a large volume role where they would be tasked with getting opposite handed batters out with relatively high frequency. Same side moving FPs and SiPs look more like the set-up of a low volume relief pitcher because they would have a clear weakness.

When an SoP is a pitcher’s best offering, it presents a tricky situation because the goal would be to create as many situations as possible to throw the SoP. This is where the primary function of a SiP would be to bridge the gap between the FP and SoP. An aSiP would pair best with an aSoP, and a gSiP with a gSoP. The ideal FP shape in the case of an aSoP/aSiP arsenal would be a gFP to help neutralize while an aFP would be best for a gSoP/gSiP set-up.

The main take away here is that the athletes strengths should build around their best pitch to the best of their abilities. Now, we’ll take a look at some real examples of this new way to classify pitches…

Jared Jones

Let’s start with one of the hottest names in baseball right now:

Jones’s arsenal consists of:

• An aFP - high velo with good rise

• A gSiP - his best pitch, tight late break with gyro movement which helps it play well versus both RHB and LHB (this is his outlier pitch)

• An aSiP - his offspeed pitch primarily used against opposite handed batters, not that great of an in-zone offering, but his gSiP helps fill that void in-zone

• A gSoP - more of a sink/drop profile, making it acceptable against both sides of the plate; shape tunnels well with gSiP, so he’s able to accentuate his best pitch even more

His aFP offers plenty of protection for his gSiP due to it’s velocity. Jones is an excellent example of someone who constructed their arsenal based on their strengths and built around their best pitch.

Adam Oller

Here we have Adam Oller’s pitch shapes from 2022. I wanted to choose a good example of poor arsenal construction because we have a bit of survivorship bias in the major leagues. The vast majority of the guys who make it to the major leagues are good at what they do. Thankfully these are old shapes and Oller made some good adjustments in 2023 (more on that later), so I don’t feel as bad critiquing him here.

The first thing you’ll notice is a lack of distinct shapes. This is typically a sign of an ineffective pitcher, especially when it happens unintentionally. The main action that needs to be taken when this is the case is the pitcher needs to clean-up their arsenal and throw more distinct shapes — which he did in 2023:

There still isn’t enough optimal separation between his FP and SiPs, but we at least have something we can work with here and analyze:

• An aFP - running, but not much rise or sink, good enough velo at 93mph

• An aSiP - shape is inconsistent in a similar way to Jones, mostly a “show” pitch for opposite handed batters

• A gSiP - tight, consistent shape, good strike pitch to bridge the gap to his best pitch from his aFP

• A gSiP - sweeping and sinking movement, by far his best pitch, not too platoon neutral due to its sweep

While Oller did take actionable steps in the right direction, there are still a few more things I’d have him do. Step one would be to adjust his aFP to add more run; it looks like his body’s natural inclination is to throw more of a running, sinking aFP than a rising one, so it would be important to play into that. This would allow for more of a movement gap between his aFP and the gSiP he has a good feel for. It helps that he flashed an aFP shape with more run and sink in 2022. The gSiP serves it’s purpose as a bridge to the gSoP so I wouldn’t change anything else there. Lastly, I’d scrap the aSiP; it would be a bit redundant with the emergence of a stronger aFP shape.

Edwin Diaz

I wanted to throw in a quality lower volume relief pitcher here, so look no further than one of the most dominant arms in the game.

Diaz’s arsenal consists of:

• An aFP - his best pitch, strong running movement and high velo

• A gSiP - very minimal cut/break, so it plays more platoon neutral than most gSiPs, able to land it for a strike

Diaz is effective with two pitches because he rarely faces more than a handful of batters in each appearance. It’s a simple yet effective arsenal that works well for the athlete’s body and his role. Here’s one more short volume example that I’d consider the inverse of Diaz…

Calvin Faucher

Faucher’s arsenal from 2022-2023 was less than ideal. Here we have a blending of gSiPs that spill into his gSoP. The gSoP is his best pitch; it’s a late breaking and steep hammer with relatively high velo. Faucher struggled to get to his gSoP due to a lack of a clear bridge gSiP that he could also land for strikes. He started to make the adjustment in 2023, but went all the way in 2024 and the results so far have been excellent (although is has been in a small sample). Here’s his new arsenal:

• An aFP - a rising and running offering with good velocity, plays well against opposite handed batters

• A gSiP - mainly a bridge pitch, but can be thrown in-zone enough and plays well against same handed batters

• His stellar gSoP - his best pitch the more drop/sink profile allows it to play a bit more platoon neutral than other gSoPs

Faucher didn’t start throwing his newest gSiP shape at a meaningful amount until 4/15/2023. His FIP in 27.2 ML innings from his debut until 4/14/2023 was 5.59. Since he started throwing that relatively new gSiP, his FIP has been 3.36 in 28 IP. Both samples are relatively small in the grand scheme of things, but the trends and improvements are significant and worth keeping an eye on going forward.

