WHY IT MATTERS

Pitch tipping is not new. Baseball has always chased the ability to pick up tells on what the pitcher is about to throw.

The new part is the mechanism for doing it and the scale it can provide.

What used to revolve around a gut-feel, a sharp-eyed scout, a dugout whisper, and/or hours of manual video can now be turned into a repeatable, systematic process for a detection problem.

• A March 2026 research paper on 119,561 professional pitches showed that body kinematics alone predicted pitch type with 80.4% accuracy before ball flight.

• The same study found that 64.9% of the predictive signal came from the upper body, with wrist position contributing 14.8% on its own.

• That matters because the edge is no longer just spotting an obvious tell.

The edge is building systems that find subtle pre-release signals faster than a human can.

ZOOM IN

In the arXiv paper — Interpretable Pre-Release Baseball Pitch Type Anticipation from Broadcast 3D Kinematics — the authors isolated the body from the ball while using broadcast video.

Enabling them to:

• Reconstruct 3D pose pitching sequences from 119,561 professional pitches.

• Detect key delivery events.

• Build a classifier from 229 kinematic features (a model that identifies based on the movements of a pitcher’s body).

➞ Where the signal actually lives

The paper found:

• 64.9% of the predictive signal came from the upper body.

• 35.1% from the lower body.

• 19% from eyes + nose was the most informative joint group.

• 14.8% from combined wrist joints.

• 12.2% from combined ankle + knee joints.

The model also looked beyond body parts to posture and balance cues:

• Trunk lateral tilt was the strongest individual signal.

• Center of gravity also showed up repeatedly among the top features.

There is meaningful pre-release information in the delivery, but it doesn’t tell the whole story.

The paper also draws a clean boundary around what it can and cannot capture.

The authors point out that grip-defined variants, like four-seam vs. two-seam fastballs, are not separable from pose alone.

Ultimately, using only body movement — no ball flight, no velocity, no spin — the model classified eight pitch types at 80.4% accuracy.

ZOOM OUT

The killer application for the type of pitch-tipping detection model we’re talking about is all about game planning.

• Hitter prep: if a model can surface pre-release cues by pitch type, hitters can go into games with decisive tells to look for — instead of hoping to spot them live.

• Advance scouting: teams can search entire video libraries for repeatable cues, then rank them by confidence, rather than relying on manual tracking and memory.

• Internal pitcher audits: organizations can check whether a prospect’s pitches carry a distinct movement signature before opponents exploit it.

• Player development: the same model that finds a tell can also help remove it by tracking whether the delivery is actually getting cleaner over time.

This is the same pattern you see everywhere else in baseball analytics.

1. A hidden signal becomes measurable.

2. Then it becomes searchable.

3. Then it becomes something teams can model, coach, and optimize.

Pitch tipping is just another problem/solution entering that pipeline.

THE BOTTOM LINE

Picking up on pitchers tipping their pitches is moving from a vague baseball-y instinct toward a measurable, modelable problem.

• 80.4% pitch-type prediction from body movement alone shows that pre-release pitch tipping is a real — not just ballplayer lore, but true signal to pick up on.

• Using single-camera broadcast video suggests the model could be viable outside of big-league lab conditions.

Perfectly highlighting where the future of baseball is headed in general.

More and more parts of the game that have historically been based on feel can be measured, modeled, optimized, and used by teams and organizations to inform actionable game planning for their players and coaches.

WHY IT MATTERS

Mike Trout’s hot week at Yankee Stadium set the baseball world on fire — and with an easy explanation: a tiny step-back with his back foot.

• This is the visible change everyone grabbed onto first

• And they should have, the step-back is real.

• You can see it and hear Mark DeRosa break it down 👇

➞ It’s more than just a step-back

The real story is the vintage contact quality, better swing decisions, and a more athletic version of Trout than we saw a year ago.

Trout’s stats through April 16:

• .416 on-base percentage / .594 slugging percentage / 1.010 OPS, 7 home runs

• .474 xwOBA / .703 xSLG / 25.4% barrel rate / 47.5% hard-hit rate / 21.7% walk rate (via Baseball Savant).

• All good for top-5 in the league at the time, this is not just a guy cheating for pull-side power.

This looks much closer to the vintage Trout package — damage, discipline, and unrivaled athleticism.

ZOOM IN

The tweak to Trout’s mechanics matters because the underlying batted-ball profile changed with it, and the series in the Bronx made the change impossible to ignore.

• Trout went 6-for-16 (.375) with 5 home runs and 3 walks in the four-game Yankees series.

• He became the first visiting player to homer in four straight days at Yankee Stadium.

• His final homer of the series traveled 446 feet and left the bat at 114.6 mph.

• Added 233 points to his OPS in one series.

The step-back move appears to have helped Trout get back to punishing mistakes like he became notorious for during his peak years.

But, CBS Sports noted that Trout used a version of this step-back earlier in his career before moving away from it.

That makes this feel less like an invention and more like a recovery, or a return to old.

ZOOM OUT

The back-foot tweak is getting the headlines because baseball likes visible, easy-to-explain narratives. You can point to a clip. You can freeze the stance.

You can say: there, that's the thing.

But to give all the credit to the step-back takes away from the work Trout has put in:

• Fox Sports reported Trout also dropped roughly 5–7 pounds this offseason to feel lighter on his feet.

• MLB.com has his 28.6 ft/sec sprint speed back in the 90th percentile.

• He prepared for and is back in center field regularly, not stashed at DH.

As compared to his recent injury-riddled seasons, where Trout looked compromised, his game narrowed, and everyone wondered if he can stay on the field? Is he now just a part-time threat?

But with Trout’s early results also coming with large rebounds in contact quality and plate discipline, it’s hard to dismiss this as a simply hot week.

THE BOTTOM LINE

Mike Trout's back foot is what everyone wants to focus on, but the more athletic-looking, complete superstar returning to form is the real story.

• The contact quality says the ball is coming off like vintage Trout.

• The swing decisions say he is seeing the game cleanly again.

• The body looks more playable, more athletic, and more dangerous.

The movement change is something real.

But bigger than the move itself is the fact that Trout is feeling athletic in ways that haven't allowed him to move like this in years

GO DEEPER

The plate-discipline shift, year over year, is the clearest evidence that this is more than a hot week:

• In-zone contact rate: 82.0% in 2025 → 93.0% in 2026 (CBS Sports).

• In-zone swing rate: 55.9% in 2025 → 61.3% in 2026 (CBS Sports).

That combination — swinging at more strikes AND making contact on more of them — is the discipline shift behind the surface power numbers.

WHY IT MATTERS

Every April, baseball analysis falls into the same trap. Somebody points to a hot batting average, a shiny ERA, or an RBI total and calls it real.

Somebody else responds with the laziest line in the sport: small sample size.

• Both sides usually miss the point— the real problem is not that the sample is small.

• The problem is that most people are looking at the wrong things, while the sample is also small.

• Some indicators become useful almost immediately because they are directly tied to player skill.

• Others take much longer because outside variables are still contaminating the picture.

The edge in April is not reacting faster. It is knowing what deserves a reaction in the first place.

ZOOM IN

The fastest clues are usually the ones closest to the underlying skill.

FanGraphs' sample-size reference — built on Russell Carleton's stabilization work — is the starting point:

• Hitter strikeout rate: informative around 60 plate appearances

• Walk rate: gets there around 120 PA

• Pitching K rate: useful around 70 batters faced

• Ground-ball/fly-ball rate: real signal around 80 BIP for hitters, 70 BIP for pitchers

• Chase rate/swing decisions: accumulates signal fast because hitters see ~4 pitches per PA

• Fastball velocity: meaningful immediately, but compare to prior Aprils, not full-season averages

➞ Jordan Walker is the cleanest early example:

His surface breakout matters, but the better story is underneath:

• Chase rate dropped from 34.1% to 28.0%

• Strikeout rate fell from 31.8% to 28.6%

• In-zone barrel rate jumped from 13.2% to 29.6%

That is not random April noise —that's a hitter making better decisions and cashing them in.

➞ José Soriano is the pitcher version:

His ERA is nice. The more useful story:

• 97.5 mph fastball velocity

• 38.8% chase rate

• 33.8% whiff rate

That is the kind of early profile you care about — the process is loud before the season line settles.

ZOOM OUT

This time of year, the calendar fools people.

A hitter can bat .360 for two weeks and still be the same hitter. A pitcher can run a sub-2.00 ERA and still be sitting on unstable ground. The surface numbers are convincing because they are familiar.

But they are also still being shaped by randomness this early in the season.

• Batting average: not informative until ~910 at-bats

• OBP: ~460 plate appearances

• Slugging: ~320 at-bats

• BABIP: ~820 BIP for hitters, 2,000 BIP for pitchers

• ERA, RBI, win-loss: scoreboard outcomes, heavily downstream from the real drivers

If a hitter has a mediocre average but improved chase behavior, stronger contact shape, and better swing decisions, that player may be in a much better place than the early stat line suggests.

If a pitcher has a 1.90 ERA but the strikeout rate is flat, walks are up, and velocity is down, the ERA is not the insight. It is the disguise.

