The Leadoff — Conventional Wisdom, Tested

What's Real and What's Noise in the Bullpen Ledger

A reliever's velocity repeats at 0.95 and his whiff rate at 0.63. His win-probability ledger repeats at 0.13 -- a coin flip -- and barely correlates with itself inside one season. The stuff is the skill; the scoreboard is the weather.

Relievers · 9 min read

Read the analysis →

Fig.1 — Called-strike rate by location
2025 baseline · inside → away

The rulebook box never moves. The shadow zone around it does — and it tightens when the batter has two strikes.

By the Numbers

Pitches analyzed

Seasons of Statcast

Umpires graded

Investigations published

The Investigations

All analysis →

Relievers

The Firemen the Save Stat Can't See

83% of the win probability relievers create happens outside save situations -- the save stat watches the smallest slice of the bullpen. We added up the wins relievers actually bank across 8,092 relief appearances, then made two independent methods fight over the names. Colin Holderman has been one of the most valuable relievers in baseball this year with zero saves; value and save credit correlate at just 0.22. We name the hidden firemen on win-probability magnitude, never on rank -- because at a half-season the order is noise.

June 30, 2026 · 9 min read

Fielding

The Fielder's Fingerprint: Two Gloves, Same OAA, Opposite Skills

Outs Above Average gives every fielder one number; it hides a direction. We split five seasons of directional OAA into the outs a fielder earns charging in versus covering deep -- his "tilt" -- and made two independent methods fight over it. The tilt is a real, repeatable skill (YoY r=0.28, ~0.29 at short, center, and left) that has nothing to do with sprint speed (r=0.04), and OAA averages it away: 37 same-position pairs in 2025 had matching OAA and opposite gloves. Lindor charges in, Baez covers deep -- same number, opposite craft. The honest limit: it's a fingerprint, not a forecast.

June 26, 2026 · 9 min read

Teams

A Bad Start Is Forever. A Good One Isn't.

Every July, fans ask which first-half teams are real. We pooled five full seasons (2021-2025, 150 team-seasons) and tested it out of sample. A team's quality is real (random-split r=0.74), but the first half predicts only about a THIRD of the second -- and almost all of that third is in the basement. A bad first half predicts the second at R2=0.33; a good one at just 0.07; the .500 muddle is noise. 82% of clearly-bad starts stayed buried vs 67% of good ones staying up -- gravity pulls down harder. The signal locks in by ~game 40 (Memorial Day); run differential and the plain record predict it best (fancy stats add nothing); bat speed is noise. Plus: the biggest frauds (2023 Angels 44-37 -> 29-52) and what the rule says about 2026.

June 21, 2026 · 8 min read

The Long Read

The Jump Tax: The Steal Is a Decision, Not a Physique

The slowest runners in baseball steal at 96%. Not with speed, not even with a big lead, but with an ambush. The jump and the legs are real, repeatable skills -- and neither tells you who banks the value. That is the decision, and it barely repeats.

The findingBaserunning

r = 0.87

how tightly a base-stealer's run value tracks his success rate -- value is conversion, not tools. The jump and speed are independent and highly repeatable, but neither out-predicts the other for value, and single-season steal value is ~44% small-sample noise.

Read the analysis →

Explore the Data

Game Reports Umpire Profiles ABS Challenges Pitcher Profiles Hitter Profiles Zone Explorer Count Calculator Open Explore →

One investigation a week. No hot takes, just the receipts.

CalledThird Weekly lands in your inbox during the season. Under 500 words — we respect your time.

Or follow on Twitter/X · Bluesky