Conclusion

Something that would be worth writing about but impossible to quantify in a meaningful way would be pitch usage and sequencing. Even though they’re largely random, they also play a role here, but again it comes down to the individual athlete and what they’re most comfortable with. I like the idea of simple and fewer labels when it comes to forms of art. Talented artists don’t really need more coaching and more labels, they need guidance occasionally and feedback to help them be the best versions of themselves. However, the most important thing might be for us to get out of the way and let them be artists.

*All graphs courtesy of Baseball Savant

Subscribe now

Share

Subscribe now

Share

If you’re interested in the philosophical things that are also boiling in my brain, I recommend reading the link I shared at the beginning of this piece.

Subscribe now

Share

Proactive framing

As I said in my first write-up about catcher symbiosis, I believe that framing might be the single largest component when determining a catcher’s overall value. There are usually a couple hundred pitches thrown in an average MLB game. Not all of them are thrown around the edges of the zone where framing occurs, but there are far more opportunities to frame a pitch for a strike than any other action a catcher will take in a game — including the time they spend in the batter’s box. This is not to say a catcher hitter or their blocking and throwing are not important, but framing might be the most important.

Blocking and throwing are tricky because they each require very specific circumstances for them to matter in the context of a game. Blocking only matters with runners on base, and the subjectiveness of official scorer ruling can sometimes make it difficult to determine whether a pitch was wild or a passed ball. The quality and accuracy of the pitch thrown also plays a significant role in blocking. Catcher throwing also only matters with runners on base, and stolen bases are sometimes more dependent on the pitcher than the catcher. Both “skills” we attribute to catchers are heavily influenced by pitchers whereas framing is much more catcher dependent.

Framing also gives us some insight as to how much a catcher is “on the same page” as the pitcher throwing the ball to them. Being proactive in framing rather than reactive helps catchers steal even more strikes.

Additional findings

I compiled data on framing strike% and K/BB ratio from 2022-2023. I wanted to explore the effects a catcher has on a pitching staff (even though some of the pitchers change year to year), so I limited the data on a few parameters:

• Only catchers that spent at least part of both the 2022 and 2023 season with one team were included

• Catchers must have received at least 250 pitches in each of those two seasons

The only team that did not have at least one catcher that met those qualifications was the Cleveland Guardians, so that’s why you won’t see any of their catchers on this really interesting table below:

A few things stand out to me here:

• Rene Pinto not only has the highest framing strike% difference relative to his team and the best overall K/BB ratio, but he also is tied for the 3rd highest framing strike% overall and has the 2nd best K/BB ratio difference relative to his team

• While Keibert Ruiz wasn’t a very good framer, he was better than the other catchers on his team and produced a much better K/BB ratio relative to those teammates — the best K/BB difference among this group

• Joey Bart was an above average framer, but did not get very good K/BB results

• Austin Nola and Martin Maldonado were not highly rated in their pitch framing, but still managed to help their pitchers maintain high K/BB ratios

Possible limitations

As I mentioned before, these things are only a piece of the puzzle. There is still a lot we don’t have public access to when it comes to how team evaluate their catchers and how well those catchers work with the pitchers in their organizations. Pitcher quality, sample size, and maybe even umpires could affect the data here. I am confident that there is some value in what I’ve put together, especially with framing strike% and K/BB ratio relative to how other catchers on the same team performed, and I wish I would’ve thought of including that in my initial piece. I think tempo and time spent in an organization are still pieces of the puzzle. I’m looking forward to continuing to explore how we can measure catcher symbiosis.

In addition to this unicorn of a knuckle curve, Megill’s fastball sits 99mph with above average carry. He also has the size and physicality (listed as 6’8, 250 on Fangraphs) to handle a starter’s workload. Why are his talents being underutilized in the Brewer’s bullpen? Teams tend to push pitchers into a relief role for a variety of reasons.