THE BIG PICTURE

The first useful question of the season is not whether the results are real. It is whether the underlying process changed first.

• Strikeout rate, walk rate, chase behavior, batted-ball shape, and velocity give you signal before box-score stats settle.

• Batting average, OBP, slugging, BABIP, ERA, and RBI are too noisy this early to carry the argument alone.

• The right framework is not" ignore everything," it is  "trust the stats that get to skill first."

That is the actual lesson of the first few weeks of the MLB season.

Not patience for patience’s sake — better filters.

WHY IT MATTERS

The first weeks of the season have shown teams are getting into hitter’s counts, but many of them are still turning into pitcher wins.

A hitter’s count is supposed to be where damage starts, but early 2026 data is pointing to a different truth:

• The count advantage only matters if the swing decision matches it.

• If you get ahead but then expand anyway, the count was just decoration.

That’s the part Alex Cora called out after Boston’s 3-2 loss to San Diego on April 4 — his line was simple:

• “We’ve got to swing less.”

• Not because aggression is bad, but because aggression at the wrong time turns it from an advantage hitter to an advantage pitcher.

ZOOM IN

Swing Decisions = the choices hitters make about which pitches to attack, take, or spoil based on count, zone, and game situation.

➞ Boston is the cleanest early example:

• 24 runs: tied for fewest in MLB through eight games on April 4, per MLB.com.

• .294 OBP: 23rd in the majors through that point.

• 8.1% walk rate: 24th in MLB.

• 28.1% strikeout rate: 26th in MLB.

That profile tells you exactly what Cora was talking about: The Red Sox weren’t just getting beat by stuff — they were giving away too many plate appearances after getting leverage in the count.

Hitters’ counts don’t create offense by themselves. They create the opportunity for offense when paired with good swing decisions.

And it seems like MLB’s youngest players might have a leg up in this area compared to the past:

• MLB’s early-season roundup noted that rookie hitters posted an .851 OPS through every team’s first six games.

• The best mark for rookies in that span since at least 1901.

Could this be more than just “young guys are hot?”

Perhaps it’s a clue that in today’s age of player development, the hitters who are surviving are the ones turning favorable counts into strong swing decisions instead of just bigger swings.

ZOOM OUT

For years, the language of hitting treated aggression as a virtue in itself:

• Win the count.

• Hunt your pitch.

• Do damage.

That’s still true — up to a point. But the best hitters aren’t just trying to swing hard once they get ahead.

They’re selectively violent.

That sounds subtle, but it changes the whole at-bat:

• 2-0 and 3-1 used to mean “green light.”

• Now they’re better viewed as information counts where the pitcher is more likely to show intent, and one take can tell you as much as one swing.

• The real edge: forcing the pitcher into a smaller attack shape before committing.

A 95 mph swing on the wrong pitch in a 2-0 count is still a win for the pitcher.

THE BIG PICTURE

The mistake a lot of hitters make in leverage counts: they think the advantage means they should swing more.

Usually, it means the opposite.

➞ A good 2-0 count buys three things:

• Pitch-type clarity: the pitcher is less likely to expand his menu of pitches.

• Location predictability: the zone shrinks toward what he can land for a strike.

• Miss protection: even if you take, you’re still likely in control of the at-bat.

Once you know how the pitcher wants to survive the count, the next swing gets cleaner.

Long story short: the best hitters aren’t succeeding because they swing more once they get ahead. They’re succeding because they are focused on better swing decsions.

READ TIME: 6 MINUTES | WORDS: 1,337

MLB's best metric for catcher framing — Catcher Framing Runs on Baseball Savant — says the gap between the best and worst catchers in 2024 was about 26 runs.

• Patrick Bailey at +16.

• Korey Lee at -10.

• A new machine learning model says the real gap was over 110 runs.

That's not a typo. That's not a rounding error.

That's a fundamental mismeasurement of an entire position.

WHY IT MATTERS

A study published in SABR's Baseball Research Journal by Mizzou physicist Charles Steinhardt and sophomore Zach Borowiak built an XGBoost model trained on 3.5 million Statcast pitches from 2019–2024.

• The model scores every pitch by expected run value, then isolates:

• What the pitcher did

• What the hitter did

• What the catcher influenced — separately.

The catcher findings blew the doors off the existing numbers.

➞ Catcher Framing Runs (CFR; the current standard)

CFR measures one thing: did the catcher turn a ball into a called strike?

• That direct effect ranges from about -10 to +16 runs per season.

• It's the number teams use to value defensive catchers.

• It's the number that justified giving Austin Hedges $4 million despite being one of the worst hitters in baseball.

But the model found that the total catcher effect — including the indirect behavioral impact on hitters — ranged from -50 to +61 runs in 2024.

If correct, the difference between the best and worst catchers was over 11 wins last season.

ZOOM IN: The Part Nobody Was Measuring

Here's where it gets interesting.

When a catcher with elite framing consistently gets that pitch an inch off the plate called a strike, hitters notice.

• Not consciously, maybe — but over the course of a game, they start adjusting.

• They expand their zone. They swing at pitches they normally wouldn't.

• And those swings tend to produce weak contact.

The catcher isn't just stealing a strike call.

He's reshaping the hitter's mental model of the strike zone as he adapts to the zone the catcher is creating, not the rulebook zone.

➞ The Data Backs This Up:

• The best framing catchers induce significantly more swings outside the rulebook strike zone than the worst ones.

• Hitters are swinging at pitches they shouldn't be swinging at, because the catcher has conditioned them to believe those pitches will be called strikes.

• The old metrics missed this entirely.

Those extra swings — the ones the hitter takes because the catcher expanded the effective zone — produce worse outcomes.

• But since the umpire never makes a call on them (the hitter swung), they don't show up in Catcher Framing Runs.

• The existing metric counts the stolen call — it doesn't count the downstream damage.

• The model estimates this behavioral effect is worth an additional 15–20 runs per year for elite framers, on top of the ~15 runs they were already getting credit for.

➞ In The Context of ABS

The fear has been that the challenge system would erode catcher framing value — if every borderline call can get overturned, why does it matter if your catcher makes a ball look like a strike?

• We broke down the ABS Challenge System's early data and found that only about 1% of pitches get challenged.

• The system's impact is behavioral, not mechanical — it changes how hitters think about the zone, not the outcome of individual pitches.

This study reveals the same mechanism running in the other direction.

• ABS tightens the zone, and hitters stop chasing.

• An elite framer expands the zone, and hitters start chasing.

• Same behavioral lever, opposite pull.

ABS can correct a call — it can't correct the swing decision the hitter already made because the catcher reshaped his zone all game.

The part of framing value that ABS threatens — the stolen call itself — turns out to be the smaller piece of the total.

The bigger piece, the behavioral cascade, doesn't get overturned by a challenge.

ZOOM OUT: Hitters Guess, Catchers Exploit

The framing-plus-behavioral effect isn't even the full story. The study also found evidence that pitch calling — the sequencing of pitches — is a separate, roughly equal source of catcher value because the model proved that hitters guess.

• When a pitcher throws the same pitch back-to-back, hitters actually perform worse — they guessed wrong, expecting something different.

• When that same pitch shows up again later in the at-bat with a different pitch in between, hitters perform better — they'd seen it earlier and adjusted.

• The logical explanation: hitters aren't just reacting.

• They're predicting what's coming next, and that prediction shapes their swing decisions before the pitch even arrives.

Though Ted Williams endorsed this approach in The Science of Hitting, coaching culture still treats "guess hitter" as a pejorative.

• But this model shows the best hitters in the world are doing it — and the data says they're not guessing enough.

• If hitters improved their approach to back-to-back repeated pitches alone, it would be worth roughly 1.8 to 4.1 wins per team per season.

➞ The Other Side of the Guessing Game

On the other side of hitters trying to anticipate what pitch is coming is the catcher calling pitches.

• Pitch calling value in the model: up to 25–30 additional runs per year for the best catchers.

• Uncorrelated with framing ability: a great framer isn't necessarily a great pitch caller, and vice versa.

• The rare catcher who does both? Jake Rogers saved 61 runs in 2024, according to this model.

• That's equivalent offensive value to what earned Juan Soto his $765 million contract(!).

And pitch calling doesn't care about ABS at all.

The challenge system corrects ball-strike calls. It has nothing to say about whether the catcher called for a changeup when the hitter was sitting fastball.

THE BOTTOM LINE

If this model holds, the "bad hitter, elite catcher" archetype isn't absorbing an offensive cost — he's potentially one of the highest-value contracts in baseball per dollar spent.

• Because the stolen strike was never the whole story.

• Behind every one is a behavioral cascade — hitters expanding their zone, chasing more, making weaker contact — that doubles or triples the real defensive value.

• The team that finds a catcher who frames and calls a game has an advantage that existing metrics have been drastically undervaluing for over a decade.

The question was never whether ABS would kill framing.

The question is whether we ever measured the full scope of what framing was doing in the first place.

DIG DEEPER

There's an ironic wrinkle and fascinating subplot that played out in spring training.

• Yardbarker's Jay Staph ran the early ABS challenge numbers and found the best framers in baseball are actually worse at using the system so far.