Command

Megill owns an 8.1 BB% at the major league level so far for his career. Among all pitchers with 100 IP since Megill debuted in 2021, this ranks in the 53rd percentile - pretty much average. An average walk rate is good enough to be a starter in the majors; Luis Castillo, German Marquez, Framber Valdez, and Spencer Strider all have similar walk rates during that same time frame and there isn’t much debate about any of those pitchers having good enough command to be starting games.

In addition to his average walk rate, Megill has maintained average command grades from PitchingBot and Stuff+:

We have two independent command metrics telling us that Megill has average command in addition to his average walk rate. We can safely say that he throws enough strikes to not hurt himself with walks or inefficient pitch counts, so his strike throwing ability should not prevent him from a spot in a major league rotation.

Throwing enough strikes doesn’t always indicate the quality of a pitcher’s command. Maybe Megill might not be walking too many batters, but his 44.8% hardhit rate (3rd percentile, min 100 IP since 2021) suggests he isn’t necessarily throwing quality strikes.

Fastballs with high IVB like Megill’s typically play well at the top of the strike zone and above it. The average fastball location height for Megill over the last three seasons has been just 2.69 feet (27th percentile). We can see his average locations last season on the heatmap below:

This heatmap is better than the past where his fastball locations were typically more in the middle of the zone, but it’s still not ideal. His optimal locations for the fastball given it’s shape would be in the upper third and above the zone. Teaching command is tricky, but sometimes it can be about finding the right external cues. For some, this looks like a universal target where a catcher sets up in the same spot every time and pitchers let the natural movement on their pitches do the work. For others, they need to have a target set up a specific way. I know from my own experience teaching and coaching (soccer, but coaching youth teams is pretty similar across many sports) that external cues are much more effective than internal cues. Optimizing Megill’s fastball locations is almost certainly something the Brewers have attempted, but I’m confident there’s a cue out there that will help him make the necessary adjustments. His base level of command is strong enough and he doesn’t have to be perfect every time because he throws another really good pitch with the potential to add a viable third pitch.

Arsenal

Another popular reason to put a pitcher in a bullpen instead of a rotation is that they don’t throw enough different pitches to effectively navigate a lineup multiple times. At surface level, we can see Megill threw just two pitches last season — his fastball and breaking ball. He has experimented with a sinker and slider in the past; the former looks like a misread on some of his fastball shapes by Statcast, and the latter was something he scrapped last season in his relief role (more on that later).

Throwing just two pitches makes it easier for hitters to sit on one pitch. Combine that with the fact that Megill has been able to locate his curveball in the zone just 39.6% of the time over the last three seasons (26th percentile among curveball zone rates, min 250 thrown), and we can generate a Sparknotes scouting report: If it’s in the zone, it’s going to be the fastball. To make things even easier on opposing hitters, Megill has thrown his fastball almost 64% of the time over the last three years. Using the fastball 60% along with his limited arsenal and suboptimal fastball locations is not an ideal usage rate. Yes, it’s a good fastball, but as we said earlier, its locations typically aren’t high enough and hitters can pretty much assume anything in the zone is going to be the fastball.

In addition to the external cue mentioned in the previous section, the next adjustment I’d have Megill make is with his arsenal by bringing back his slider and adjusting the shape to flatten it out. The slider Megill threw in the past did not have enough movement separation from his curveball:

Achieving more of a bullet slider/cutter shape would allow for more optimal separation, making it a true “bridge” pitch between the fastball and curve. This new slider/cutter shape doesn’t have to be anything overpowering or unique; Megill just needs another pitch he can land for strikes if we want him in a starting role. Tyler Glasnow is a good comparison here. Similar fastball shapes and locations as Megill, and they both have nasty curveballs that are rarely in the zone. Glasnow implementing a cutter/slider shape helped him reach a new level as a starter. He was able to go deeper into games because he was more efficient with his strike throwing; he could now throw two pitches in the zone except for one. Additionally, Glasnow was no longer a two-pitch pitcher and could attack hitters differently more than two times through a lineup — which also allowed him to go deeper into games and have more success as a starter. Like Glasnow, Megill’s third pitch wouldn’t even have to be that great, just throwing it would make his better pitches more effective. He would also only need to throw his new third pitch at least 20-25% of the time. While it might take some time to get a feel for sequencing his revamped arsenal, the rest of the time he can split between his fastball and curve.