• The top five framers last season: 58.6% challenge success rate.

• The bottom five: 61.5%.

➞ Trevor May’s Theory

The former MLB reliever has stated that it’s possible the elite framers have gotten so good at making balls look like strikes that they've started fooling themselves.

• They frame a pitch just off the plate, their brain registers it as a strike, they challenge — and the robot says no.

• The very skill that creates the behavioral value this study identified is now a cognitive blind spot when a machine gets the final word.

Catchers are learning they can't necessarily always trust their own perception of the zone they've been manipulating for years.

By The Numbers

• +61 to -50 runs: Full catcher score range in 2024, vs. +16 to -10 for Catcher Framing Runs alone

• 11+ wins: Estimated gap between the best and worst catchers

• 15–20 runs/year: Additional value from the behavioral effect (hitter zone expansion) that existing metrics don't capture

• 25–30 runs/year: Pitch calling value for the best catchers — uncorrelated with framing skill

• 3.5 million pitches: XGBoost model training data (2019–2024 Statcast)

• 58.6% vs. 61.5%: ABS challenge success — best framers vs. worst framers

➞ The Study

"Hitter and Catcher Adaptation in Major League Baseball" by Charles Steinhardt and Zach Borowiak, published in the Fall 2025 Baseball Research Journal (SABR).

• Borowiak presented the research at Undergraduate Research Day at the Missouri Capitol earlier this year.

WHY IT MATTERS

ABS is a zone-discipline accelerant. It doesn't just fix bad calls — it recalibrates how hitters approach every at-bat, including the 99% of pitches that never get challenged.

• Pitches just off the plate that normally generate chase swings? Hitters took them.

• One pitcher told Baseball America his first outing made it clear: you've got to go for in-zone whiffs instead of chases, because these hitters know the zone.

• MLB hopes this breaks the 70-year walk-rate freeze — stuck between 7.6% and 9.6% while strikeout rates keep climbing.

• If pitchers can't nibble the edges knowing umpires will expand for them, the entire approach to at-bats shifts.
  

ZOOM IN

The numbers from Triple-A's 861,000 pitches and spring training 2026:

• ~50% overturn rate: umpires are right half the time, even when players cherry-pick their best opportunities

• 55% fielder success vs. 45% batter success: catchers calibrate across nine innings; hitters get 4-5 ABs to learn the zone

• Only 1% of pitches drew a challenge: the system's impact is behavioral, not mechanical

• Challenge rate doubles from 1st inning (2.1%) to 9th (~5%): but success rate drops. Early challenges are confident; late ones are desperate
  

Teams are already splitting on philosophy this spring.

• The A's let players police themselves: 69.2% success rate, best in baseball.

• The Yankees challenge everything: 3.8 per game, 52.6% success.

• The Dodgers sit at 22.7%.

The teams (and players) treating challenges as a strategic asset, not a reflex, are already separating themselves.

ZOOM OUT

Those numbers play out in real shifts on the field.

Catcher framing isn't dead — it's evolved.

• 98%+ of pitches still get called by human umpires.

• A catcher who steals borderline strikes forces opponents to burn challenges. A poor framer forces his own team to waste them.

• Framing becomes a challenge-management tool — different mechanism, still valuable.

• Full ABS would obliterate framing overnight, which is exactly why MLB chose the challenge model.

Veteran privilege is gone.

• Walker Buehler said the quiet part loud:

"I think starting pitchers that have pitched for a long time deserve certain parts of the plate that other guys don't get."

• He's describing the “Maddux Zone” — umpires granting expanded zones to established aces.

• ABS is a leveling mechanism that doesn't care about career or status.
  

THE BOTTOM LINE

ABS isn’t just a replay system. It’s a behavioral reset.

The data is unambiguous. When the zone is enforced:

• hitters stop chasing

• pitchers stop nibbling

• and the game’s entire hitter/pitcher matchup evolves.

The teams that figured this out sooner rather than later — treating challenges as a strategic asset, retraining zone discipline — will develop a measurable edge.

Watch on YouTube

WHY IT MATTERS

The BAR hitter dashboard helps find the patterns and tendencies Shohei exploited to dominate the Brewers on both sides of the ball.

• See how each pitch moves: Movement Profile shows the up/down and in/out break for every pitch a pitcher throws.

• Learn what’s likely coming in each count: Pitch Type by Count breaks down what a pitcher likes to throw and where, based on the count in the at-bat.

• Find your zones to hunt: Expected wOBA (Weighted On-Base Average) by Pitch Location identifies a hitter’s most productive zones while giving an idea of where the pitcher will likely attack them.

SHO-TIME: INSIDE OHTANI’S NLCS GAME 4 TAKEOVER

Let’s look at Game 4 through the eyes of Shohei — and some Brewers hitters who faced him — using the BAR hitter dashboard to reveal how Shohei delivered the most complete performance in MLB history.

3 Home Runs

The Movement Profile, Pitch Type by Count, and xwOBA by Pitch Location visuals each provide a piece of the puzzle to game-planning for the best swing decisions.

➞ Leadoff HR vs Quintana

- Movement Profile: Jose Quintana’s sinker stays up and runs in on lefties more than the typical MLB sinker.

He throws it more than half the time to lefties, mixing in a four-seam fastball with similar movement/velocity, and a slurve with above-average break that also stays up more than usual.

- Pitch Type by Count: In 3-2 counts (where Shohei first homered), Quintana’s sinker/four-seamer usage spikes while slurve usage drops.

Combined with the two most frequent pitch locations, we can determine Quintana likely attacks with velocity middle/in or a slurve middle/out.

- xwOBA by Pitch Location: Despite a smaller hot zone vs lefties compared to all pitchers, the xwOBA heatmap shows that even the locations where Shohei cools off, he still outperforms the league.

- The Plan: Knowing Quintana’s pitch movement and tendencies in 3-2 counts, Shohei can protect the zone while hunting Quintana’s pitches that come back over the plate.

➞ Out of the stadium vs Patrick

- Movement Profile: Chad Patrick throws 84% fastball variations — a four-seam and sinker with above-average run from lefties, plus a cutter that cuts more than usual.

When he does throw something soft, it’s a changeup with heavy arm-side run.

- Pitch Type by Count: You never want to be in a 3-1 count to Shohei, but especially not Patrick when his pitch mix and location become more predictable.

It’s cutters down & in or sinkers that come back over the inside part of the plate to lefties.

- xwOBA by Pitch Location: Shohei’s heatmap shows deep, dark reds on the inside parts of the plate, exactly where Patrick likes to go in 3-1 counts — that should tell you everything you need to know.

- The Plan: Hunt for inside velocity.

➞ The hat trick vs Megill

- Movement Profile: A true two-pitch pitcher, Trevor Megill throws a 98-100mph four-seam fastball with exceptional rise 63% of the time to lefties and a knuckle curve with below-average break the rest of the time.

- Pitch Type by Count: Megill had the advantage in a 1-2 count, where his pitch mix is less predictable (four-seam 53%, knuckle curve 47%).

But he clearly doesn’t attack in & off the plate to lefties in this count and repeatedly misses in the heart of the zone.

- xwOBA by Pitch Location: Again, looking at Shohei’s intimidating heatmap, this immediately warns that Megill is toast if he ends up throwing something in the very spot he’s prone to in this count.

- The Plan: With only two pitches to worry about, Megill’s tendency to miss over the plate and rarely bust lefties inside lets Shohei set his eyes on the heart of the zone, then just see the ball and react.

10 Strikeouts

Using the same visuals, the BAR hitter dashboard shows Shohei’s pitch mix, movement, and how he struck out the side in the first inning — setting the tone for the rest of the game.

➞ Shohei’s Stuff

- Movement Profile: Good luck to righties. Shohei throws his 98–100 mph four-seam fastball and his devastating sweeper at nearly identical rates — the latter breaking up to 20”, compared to the standard 6” to 10”.

Lefty batters have it slightly easier. They’ll see a four-seamer 44% of the time, but after that, Shohei’s cutter, slider, and elite curveball all show up at nearly equal rates.

➞ Shohei Blows Chourio Away

- Pitch Type by Count: You don’t want to be down 0-2 against Shohei. Righties now have to be just as ready for the slider as the four-seamer or sweeper.

Location-wise, he’s living down & away with breakers, up and away with heat.

- xwOBA by Pitch Location: That’s a problem for Chourio. His heatmap reveals his weakest zone is up & away — exactly where Shohei likes to go.

- The Plan: Chourio needs to bear down and be ready to battle on the outside edge. Or else it’s light out.

➞ Shohei Paints vs Yelich

- Pitch Type by Count: In 2-2 counts, lefties lose their edge in Shohei’s pitch mix. Now it’s 58% four-seamer or slider, and 34% curveball or sweeper.

Along with four pitches to defend, they have to cover the whole zone — though Shohei tends to dot the down and away.

- xwOBA by Pitch Location: Bad news, again. The heatmap shows Yelich has a soft spot where Shohei likes to go in this count.

- The Plan: Yelich also needs to bear down and battle, but focus more down & away than Chourio. Or else it’s light out, again.

➞ Shohei Makes Contreras Chase

- Pitch Type by Count: Another 0-2 count, Contreras now faces the same issues Chourio did.

- xwOBA by Pitch Location: Rough. Contreras’ heatmap shows glaring holes down & away, and up & in. Remember where Shohei likes to go in 0-2 counts?

- The Plan: Contreras better get ready to go play defense.

THE BOTTOM LINE

This look at Shohei’s NLCS Game 4 through the BAR hitter dashboard shows how you can game plan to make better swing decisions.

That’s half the battle in good hitting.

Who should I break down next?

Let me know which hitter performance or hitter/pitcher matchup you want broken down in the BAR hitter dashboard.

ZOOM OUT: NOW YOU SEE JAVY

I first saw it in 2013: a grainy MiLB clip titled "Javier Baez 4th Home Run Daytona Cubs." A moonshot from a whippy swing with extreme movements.

• Once a full-time Chicago Cub from 2016-2021, the most aggressive bat tip I'd ever seen led Javy to rank near the top of MLB in maximum exit velocity (Max EV), average exit velocity (Top 8th EV Average), and bat speed (Bat Speed).

Max EV

• Highest recorded speed of a ball off a hitter's bat.

Top 8th EV Average

• Average exit velocity of a hitter's top 12.5% of hardest batted balls.

Bat Speed

• Estimated in-game bat speed derived from the top 12.5% of exit velocities.

• Uses the top 8th EV Average and Robert K. Adair's model for deriving bat speed from exit velocity.

• 2021 Season (Cubs 91 games; Mets 47 games)

  • Max EV: 96th percentile

  • Top 8th EV Average: 94th percentile

  • Bat Speed: 94th percentile
    

• No surprise: Javy's violent swing mashes baseballs and generates impressive bat speed.
  

• Surprise: he did it while making better swing decisions and finding more barrels than the rest of the league.

ZOOM IN: NOW YOU DON’T

At the 2021 Trade Deadline, El Mago moved his act to New York before signing an offseason deal with the Detroit Tigers (2022-2027).

• Despite a spellbinding walk-off winner on Opening Day, it would soon be clear that Javy left some magic behind in Chicago.

One sign was the drop in Javy's ability to routinely hit balls with an optimal exit velocity and launch angle — otherwise known as a Barrel.

• After hovering around the 90th percentile since 2018, Javy's percentage of hits that qualified as a barrel (Barrel%) slid to the 67th percentile in his first year with Detroit (2022).

• From there, the descent continued, dropping to the 35th percentile in 2025 (as of 8/31).

Barrel%

• The rate at which a player's batted balls have an exit velocity and launch angle that has resulted in a minimum .500 batting average and 1.500 slugging percentage since 2015, when Statcast was implemented MLB-wide.
  

• Starts at 98mph with 26-30º; widens 1º per side with every +1mph until reaching 8-50º at 116mph+.

But it wasn't the balls Javy was hitting that gave him away. It was the balls he was swinging at with no chance of hitting that marked El Mago's disappearance.

• Those swings are jarring, given that Javy's ability to swing or not at quality pitches (xwSwingDecisions) hadn't ranked below the 95th percentile since his first full year as a Cubbie.
  

xwSwingDecisions

• The expected runs gained or lost above average by swinging/not swinging at a given pitch.

• League-adjusted; 0 = MLB average (the higher the better).

• Javy's last four seasons in Chicago (xwSwingDecisions)

  • 2018: 99th percentile

  • 2019: 98th percentile

  • 2020: 95th percentile

  • 2021: 98th percentile
    

• Javy's first four seasons in Detroit (xwSwingDecisions)

  • 2022: 90th percentile

  • 2023: 81st percentile

  • 2024: 34th percentile

  • 2025 (as of 8/31): 64th percentile

Javy's swing decisions have trended back in the right direction this season — perhaps the murmurs of El Mago's return are no coincidence.

EL MAGO REAPPEARS?

The first half of the 2025 season saw excitement for El Mago returning to form. Plus, a renewed energy for his defense, thanks to a move to center field.

By the dog days, Javy's underlying hitting metrics, although rebounding in some areas, show we are still waiting for the actual return of Chicago's El Mago.

• Javy's hitting style has never allowed him to be particularly precise with his swing — topping out around the 60th percentile with his swing efficiency (Smash Factor) even in his best years with the Cubs.
  

Smash Factor

• Borrowed from the golf world via Driveline Baseball, measures the efficiency of a hitter's contact with a batted ball or series of batted balls.

• Superior to other contact quality metrics (line drive percentage, hard hit rate, whiff percentage, etc) because it controls for a hitter's bat speed.

  • According to Driveline, most batted balls have a Smash Factor between 0.9 and 1.2 (the higher the better contact quality).

• But, as a Tiger, Javy's Smash Factor has been toward the bottom of the league.

  • 2022: 11th percentile

  • 2023: 16th percentile

  • 2024: 8th percentile

  • 2025 (as of 8/31): 22nd percentile
    

If paying attention, you noticed the tick up in both Smash Factor and xwSwingDecisions this season.

Which is interesting considering that Javy is indeed having his best season in Detroit so far, yet the other four previously discussed metrics are at all-time lows for him.

THE BIG PICTURE

To paraphrase the greatest hitter who ever lived—Ted Williams—effective hitting is 50% "proper thinking" and 50% "hitting to your style".

• Let's call it 50% swing decisions and 50% mechanics.
  

That aligns with our Javy analysis, as he’s been able to improve his stats in 2025, despite only performing above-average in xwSwingDecisions.

• Ultimately, this type of analysis alone doesn't help a hitter actually improve.
  

• Future installments of BAR will dive into analyses that can help you make better swing decisions (hitter game planning) and take better swings (swing design).

The monkey reference (Marcel from Friends) comes from Tango’s intention to build the most basic forecasting system, using as little intelligence as possible.

WHY IT MATTERS

Since MARCEL’s introduction in 2004, player projection systems have become increasingly complicated — yet often aren't much more accurate than Tango's simple approach.

Let's take a step back, return to MARCEL's three-step foundation, and then focus on improving only the lowest-hanging fruit.

1. Previous Stats: MARCEL takes the player’s results from the last three MLB seasons, then weighs them by recency — last year's stats are multiplied by 5, two years ago by 4, and three years ago by 3

   1. The total for any given stat is then divided by 12 (because 5+4+3 = 12).
      

2. Regression Towards the Mean: To account for randomness and varying sample sizes, MARCEL assigns the result from Step 1 a reliability score based on the number of appearances compared to the rest of the league.

   1. The reliability score is then combined with the league average result to give us a regressed prediction for the player.

      1. The more appearances, the higher the reliability score, and the less we need to regress the player to league average.
         

3. Age Adjustment: Finally, MARCEL adjusts the stat line produced by the first two steps based on the player’s age since human performance is known to vary based on our age.

   1. For players under 29, increase by 0.006 per year below 29.

   2. For players over 29, decrease by 0.003 per year above 29.

So, can we take MARCEL up a notch without pushing it too far?

BETTER THAN MARCEL

Steps 1 and 3 contain the easiest levers we can pull to improve MARCEL while maintaining the principles of simplicity — as described below for predicting On-base Plus Slugging (OPS).

Step 1 | Previous Stats Upgrade

Using the baseballr package and Baseball Savant's API in R, we can enhance Step 1 of the player’s predicted OPS by factoring in his Statcast swing metrics from the three seasons of concern.

• The simplest way to do so is by accounting for his:

  • Barrel Rate (%) - How often the player hits a ball ≥98 mph with a ∼26-30° launch angle — barrels have led to a minimum .500 batting average and 1.500 slugging percentage, making it a strong gauge for assessing OPS.

    

  • Out of Zone Swing Rate (%) - Tracking Barrels inherently accounts for both the ability to get on-base and slug, so we’ll enhance Step 1 just a bit more with the simplest measure of plate discipline and walk potential — how often the player swings at pitches outside the strike zone.

Step 3 | Age Adjustment Upgrade

We can implement a simple improvement to MARCEL's age adjustment by utilizing the Mixed GAM Computation Vehicle (mgcv) package and gam() function in R.

• Tom Tango’s approach to measure player aging and determine the -.0.003 and 0.006 adjustment figures used in MARCEL is called the Delta Method.

  • A non-parametric, arithmetic model that projects OPS to follow the curve pictured below as a player ages:

    

• In the years since MARCEL, a semiparametric regression model known as a generalized additive model (GAM) has proven to be an improvement on the Delta Method for projecting the effects of aging on player production.

  • As rated by Mean Absolute Error — a measure of average difference between forecast and actual.

THE BIG PICTURE

A 14th-century problem-solving principle known as Occam’s razor recommends you search for explanations while using only the smallest possible set of elements.

Tom Tango clearly understood this call to simplicity when he created the Marcel the Monkey Forecasting System (MARCEL) for MLB player projection.

Whether using one of today’s complex forecasting systems or our simple adjustments to his model, you’ll struggle to meaningfully outperform plain-old MARCEL.

WRITER’S NOTE: this article was written just prior to the contract extension reached by Vladdy Jr. and the Blue Jays on April 7, 2025.