Let’s say Megill does find an external cue that works for him and he picks-up an okay cutter/slider shape that he can throw enough strikes with. What else could hold him back from being a starter?

Platoon Neutrality

Some pitchers don’t last as starters because they are unable to effectively handle opposite handed batters. The verticality of Megill’s arsenal in his high carry, low run fastball and almost 12-6 movement on his curveball are naturally fairly platoon neutral.

During Megill’s career, right-handed batters have slugged .528 (21st percentile min 50 PA 2021-2023) and .500 (12th percentile) against his fastball and curve respectively. Left-handed batters slugged just .385 (67th percentile) and .229 (83rd percentile). against his fastball and curve. Despite the relatively small sample size of his major league career, we can already see that Megill is showing signs of being able to handle LHB which makes sense given his pitch shapes. The issue with his subpar performance against RHB has nothing to do with his shapes or platoon neutrality; it’s mainly about hitters knowing that any pitch in the zone is almost always going to be a fastball and most of his fastballs are not located high enough to be effective.

The addition of a cutter/slider shape discussed earlier would only help him handle left-handed hitters. The less spin efficiency on it, the better so it doesn’t break too much. Glasnow again is a good example here; his bridge pitch had less than 20% of active spin over the past two seasons. Megill’s slider back in 2022 wasn’t too far off at 34%, but even that may have been too much for him. An optimal shape for Megill would have more gyro spin and very little side spin. It’s easier said than done obviously, but being in an environment that is more focused on the process than product where he can tinker with different grips under the guidance of a pitching coach who understands best practice in coaching (and possibly pedagogy) could help Megill unlock another level as a starter.

To sum everything up, Megill could be an interesting (and possibly great) starting pitcher because:

• He has the size and physical attributes to be a starting pitcher

• He throws enough strikes

• He could throw his fastball higher in the zone, and above it sometimes

• He could pick-up his cutter/slider again and tweak the shape until it has enough movement separation from the curve

• He could throw that new cutter/slider in the zone, and use it roughly 20-25% of the time

• He is already platoon neutral with his current shapes, and the addition of a cutter/slider in the similar mold mentioned earlier would only add to that neutrality

Subscribe now

Share

Which player was more valuable in 2014 on an fWAR per 500 PA scale: Mike Trout or Rene Rivera? If you said Rivera, you’d be correct; his 7.93 fWAR/500 PA actually led all position players that year (min. 250 PA). He played in just 103 games to accumulate a fantastic 5.22 fWAR, and he maintained a respectable-although-not-too-impressive 113 wRC+. So how did Rivera manage to have one of the most valuable seasons of any position player over the last 15 years despite being just a decent hitter? He was excellent defensively - especially in his ability to steal strikes via framing.

Subscribe now

Share

Catcher framing is something that we have been able to measure and quantify over the last decade thanks to advances in baseball technology. But even before that, teams have had their own internal methods of evaluating catcher framing. Other advanced defensive metrics related to catching have made their way into the public sphere. These measurements give us some insight into just how valuable catchers can be to their teams, and the vast majority of a catcher’s job involves their defensive duties.

Growing-up a Rays fan, I never expected much offensively from the catchers on the roster. The rationale for the team giving valuable playing time to guys like Rivera, Jose Molina, and Gregg Zaun is pretty clear in hindsight. Rays fans are well aware of Ben Zobrist’s 2009 season when he amassed a Trout-esque 8.70 fWAR (7.26 fWAR/500 PA) thanks to his versatility, defense, and a whopping 152 wRC+. His fWAR/500 PA led all position players that year, just beating a pair of catchers - Joe Mauer (6.93) and Gregg Zaun (6.16). But my personal World Series was 2012 Jose Molina. Below is a list of the top ten position players by fWAR/500 PA for the 2012 season (min 250 PA).

At first glance, you might notice six catchers here. You might also notice Jose Molina’s abysmal 81 wRC+ and Ryan Hanigan’s (who the Rays grabbed a few seasons later in 2014) 88. Catcher defense, especially framing, is extremely valuable. Four of the top ten seasons by fWAR/500 PA since 2008 came from catchers.

Proactive Framing

I believe that framing might be the single largest component when determining a catcher’s overall value. There are usually a couple hundred pitches thrown in an average MLB game. Not all of them are thrown around the edges of the zone where framing occurs, but there are far more opportunities to frame a pitch for a strike than any other action a catcher will take in a game. This is not to say blocking and throwing are not important, but framing might be the most important.