• Yankees: Already scrambling after Soto’s departure, the influx of injuries to centerpiece players leaves lots of uncertainty in the Bronx.

• Red Sox: It’s only a matter of time before the BoSox take a leap forward, but they’re still young and mostly unproven — while position/performance drama grows and orbits (a)round their homegrown star.

• Orioles: The hype was real for a minute, but last year’s regression, doubts about their top prospect, and lack of moves this offseason have raised questions about what’s next for Baltimore.

• Rays: You admittedly can never sleep on the Rays, but we certainly seem to be beyond their magic era of the past.

Despite this apparent window to step up, the Toronto Blue Jays and Vladimir Guerro Jr.’s self-imposed Opening Day deadlines to agree on a contract extension have come and gone.

But, as you can hear from Ken Rosenthal, these deadlines are fake, and the Blue Jays desperately need to make a move ↓

WHY IT MATTERS

The Blue Jays are in a prime position to become one of the AL's dominant forces despite their underwhelming 2024 season.

• Roster flexibility: Only four players signed through 2026 (prior to the Alejando Kirk extension on Opening Day)

• Revenue streams: "Canada's team" boasts strong TV deals and revenues while off-season renovations to the Rogers Centre are set to boost income further.

• Fan engagement: Passionate but wavering due to recent plateaus and uncertainty about the future of their homegrown stars — Vladdy Jr. and Bo Bichette.

• Investment capability: Recent free-agent pursuits (Ohtani and Soto) show an ability and willingness to invest in the roster.

THE BIG PICTURE

Landing coveted free agents like George Springer in the past required clarity in the team's future — recent misses point to a perception of instability.

➞ Vladdy Determines the Blue Jays’ Future

• He's not just a star; he's Canadian, the franchise cornerstone, and deeply connected to fans.

• Committing to Vladdy signals that Toronto is a destination for players.

• Vladdy is on a Hall of Fame trajectory — extending him now helps solidify our future outlook.

Despite some expected ups and downs since debuting in 2019, Vladimir Guerrero Jr. has firmly established himself as one of MLB's premier offensive players with an elite combination of approach, power, and swing quality.

• Plate Discipline: Vladdy's K% (15.6%) is exceptional for a power hitter —outperforming Freeman, Tucker, Soto, Harper, and Judge.

  • BB/K ratio (0.63, 22nd overall since 2019) ranks alongside Harper, Tucker, and Betts.

  • Whiff% compares favorably to Soto, Alvarez, and Machado — significantly outperforming sluggers like Judge, Ohtani, and Alonso.

• Swing Metrics: Vladdy leads MLB in Hard Hits (95+ mph) since his debut and maintains a Hard Hit% trailing only Judge, Acuña Jr, and Alvarez.

  • Max Exit Velocity (118.9 mph, 6th) and Avg. Exit Velocity (92.8 mph, 8th) put him among MLB's most feared sluggers, including Stanton, Tatis Jr., Schwarber, and Ohtani.

• Expected Stats: The strength of Vladdy's underlying swing is reflected further in his career xwOBA (.378), xSLG (.567), and DRC+ (135).

  • Ranking him consistently among the league's best hitters, even in his "down" years.

  • Vladdy's expected results match Judge, Soto, Alvarez, Ohtani, Harper, Freeman, Tucker, and Witt Jr.

• Weighted/Adjusted Offense: Vladdy's actual production confirms the impact of his swing metrics and expected results.

  • He ranks in the top 14 in wOBA (.367, 14th), wRC+ (137, 12th), Batting Runs (154.6, 9th), and OPS+ (137, 10th).

  • Placing him among MLB's most productive hitters alongside Judge, Soto, Ohtani, Harper, Freeman, Betts, Alvarez, Acuña Jr., Trout, and Tucker.

• Consistency & Resurgence: Despite inconsistent results, Vladdy's consistently elite approach and swing have fueled career results most similar to Hall of Famer Eddie Murray (936.6, Bill James' Similarity Scores).

  • His 2024 resurgence (5+ fWAR, 6+ bWAR, 6+ WARP, ~.400 wOBA, 160+ wRC+) reaffirmed his potential to become one of MLB's all-time hitters.

BY THE NUMBERS

A recent article by Mike Petriello, MLB.com writer and Director of Stats & Research, places Vladdy among a group of young stars on the path to the Hall of Fame — citing his 17 career fWAR.

➞ Building on this, we leveraged two industry-standard projection systems, Steamer and Depth Charts, to map that potential Hall of Fame path through Valddy's age-40 season.

• Projection Methodology: Leveraging Artificial Intelligence & Machine Learning (AI/ML), Tom Tango's regression toward the mean, and Jonathan Judge's general additive model (GAM) Across aging curve were applied to three datasets to project Vladdy’s stats through the 2039 season:

  • (1) Vladdy's 2019–2024 statistics

  • (2) Vladdy's 2019–2024 statistics + Steamer 2025

  • (3) Vladdy's 2019–2024 statistics + Depth Charts 2025.
    

➞ Weighting and averaging the three projections gave us a comprehensive outlook for Vladdy's career stats from 2019 through the 2039 season.

• Hall of Fame Case: Confirming Petriello, our projections give Vladdy a career fWAR of 71.81 — just beyond the typical Hall of Fame range of 50-70.

  • Good for 56th all-time using today's rankings and directly behind Eddie Murray (72 fWAR).

  • Making Murray Vladdy's most similar player through age 25 and according to his projected career stats.

• Other Projected Career Totals of Note:

  • 3,317 H: 12th all-time.

  • 534 HR: T-19th all-time.

  • 1,735 R: 25th all-time.

  • 1,804 RBI: 23rd all-time.

  • 133 wRC+: T-130th all-time.

  • Slash Line: .348 OBP / .475 SLG / .339 wOBA.

KEEP VLADDY IN TORONTO

The following are three potential contract proposals to keep Vladdy in Toronto long-term:

1) Bottom Tier | Most Team Friendly

• Length: 14 years (2025–2038)

• Guaranteed: $385,000,000

• Deferrals: 10% of each guaranteed year's salary, paid 2039–2048 (no interest)

• Opt‐Out: None

• No‐Trade Clause: Partial (10‐team block), expands to full at 10-and-5

• Team Options: 2037 & 2038 at $25,000,000 each vs. $10,000,000 buyout each

• Projected Value: ∼$7.025M per win (54.8 WAR projected over 14 seasons)

  • Below the $8.5M cost per WAR found by Ben Clemens

Salary Structure

• Signing Bonus: $10,000,000 (split $5M in 2025, $5M in 2026)

• Base Salaries (2025–2036): $355,000,000 total; each year has 10% deferred, 90% paid at the time.

• If Both Options Declined: Vladdy receives 2 × $10M = $20M in buyouts, plus the earlier $365M (base + bonus).

• Total Guaranteed: $385,000,000

• CBT AAV: $385M ÷ 14 ≈ $27.5M

Incentives

• Plate Appearances: +$1M at 550 PA, +$1M at 600PA (max $2M/year)

• Awards: $1M for MVP win, $500k for top‐3 finish

• Postseason: $1M for ALCS victory, $1M for WS victory (max $2M/year)

Potential Total

• Team Options: If exercised at $25M each instead of $10M buyouts, Vladdy would gain $30M above the guaranteed amount.

• Incentives: Over 14 years, a maximum of $70M is possible if all performance bonuses are triggered annually — conservative estimate assuming strong but not perfect performance lands between $15M–$25M in total bonuses.

• Rough Grand Total: $385M + $30M + ~$20M = $435M (no upper cap, just an estimate).

➞ Vladdy's Acceptance: 4/10

• Pros: Long-term security at $385M is still enormous.

• Cons: Extremely lengthy (14 years), no opt-out, partial deferrals (10%). Lower annual cash flow and no chance to re-test free agency in his prime.

➞ Blue Jays' Favorability: 8/10

• Pros: Low(ish) AAV ($27.5M for CBT), partial deferrals, no opt-out. The club secures a franchise cornerstone for nearly his entire career on a predictable budget.

• Cons: Significant total outlay and 14-year commitment if performance declines.

2) Mid Tier | Soto-Style Deal

• Length: 12 years (2025–2036)

• Guaranteed: $425,000,000

• Deferrals: 0%

• Opt‐Out: After Year 7 (2031), team can void opt-out by increasing salaries by $5M/year in 2032–2036.

• No‐Trade Clause: Full from Day 1

• Team Options: 2035 & 2036 at $40,000,000 each vs. $15,000,000 buyout each

• Projected Value: ∼$7.9M per win (53.62 WAR projected over 12 seasons)

Salary Structure

• Signing Bonus: $15,000,000 ($10M in 2025, $5M in 2026)

• Base Salaries (2025–2034): $380,000,000

• If Team Declines 2035–36: 2 × $15M = $30M => total $425M (with the $15M bonus included).

• CBT AAV: $425M ÷ 12 ≈ $35.4M

Incentives

• Plate Appearances: +$2M at 550 PA, +$1M at 600PA (max $3M/year)

• Awards: $1M for MVP win, $500k for top‐3 finish

• Postseason: $1M for ALCS victory, $2M for WS victory (max $3M/year)

Potential Total

• Team Options: Exercising both at $40M each instead of $15M buyouts yields an additional $50M above the guaranteed amount.

• Incentives: Over 12 years, a maximum of $84M is possible if bonuses are maxed out each season — a realistic scenario based on regular All-Star–level production and team contention would yield $20M–$35M in bonuses.

• Rough Grand Total: $425M + $50M + ~$30M = $505M (no upper cap, just an estimate).

➞ Vladdy's Acceptance: 6/10

• Pros: High guaranteed ($425M), shorter term (12 years) than the low-tier deal, no deferrals, and a mid-contract opt-out that forces the team to pay more or let him walk.

• Cons: Still a lengthy contract; if he prefers an even earlier opt-out or an even bigger headline figure, he may hold out for more.

➞ Blue Jays' Favorability: 7/10

• Pros: They get Vlad for at least 7 years and can void the opt-out if they're willing to pay a higher salary afterward. No deferrals keep future liabilities simple.

• Cons: Higher AAV ($35.4M) from Day 1, plus the club risks losing him at the opt-out point or increasing salary obligations significantly.

3) Top Tier | Ohtani-Style Deal

• Length: 10 years (2025–2034)

• Guaranteed: $495,000,000

• Deferrals: 70% of each nominal salary (no interest), paid 2035–2049

• Opt‐Out: After Year 4 (2028), Blue Jays can void by adding $5M/year for 2029–2034

• No‐Trade Clause: Full from Day 1

• Team Option: 2035 at $50,000,000 vs. $10,000,000 buyout

• Projected Value: ∼$9.7M per win (50.84 WAR projected over 10 seasons)

Salary Structure

• Nominal Salary: $49,500,000 annually × 10 = $495,000,000

• Current Payment Each Year: 30% of $49.5M = $14,850,000

• Deferred Each Year: 70% = $34,650,000 (paid 2035–2049)

• CBT AAV: $495M ÷ 10 = $49.5M

If the team declines in 2035, Vladddy will receive a $10,000,000 buyout and still collect all deferred amounts. The guarantee is $495M.

Incentives

• Plate Appearances: +$3M at 550 PA

• MVP: $2M for MVP win, $1M for top‐3 finish

• Postseason: $2M for ALCS MVP, $3M for WS MVP (max $5M/year)

Potential Total

• Team Option (2035): $50M vs. $10M buyout = +$40M if exercised.

• Incentives: Over 10 years, a maximum of $100M in bonuses is possible if Vladdy consistently posts elite seasons and racks up awards — a strong but flawed outcome would realistically yield $50M–$70M in total incentives.

• Rough Grand Total: $495M + $40M + ~$60M = $595M (no cap).

➞ Vladdy's Acceptance: 7/10

• Pros: Massive headline sum ($495M). Early opt-out after Year 4 means a second bite at free agency unless the team bumps his pay. Even with deferrals, he sets a historic benchmark in nominal total.

• Cons: Actual annual cash-in-hand is only 30% of $49.5M — no interest added.

➞ Blue Jays' Favorability: 7/10

• Pros: Extremely high nominal figure but minimal present payroll impact (only $14.85M/year in real time). The team can keep building around Vladdy ASAP.

• Cons: Huge deferred liability. If Vladdy declines or gets injured, still on the hook for massive sums down the road. The early opt-out also provides less long-term certainty than the other deals.

➞ Best Choice

While the top-tier proposal carries long-term liabilities, its short-term benefits are undeniable.

• Vladdy can tout the headline of potential to exceed $560M, while heavy deferrals provide payroll flexibility—absorbing just $14.85M/season—freeing resources to strengthen the roster elsewhere.

• Combining these advantages with the opportunity to make Vladdy a lifelong Blue Jay makes this our best choice.

THE BOTTOM LINE

The Blue Jays face a pivotal moment.

The American League lacks dominant teams and star players, the AL East is in flux, and the face of the franchise projects to be one of MLB's all-time great hitters.

But he could be on the verge of leaving in free agency — which would be especially catastrophic with rumors already linking him to both the Red Sox and Yankees.

A long-term extension for Vladimir Guerrero Jr. is essential.

GO DEEPER

In using AI/ML, I acknowledge the likelihood of significant errors in projecting Vladdy’s future stats.

If this were more than a mere thought exercise, more vigor would be applied to verify proper calculations and model training.

➞ Career Projection Weightings

• 25% projection from Vladdy's 2019–2024 statistics

• 37.5% projection from Vladdy's 2019–2024 statistics + Steamer 2025

• 37.5% projection from Vladdy's 2019–2024 statistics + Depth Charts 2025

➞ Full Career Projections

➞ Top 10 Most Similar Based on Career Projections (Using Bill James' Similarity Score Formula)

1. Eddie Murray (891), 72.0 career fWAR

2. Rafael Palmeiro (846), 70.0 Career fWAR

3. Miguel Cabrera (840), 68.8 Career fWAR

4. Carl Yastrzemski (833), 94.8 Career fWAR

5. Dave Winfield (814), 59.9 Career fWAR

6. Adrian Beltre (784), 83.5 Career fWAR

7. Albert Pujols (774), 89.9 Career fWAR

8. David Ortiz (748), 51.0 Career fWAR

9. Frank Robinson (745), 104.0 Career fWAR

10. Reggie Jackson (728), 72.7 Career fWAR

Let’s start with the obvious question: what's a distribution?

Following our insight into summary statistics, we can define distribution as "how the data is spread out across all possible values."

Revealing:

• Shape (the 'curve')

• Clusters (central tendency: mean, median)

• Spread (dispersion: standard deviation, range).

➞ The 80/20 Rule

A commonly known distribution is Pareto's Law/Distribution, widely considered one of life's power laws.

It states that 80% of outputs come from 20% of inputs for many relationships, and you see it across all walks of life:

• Wealth: 80% of the wealth is held by 20% of the population.

• Nature: 80% of a garden's yield comes from 20% of planted seeds.

• Productivity: 80% of results come from 20% of the effort.

Now that we have a firm grasp of distributions, the next logical step, if you're thinking along, is to decide when to use…

➞ Central Tendency Statistics vs. Dispersion Statistics

The reductive quality of summary statistics makes this a crucial consideration for two main reasons:

1. You're vulnerable to picking the wrong statistic to understand the data best.

2. You can easily miss significant differences between the groups in your data.

That's why it's a great idea to visualize the distribution of your data before committing to any one summary statistic.

How to best visualize a distribution depends on the type of variable you're concerned with, often requiring you to transform variables.

WHY IT MATTERS

Your variables can be categorical or numerical.

Categorical: Variables that can have only one of a small set of possible values (e.g., baseball players' batting order can only be 1-9) and are commonly visualized using a bar chart.

Numerical: Variables that can have a wide range of values (e.g., the number of home runs a batter hits in a season) and are commonly visualized using a histogram or density plot.

Though the specifics of creating the visualizations mentioned are beyond the scope of this article, we will touch on transforming variables so they can be appropriately plotted.

➞ The Transformers

• Factor: Turn values with a fixed set of possible values that can be placed into a 'natural' order into ordered factors with distinct ranks/levels/tiers.

  • Returning to our batting order example, the four-hitter can only return the value of 4 for his position in the lineup without first needing to hit 1-3; always comes before the five-hitter; always follows the three-hitter.

  • Therefore, it makes sense to transform baseball players' batting order values into defined ordered factors.
    

• Log(arithm): Compress wide-ranging data that is heavily skewed in one direction and/or exemplifies exponential growth to make the data more linear and easier to visualize and interpret.

  • Player salaries make a strong case for log transformation because they often follow a heavily right-skewed pattern (superstar contracts that lie far above the majority).

  • The log tames outlier salaries and clarifies the relationship between players and their pay.
    

• Trimming: Remove or adjust extreme outliers distorting the distribution pattern.

  • The relationship between strikeouts and earned run average (ERA) for baseball's starting pitchers is a perfect example of the benefit of trimming.

  • The visualizations below show that merely trimming the years removed extreme outliers in strikeout and ERA that likely came from previous baseball eras.

    • Creating a more linear and predictable relationship between a modern-day starting pitcher's annual strikeout totals and ERA.

MLB Starting Pitcher's Strikeout by ERA (2000+)

MLB Starting Pitcher's Strikeout by ERA (2000+)

These are just a few transformation techniques that help you spot hidden insights, guide more decisive decisions, and communicate your findings with clarity.

THE BIG PICTURE

Distributions' real power lies in their ability to allow you to spot outliers, detect patterns, and uncover deeper insights.

But this is only possible if you use proper summary and transformation strategies to sharpen your focus, enable more precise understanding, and, most importantly, effectively share actionable takeaways.

For, once again, communication is the ultimate frontier in baseball analytics.

In our previous Choose Your Own Adventure, we considered four different perspectives on how an MLB GM may tackle real-world problems using t-tests and ANOVAs (data analytics tests for comparing averages between groups).

• Today, we will embark on a similar adventure, but we're moving from the front office to the dugout.
  

We are now data-driven MLB Managers looking to inform key decisions for tomorrow’s rapidly approaching 2025 season.