Framing also gives us some insight as to how much a catcher is “on the same page” as the pitcher throwing the ball to them. Being proactive in framing rather than reactive helps catchers steal even more strikes. A few factors go into what I would consider proactive framing.

Tempo

In this context, tempo would refer to the speed and rhythm in which a catcher and pitcher execute their game plan. This doesn’t necessarily mean they need to work fast, but a catcher and pitcher who are in-sync would work relatively quickly, at a speed where the pitcher feels most comfortable, and that speed would remain relatively consistent from pitch to pitch. Catchers who work well with their staff would be able to anticipate the pitch and location at which the pitcher wants to throw it because both players would be in-tuned with the team pitching philosophy in addition to the catcher knowing that pitcher like the back of his hand. A catcher being especially aware of how to attack a specific hitter while working with a specific pitcher would be even more adept at maintaining a good tempo.

While we do have public access via Baseball Savant to pitcher tempo, we currently do not have any data on catcher tempo unfortunately. This is something that would help provide insight into catcher “symbiosis” (I’m assigning my own denotation here; symbiosis refers to how well a catcher works with their staff). I believe the catchers with relatively quick tempos would predominantly be the ones who work the best with their pitchers and probably have higher framing strike rates as a result.

K/BB Rate

Strikeout to walk rate is absolutely more of a pitcher statistic, but I believe it still has some value in highlighting catcher symbiosis. Pitchers who feel more comfortable with their catchers can attack the strike zone, and the catchers are well aware of how their pitchers want to attack the zone. Catchers can help their pitchers avoid walks by framing well and adhering to team pitching philosophy as well as individual plans for attacking hitters - also leading to more strikeouts. K/BB can be influenced by a catcher’s framing ability, but it can also be influenced by how well a catcher calls pitches. Game calling and framing strike rate combine to show catcher symbiosis, and K/BB rate gives us some imperfect insight as to how well a catcher works with their pitchers.

Time Spent in an Organization

Like K/BB rate, this is not the ultimate way to measure catcher symbiosis, but I still believe it’s a piece of the puzzle. The concept is simple, a catcher who has spent more time with their organization would most likely be more skilled at executing team pitching philosophy. This would not just be evident in framing strike rate, but K/BB as well - and even tempo if we had public access to it. A catcher who is skilled at framing could still find success bouncing around from team to team, but looking at all the factors mentioned above from a more holistic view would help paint a picture as to which catchers are the best at working with their pitchers

Putting the Puzzle Together

While we don’t have public access to all the necessary data and their are things individual organizations do to evaluate how well catchers work with pitchers, we can still work with what we have to do some analysis. Below is a graph comparing framing strike rate (x axis), K/BB rate (y axis), and the time a catcher spent in their most recent organization (bubble size). I created this using data from Baseball Reference and Baseball Savant. For a catcher to be qualified in this sample, they needed to receive at least 250 pitches in both 2022 and 2023 for the same team. I chose these parameters in order to get a framing strike rate in a meaningful sample, and I wanted to only include catchers who had more than one season of continuity working with mostly the same pitching staff to help control for catchers who might be trying to learn a new staff or pitching philosophy.

I wouldn’t say it’s statistically significant, but most of the catchers with more time spent in their current organizations as of the 2023 season were better than average at framing and had better than average K/BB rates with the exception of a few outliers.

We can also see there is some sort of relationship between both K/BB and time spent in an organization, and framing strike rate and time spent in an organization.

Next Steps, ABS, and More Questions

It’s apparent that there’s still a lot of work to do in measuring catcher symbiosis. It might be years before we have enough public data to determine which catchers are the best at working with their pitchers, but I believe we can start putting the puzzle together now. I plan to continue looking for ways to measure catcher symbiosis, and I will update the data I have here following the conclusion of the 2024 season. I will also look into both K/BB rate and framing strike rate among catchers that have changed organizations between seasons and compare that data with catchers who stayed with their organizations.

Another side project I’m looking forward to is how the league implementing automated balls and strikes (ABS; aka robot umpires) could impact catcher value. Framing (which I still believe is the carrying tool for catcher value) would become less important, but I don’t think ABS would make catcher symbiosis less important. In fact, implementing ABS might help accentuate a catcher’s ability to call a game and work with their staff by controlling for catchers who are good at framing no matter who they are working with.