• Fortunately, analytics such as linear regression allow us to use existing real-world data to identify relationships between variables and predict future outcomes.

WHY IT MATTERS

In our data-driven era, MLB Managers can't rely purely on instincts.

We must utilize all the tools and resources available to us to decode past performance, project future outcomes, and build winning lineups.

➞ Player Projections: Predicting future outcomes based on previous results is a point of strength for linear regression — making it the perfect tool to project a player’s future performance based on stats from previous MLB seasons.

• If we want an accurate projection that takes into account the realistic nature of baseball (and life)—things like random chance, unpredictable fluctuations in skill, and age—we can't assume a player will simply copy and paste his past outcomes into the future.
  

  • A simple way to think about it: Outcomes = Talent + Randomness*
    

• Thankfully, we can stand on the shoulders of baseball analytics professionals who came before us, like Tom Tango, who found the best way to account for the inherent randomness and unpredictable external factors is with regression towards the mean.
  

  • In Tango's The Book, he clarifies that to project a player's future performance, we need to adjust his past performance toward the average performance of similar players, not all players. While this helps us limit the effects of randomness on our projections, we still have things like age and league/park factors to consider
    

➞ Team Projections: Linear regressions actually fail to predict a team's win-loss record (winning percentage) since a one-unit change in Runs Scored or Runs Allowed does not lead to a consistent (linear) change in winning vs. losing.

• Bill James is credited with being the first to recognize the nonlinear nature between runs scored, runs allowed, and wins. Instead, he found that the relationship is best expressed by a math formula similar to one you likely learned as a tween — the Pythagorean expectation.

• Though not linear, this formula allows us to predict a team's season win total with just their total runs scored and runs allowed.
  

  • So, we can use the linear regression process from our player projections to determine projected runs added (or lost) by a player, then factor that into our team's runs scored/allowed for the Pythagorean expectation, giving us a projected winning percentage for the season.
    

pythagorean expectation formula:

➞ Aging Curve: Another common analytic technique of interest is projecting the effect of age. Once again, this is a nonlinear relationship since a player's performance does not change at a constant rate over time.

• There’s no shortage of debate on the best method for projecting the effect of player aging; nonetheless, it's generally accepted that humans can't physically perform as well at 36 as they can at 26 because they’re more likely to get hurt and perform worse even when they are healthy.
  

• Though no two players age exactly the same way, these consistent trends in aging need to be accounted for to properly project future production.
  

  • Below is an aging curve graph for OPS+ from 1988 to 2016 with clear peaks in the mid-20s (showing multiple aging curve methods; click here to read more).

The idea is to use an aging curve to project the effect age has on a player, factor that into our linear regression for player projection, determine projected runs added (or lost) by a player, then factor that into our team's runs scored/allowed for the Pythagorean expectation, giving us a projected winning percentage for the season.

ZOOMING IN

➡︎ Toronto Blue Jays

A major storyline this spring training was the reluctance of the Blue Jays to extend their homegrown star, Vladimir Guerrero Jr. It likely comes down to uncertainty about which version of him you can expect — the recent MVP-caliber player, the borderline replacement-level player, or the league-average player from 2022 (see stats here)?

• Research Question: What level of production can we expect from Vladdy in 2025?

• Prediction Process: Randomness is especially relevant to our Vladdy situation when we zoom out and realize that A) six years of stats is a relatively small sample size, and B) the inconsistency within those six years further muddies things.

  • Once again, we can lean on those who came before us by using existing, popular, advanced methods of projection (all of which are still linear regressions) such as ZiPS, Steamer, and Depth Charts (amongst others).

    • Depth Charts is FanGraphs' (FGDC) in-house projection system, an average of ZiPS and Steamer plus a bit of customization for playing time estimates — see Vladdy's 2025 Depth Charts projection in the green column below:

      

➡︎ Arizona Diamondbacks

One of the prized free agents this past off-season was fantastic starting pitcher Corbin Burnes, whom the D-backs locked in for the next six years with the richest contract in the franchise's history ($210 million).

Between his salary and impressive stats (see them here), one might think it's obvious as to who is the club's ace — but let's not forget the outstanding pitcher who's been the head of our pitching staff for the last 6 seasons, Zac Gallen (see stats here).

We can't let ephemeral factors like who makes more money sway our decision as an analytically minded Manager.

• Research Question: Will Burnes or Gallen, as our number one starter, lead to more team wins in 2025?

• Prediction Process: Building on the player projection process, with a bit of extra work from our analytics team, we can project both pitchers' 2025 total runs allowed using the anticipated matchups of our ace.

  • The two separate scenarios will likely provide us with different total runs allowed for our ace matchups. We then plug these into two separate Pythagorean expectations to compare possible 2025 winning percentages.

➡︎ Pittsburgh Pirates

It's been a while since Pirates fans had a ball club to get excited about, but things seem to be improving in recent years as young, exciting, uber-talented ballplayers make the big team — Exhibit A) Oneil Cruz.

Though his talent is undeniable, questions remain about how to get the most out of the budding star, as shown by our decision to move him from shortstop to center field for the last 23 games of the 2024 season and the 2025 season.

• Research Question: What effect will Cruz's position change have on his value in 2025?

• Prediction Process: Positional adjustment is an essential part of player evaluation (and comparison) that we have yet to discuss. In short, it would be foolish to compare the defensive value of a shortstop to that of a right fielder because they field a totally different kind of batted ball at a totally different rate.

  • So, we correct for the inherent difficulty of playing some positions compared to others, adding runs for players at more demanding positions while subtracting runs for players at easier positions.

    • The shortstop adjustment is +7.5 runs per full season, while the right field adjustment is -7.5 runs per full season — see FanGraph's full positional adjustment chart below:
      

➡︎ Kansas City Royals

The Royals are an exciting, up-and-coming team in the big leagues, largely thanks to their 24-year-old phenom shortstop, Bobby Witt Jr., who seemingly has no ceiling as he inches closer to the title of undisputed best player in the game.

Witt Jr. and the Royals' long-term future seems bright (2024-2034, 11 years, $288.778M), but the fact remains that the backbone of our team is Salvador Perez — as has been the case for the last 13 years (see stats here).

His longevity, consistency, and offensive production as a catcher are nearly unprecedented. But as we enter his age-35 season, it's our duty as the Manager to always consider what's in the best interest of the team, not any one player.

Thus, we can’t assume Salvy will be the first person in history to defeat Father Time.

• Research Question: Could Salvy's age start to impact his value to the club in 2025?
  

• Prediction Process: More straightforward than our D-backs and Buccos processes, we're simply looking to apply our preferred aging curve method to Salvy's MLB body of work (and regression toward the mean) to get an idea of how many runs and wins Salvy projects to add to the team despite moving into the age seasons often hallmarked by rapid decline.
  

  • You should notice that this is essentially just the player projection process described earlier. In fact, all of the previously noted popular player projection models (see Blue Jays section) apply their own methodology to account for the effects of age.
    

    • Our in-house analytics department likely has a proprietary preferred method for projecting the aging curve for Royals players. But, as the Manager, we have to weigh not only the tangibles but also the intangibles of Salvy as a leader and irreplaceable presence in our clubhouse culture.

THE BIG PICTURE

Similar to our MLB GM perspective, relying solely on the guesswork and instincts of "baseball lifers" is a relic of the past.

Managers today must include analytics in their decision-making toolbox, ignoring the power of something like linear regression and its ability to predict future outcomes will likely come with significant losses.

Happy Opening Day tomorrow — here’s to a great 2025 season of baseball!

1. New York Yankees: After a whirlwind offseason, the Yankees insist they're in a better position to win compared to last year following their bevy of signings and trades that project a more complete and all-around better roster than 2024.

   1. However, a significant question remains about who will play third base. So, let's explore an analysis a Yankees GM might conduct when considering players to bring in to fill their needs at third base.
      

2. Boston Red Sox: Speaking of third base, the Red Sox made a surprising move this offseason when they signed star third baseman Alex Bregman to allegedly play second base.

   1. This quickly caused a stir in Red Sox Spring Training, as team officials and media members hinted that the plan was not 100% set on Bregman playing second base.

   2. Meanwhile, current star third baseman Rafael Devers insisted he wouldn’t be switching positions. Thus, we will explore an analysis a Red Sox GM might conduct when considering which is the better option at third base.

3. Miami Marlins: Starting with the fish, we'll then shift our focus from roster construction and winning to roster construction and revenue.

   1. A playoff participant just two years ago, the Marlins have elected to go with an entire team of replacement-level players (0 WAR - Wins Above Replacement) — at least once they trade away their final valuable pitchers.

   2. Why don't we explore an analysis a Marlins GM might conduct when considering this approach to building a team?

4. Los Angeles Dodgers: In a seemingly alternate universe lives the 2025 Dodgers — putting the entire league to shame as they spare no expense on players.

   1. Going all-in on a roster stacked with stars sure seems fun (MLB The Show, anyone?), but we can make a down-to-earth analysis a Dodgers GM might conduct on whether there's actual viability in spending literally billions of dollars for on-field talent.
      

WHY IT MATTERS

Neither winning nor making money in the MLB is guesswork anymore.

Proper research questions and statistical tests (like t-tests and ANOVAs) give GMs the evidence to determine whether roster moves genuinely move the needle on performance and profits.

➡︎ Yankees

Given Yankee Stadium's lefty-friendly dimensions, a responsible Yankees GM should always be considering left-handed hitters as our offensive preference. Especially when our current power production predominantly comes from our monster righties, Judge and Stanton.

• Research Question: Does signing a left-handed third baseman yield a significantly different impact on the Yankees' winning percentage compared to a right-handed third baseman, given Yankee Stadium's park effects?

• T-Test: With two distinct groups to compare, left-handed and right-handed third basemen, a t-test works perfectly. It's explicitly designed to compare the difference in mean between the two groups. By running a t-test on the two groups (lefty vs. righty 3B), we can see if a difference in projected production adjusted for Yankee Stadium indeed translates to more wins.

  • Data Setup: Gather MLB third-base candidates' annual offensive metrics (e.g., wRC+, OPS) over multiple seasons, adjusting to Yankee Stadium. Label each candidate "Lefty" or "Righty." Then, estimate how each candidate's production translates into additional runs and wins created for the Yankees.

  • Model: Compare winPct ~ 3BHandedness (winning percentage for two groups: lefty vs. righty).

➡︎ Red Sox

As the Red Sox GM, we need to at least consider the possibility that our team could be better off from a winning percentage perspective with a defense-first third baseman like Bregman or an offense-first third baseman like Devers.

• Research Question: Does a defense-heavy third baseman produce a significantly different win rate than a bat-first third baseman for the Red Sox?

• T-Test: Once again, we compare two distinct groups (defense-first and offense-first third basemen), so a t-test is the best way to analyze run prevention (defensive metrics) vs. run creation (offensive metrics) makes a meaningful difference in projected runs scored and runs allowed.

  • This would allow you to calculate and compare the Red Sox's projected winning percentage for the different scenarios using Bill James' Pythagorean Winning Percentage — read more about it here.

    • Data Setup: Compile defensive metrics (DRS, OAA) and offensive metrics (wOBA, HR, wRC+) for available 3B. Split them into "Defense-First" or "Offense-First" categories based on thresholds.

    • Model: Compare winPct ~ 3BType (winning percentage for two groups: defense-first vs. offense-first).

  

Roster Construction & Revenue

➡︎ Marlins

It's no secret that our Marlins leadership group has decided that yet another rebuild is the best option for the organization's future. As the GM leading the rebuild, it is up to us to determine the best approach that balances reconstructing the roster while not intentionally throwing away entire seasons with zero intention of being competitive and no regard for our fans (or, you would think).

• Research Question: Does a roster of replacement-level players, instead of a standard mix or even league-average players, get penalized for lack of competitiveness by way of generating significantly less revenue than other teams?

• ANOVA: Since we need to compare the differences in averages for one factor (revenue) at multiple levels (roster types), an ANOVA is a better fit for our testing. Allowing us to compare the effects of our three roster types (replacement level, standard mix, and league average) on revenue generation.

  • Data Setup: Look at team-level revenue (ticket sales, local TV ratings, merchandise) from past seasons while categorizing each team's roster into "Replacement," "League Average," or "Standard Mix" based on team makeup of player all-star appearances, WAR > 3, and salary.

  • Model: Compare Revenue ~ RosterType (revenue for three roster types: Replacement, Standard, Average).
    

    
    

➡︎ Dodgers

An even worst-kept secret across the MLB landscape is that our Dodger leadership group has no ceiling on our budget for player acquisition. However, that doesn't inherently mean we should look to load up on as many stars as possible. In our proper due diligence as the GM designing the organization's path forward, we should determine if investing heavily in talent directly boosts revenue as well.

• Research Question: "Does going 'all-in' on a superstar roster produce significantly higher revenue compared to more roster approaches of a standard mix, league-average players, or replacement-level players?

• ANOVA: Now comparing four different groups around one factor, an ANOVA is still our best-fitting test. Assessing the average revenue generated by each of our roster types can inform us whether there is any downside to our all-in strategy that we've been overlooking.

  • Data Setup: Collect annual revenue data for teams with different roster compositions (star-laden, league-average, replacement-level, standard mix). Classify them by team makeup of player, all-star appearances, WAR > 3, salary, and team total WAR.

  • Model: Revenue ~ RosterType (revenue for four roster types: Stars, Replacement, Standard, Average).

THE BOTTOM LINE

These four GM scenarios highlight how analytics—through t-tests and ANOVAs—can guide real decisions on the field and in the books.

But I hope to have also highlighted a broader competitive backdrop: are some teams realizing they can forget about winning and just cashing in on revenue sharing?

• MLB super-agent Scott Boras says it's time to leverage the data-driven methods used across the game to develop a new team competitiveness metric that holds owners accountable for using revenues to build competitive rosters.

• Allowing the fans and the entire baseball world to see who's truly playing to win and who's gaming the system.

GO DEEPER…

Watch Boras discuss this issue in this recent clip from Foul Territory:

Seymour Sarason, the father of community psychology, unknowingly foreshadowed this divide in his 1971 book The Culture of School and the Problem of Change, stating:

• Outsiders tend to underestimate the complexity of the culture that they want to change.

• Insiders instinctively resist change.

• Creating a mutual ignorance that leads to a polarizing relationship

Sound familiar? Like jocks being skeptical of nerds who "never played the game."

But Sarason did find an approach for mending such a relationship:

Introducing an ideal go-between ⎯ someone embedded in the community and culture that the outsiders wish to change.

WHY IT MATTERS:

Fast-forward to 2023.

MIT researcher Adam Berinsky found that partisan groups tend to find one kind of messenger particularly credible:

the unlikely source.

In the baseball world, there is no source more unlikely than someone versed in advanced analytics who also spent time in uniform.

Former San Diego Padres Manager Andy Green plainly outlines the power of a "baseball guy" who’s also fluent in data:

Unless the message gets translated by a guy who speaks both languages [analytics and on-field baseball], it usually falls on deaf ears.

ZOOMING IN:

In true poetic baseball fashion, the power of combining Sarason's ideal go-between and Berinksy's unlikely source can be seen in a head-to-head matchup of the game's most historical rivalry: Boston Red Sox vs. New York Yankees.

➡︎ Red Sox | A Catalyst for Success

In 2015, the Red Sox hired Brian Bannister, a former MLB pitcher who retired in 2011, as a pro scout and analyst.

• During Bannister's MLB career, he was documented as possibly the first active MLB player to openly adopt and incorporate the advanced statistics (sabermetrics) taking over baseball into his game.

That same year, Bannister crossed paths with aging big league pitcher Rich Hill.

• At 35, Hill was finding it increasingly difficult to stick with an organization — making him the perfect amount of desperate to listen to someone like Bannister, who had an unconventional idea backed by spin rate data for a way Hill could reinvent himself as a valuable MLB pitcher.
  
  

  ---

Long story short: it worked.

• After a couple of mid-August meetings with Bannister, Hill was starting for the Red Sox by early September.

Hill's first MLB start since 2009 began an epic four-game run to end the 2015 season:

• Sep. 13 at Tamp Bay Rays: 7 innings, 10 strikeouts, 0 runs.
  

• Sep. 20 at Toronto Blue Jays: 7 innings, 10 strikeouts, 3 runs.
  

• Sep. 25 vs. Baltimore Orioles: 9 innings, 10 strikeouts, 0 runs.
  

• Oct. 1 at New York Yankees: 6 innings, 6 strikeouts, 2 runs.

The following offseason, just a few months removed from meeting Bannister, Rich Hill signed a one-year contract with A's that topped his career total earnings to that point.

• He then inked a three-year, $48 million deal with the Dodgers following the 2016 season.

Hill credits Bannister as the catalyst for late-career success:

[he] opened my eyes to the creative side of pitching...that was really the first time that I had heard anything about that.

• The fact that this message came from someone who had been in his shoes made all the difference.

The catalyst effect from an ideal go-between/unlikely source combo like Bannister wasn't limited to individual player success.

• By 2018, Bannister was vice president of pitching development and assistant pitching coach for the Red Sox as they defeated the Dodgers (and Rich Hill) in the World Series that year.

➡︎ Yankees | Lost in Translation

The 2023 Yankees recorded their worst season since 1992.

• General manager Brian Cashman described the season as a "disaster," and the Yankees' analytics department became a central topic of discussion.
  

📌 Watch Brian Cashman describe the 2023 Yankees' "disaster" season.

This wasn't a new phenomenon.

• Previous accusations suggested the team's heavy reliance on the analytics department was directly responsible for a string of questionable acquisitions that backfired.

Things started to boil over following the "disaster" season when Aaron Judge spoke publicly about the team's analytics.

• Was there a problem with the actual number being provided?
  

Not according to the Yankees' captain:

BIG PICTURE:

There's an engineering term for the disproportionate difficulty of the final step in delivering network services to an end-user: the last mile.

While building a network can be relatively easy, the challenge comes at the very end when trying to get into houses and offices for people to actually use.

Baseball's last mile is the gap between the front office "nerds" and the dugout "jocks."

The ultimate frontier for baseball is communication.