Overall, analyzing catcher symbiosis has created more questions than answers. It’s exciting to think about thie next frontier in determining catcher value through this lens, and I personally enjoy exercises like this where looking at the data - both anecdotally and the things we can quantify - forces us to try to piece things together rather than providing a direct answer.

Subscribe now

Share

I will take a second to pat myself on the back for highlighting Josh Lowe as a second year player who could take a step forward when I wrote about a possible Josh Lowe breakout back in February of 2023 for DRaysBay. It turns out I was pretty spot on, but I wanted to revisit what I noticed about rookies in the opening paragraph:

“From 2015 through 2022, rookies slashed an uninspiring .238/.303/.387 with an 86 wRC+ and 1.03 fWAR/600 PA. Let’s compare that to the non-rookies during that same time: .252/.322/.420 with a 100 wRC+ and 2.00 fWAR/600 PA”

Looking back at the 2023 season, we once again saw that rookies were not very good. Rookie position players struggled to produce at the plate while rookie pitchers saw fewer meaningful innings than non-rookies.

With the pitching side of things, overall workloads and usage in higher leverage situations certainly contributed to the significant difference in fWAR. However, it’s still apparent that trying to rely on rookies is most likely going to lead to less-than-ideal results.

Taking the time to let rookies and young players grow into various roles (usually with the safety net of a few veteran players - even if those veterans aren’t very good themselves) allows for a more safe integration of young, unproven talent.

The speed at which various organizations promote and push their prospects into larger roles is going to be interesting to watch in 2024. The unending dance of weighing risk versus opportunity cost has quietly become one of my favorite things about the game. There is no magic number of games or average amount of service time needed on a given roster for a team to be successful, but experience does play a factor in raising the floor of every team.

Share

Subscribe now

Share

Subscribe now

At the conclusion of the 2023 season, the league average payroll of an MLB team was $165 million. The current projected average payroll heading into Opening Day 2024 is $156 million. While there is still time for teams to spend in free agency and even negotiate contract extensions to get those numbers up, spending is still down roughly 5.3% year-over-year. Most fans are well aware of which teams consistently spend at the extremes of each end of the payroll spectrum. Teams like the Dodgers, Mets, Phillies, and Yankees have consistently found themselves spending the most while the Pirates, Rays, Athletics, and Orioles have spent the least on payroll in recent years. However, we can see some interesting trends if we look at team spending in heading into 2024 in relation to what individual organizations spent the previous season.

The table below shows projected team payroll for 2024 relative to what they spent in 2023. The Orioles, Royals, Dodgers, Rays, and Reds have shown the greatest willingness to invest in their 2024 rosters while the Padres, Athletics, Angels, Twins, and Giants have cut costs the most. It’s important to note that these numbers include projected arbitration salaries. We can see a team like the Mets, who are projected to have the second highest payroll in 2024, actually slashed spending by more than 3 times the league average. The overall trend of cutting payroll costs could be an attempt by teams to extract more value from their rosters.

The next table shows how much each team is spending on a projected per fWAR basis. The Athletics, despite having the 4th lowest projected fWAR total, are actually getting the most value from their roster by spending just $1.59 million per projected fWAR. They’re followed by the Pirates, Orioles, and Rays; all teams expected to accumulate more fWAR than the league average team while also getting a good value for that fWAR due to their lower payrolls. The Rockies are the only team spending more than double the league average for projected fWAR on their roster.

It will be interesting to see how team payroll spending progresses through Spring Training as teams continue to add free agents and even extend players currently on their rosters. Despite the current trend, I do expect the league average payroll to grow and possibly even eclipse the average from the previous season. The teams who are showing a willingness to invest in themselves early on deserve some recognition, and I’ll be watching how much certain teams are willing to spend on a projected per fWAR basis going forward.

Share Ben Whitelaw Baseball

The more I’ve asked this question, the more I’ve come to understand that there is too much to learn. Being a lifelong learner is an endless journey. There will always be another frontier. Nobody has all the answers, so it’s much more important to ask the right questions.

One question I’ve been mingling with lately is, “What makes a breaking ball effective?”

There are dozens of data points we have at our disposal to attempt to answer that question with possibly hundreds more that we don’t even know we should be looking at yet. There really isn’t a hierarchy among the different data points either, so being able to look at different data points as pieces of a larger puzzle is maybe more important than examining certain traits under a microscope or in a vacuum.

So what makes a breaking ball effective? Recently, I’ve been exploring how different angles and locations of pitches can help generate different outcomes.

Subscribe now

Trevor Megill and his outlier breaking ball

If you’ve talked to me about baseball at some point since June, you’re probably aware of my interest in Trevor Megill. I thought he would end up being a big time acquisition at the trade deadline because he doesn’t appear to be a significant part of the Milwaukee’s future plans.

Megill is a 29 year-old RHP currently with the Brewers. His large frame at 6'8 allows for a unique delivery with above average extension that adds roughly half a mile per hour to the perceived velocity on his pitches. Megill’s upper 90s fastball has above average velocity and has slightly above average vertical movement. Fastballs with more of a rising shape than a running or sinking one tend to be platoon-neutral — making him someone who could eventually pitch in a multi-inning role or higher leverage. His breaking ball, labeled as a knuckle curve, is one of three breaking balls this year to average a vertical approach angle steeper than -9.5 degrees at a velocity of at least 85 mph. Megill’s curveball is on the extreme end of a spectrum of steeper and harder than average breaking ball shapes that have helped some players find success; like Zac Gallen, Tyler Glasnow, and David Robertson.

Megill put together some respectable numbers this season and has been fairly serviceable throughout his career, but I believe some minor adjustments could help him reach a higher level. The following suggestions are easier said than done, but I’m confident they could increase his effectiveness. The data only helps if it can be turned into actionable steps and development. Sometimes a subtle shift in mindset or philosophy on the back-end of game planning can lead to significant strides when it’s put into action.

First, his breaking ball is his best pitch and needs to be thrown more. Megill threw his less than 30% last season while he may have relied a bit too much on his fastball — throwing it a little more than 70%. Even a 60/40 split would likely keep enough hitters from sitting on the fastball. However, it needs to be determined why Megill is throwing his best pitch at such a small percentage of the time. The reasons could range anywhere from not feeling confident in himself or his catchers, or maybe he’s not fully aware of how unique his breaking ball is. Answering this question is the first step towards creating actionable steps to increase his breaking ball usage.

Megill’s average pitch locations could use some adjusting as well. His fastball would be more effective up in and above the zone while he could take advantage of the unique depth on his breaking ball by getting it below the zone more often. Throwing the breaking ball in the zone can be effective on the first or second pitch of an AB to steal a strike, but an average location below the zone would generate more swing and miss and poor contact because hitters have to protect against Megill’s upper 90s fastball. Additionally, a lower breaking ball location would play into his strengths by creating an even more steep VAA. The potential barrier to sub-optimal pitch locations could be helped with a universal target from a team’s catchers or targeting optimal locations for different pitches in various counts.

If a transformation into a multi-inning or even starting role is something Megill would be comfortable with, he would have to get a feel more for a cutter or slider to “bridge” the gap between his fastball and breaking ball. The implementation of bridge pitches has been interesting to watch unfold across the league over the last few years. Megill had a cut-slider shape he used a bit in previous seasons, but almost completely abandoned it last season. That’s fine for a short relief role because his other pitch shapes are naturally fairly platoon neutral, but implementing a cut-slider pitch again would be necessary if he’d like to explore a more stretched-out role.

The issue with the cut-slider in the past has been the shape; it hasn’t been distinct enough from his breaking ball in its movement. Again, this data doesn’t mean much if it can’t be translated into actionable steps from a coaching staff or player development team. Megill’s unique build and natural tendencies would ultimately determine the next steps for him in designing an effective bridge pitch.

For what it’s worth, he did alter his horizontal release point towards the end of last season to create less of a horizontal angle to the plate. There were no significant changes to his pitches, but this adjustment could theoretically help him be even more platoon neutral. Megill really only started doing this in August, so the sample for him isn’t large enough to draw any significant conclusions yet. Nonetheless, it could be effective in theory to neutralize LHB.

While his future role in Milwaukee is a bit unclear at the moment and he has played for three different organizations over the last three seasons, Trevor Megill looks like a strong candidate to take a step forward in 2024.

Share

All data and graphs came from either Baseball Savant or Alex Chamberlain’s Pitch Leaderboard v5, but I made the first graph myself using Google Looker Studio.