Player development / Development Tracker
Who is getting better, who is slipping, and what should each player work on?
Every Pittsburgh player with pitch tracking (MLB, Triple-A Indianapolis and Bradenton in the Florida State League), checked month by month. A
change makes this list only if it’s large enough to matter and clear of noise (about 3 standard errors). A 2025 → 2026
change in a rate the strike zone moves (zone rate, chase, whiffs, walks, strikeouts) is measured net of the league’s own change, because 2026’s ABS zone
moved them league-wide; velocity, bat speed and exit velocity are raw changes, and Stuff and Location are graded against each season’s league. Each player then gets a short plan
built from the models: what to fix, a target that beats what regression alone would deliver, and a check timed to his sample. Only Triple-A and the
FSL have public tracking; High-A Greensboro, Double-A Altoona and the complex leagues don’t, so those players aren’t here.
Players tracked 159 MLB 47 · Triple-A 68 · Bradenton 95 (46 at more than one level)
Clear improvements 6 In-season and 2025 → 2026 (zone-dependent rates net of the league)
Clear declines 20 Players staying in 2027; 4 more from players leaving
What changed
Sorted by strength of evidence. “net” = measured against the league’s own 2025 → 2026 change. Hover or tap a row for the samples behind it.
30 changes
- ▼ Paul Skenes MLB fastball velocity 98.2 mph → 96.8 mph −1.4 mph 2025 → 2026
- ▼ Kirby Yates MLBleaving (free agent) fastball velocity 92.8 mph → 90.8 mph −2.0 mph 2025 → 2026
- ▼ Noah Davis Triple-Aleaving (likely minor-league free agent (estimate)) Stuff 92 → 83 −9 2025 → 2026
- ▼ Carmen Mlodzinski MLB Stuff(−8 by a model that never saw him) 103 → 94 −9 2025 → 2026
- ▼ Paul Skenes MLB Stuff(−11 by a model that never saw him) 114 → 104 −10 2025 → 2026
- ▲ Adolfo Oviedo Bradenton four-seamer velocity 91.0 mph → 93.3 mph +2.3 mph 2025 → 2026
- ▼ Noah Davis Triple-Aleaving (likely minor-league free agent (estimate)) fastball velocity 94.6 mph → 93.0 mph −1.6 mph 2025 → 2026
- ▲ Adolfo Oviedo Bradenton cutter velocity (primary fastball) 86.5 mph → 87.9 mph +1.4 mph 2025 → 2026
- ▼ Carmen Mlodzinski MLB fastball velocity 96.1 mph → 95.0 mph −1.0 mph 2025 → 2026
- ▼ Carmen Mlodzinski MLB sinker velocity 95.9 mph → 94.4 mph −1.5 mph 2025 → 2026
- ▲ Evan Sisk MLB sinker velocity (primary fastball) 90.0 mph → 91.1 mph +1.2 mph 2025 → 2026
- ▼ Noah Davis Triple-Aleaving (likely minor-league free agent (estimate)) sinker velocity 94.6 mph → 93.1 mph −1.5 mph 2025 → 2026
- ▲ Brandon Cain Bradenton fastball velocity 93.3 mph → 94.6 mph +1.3 mph since Jun 9
- ▼ Michael Darrell-Hicks Triple-A sinker velocity (primary fastball) 95.0 mph → 94.1 mph −0.9 mph 2025 → 2026
- ▼ Paul Skenes MLB sinker velocity 97.6 mph → 96.6 mph −1.0 mph 2025 → 2026
- ▼ Isaac Mattson Triple-A Stuff 101 → 91 −10 2025 → 2026
- ▼ Noah Murdock Triple-A Stuff 97 → 89 −8 2025 → 2026
- ▼ Ryan O'Hearn MLB bat speed 71.1 mph → 70.0 mph −1.1 mph 2025 → 2026
- ▼ Mason Montgomery MLB Stuff(−7 by a model that never saw him) 117 → 110 −7 2025 → 2026
- ▲ Noah Murdock Triple-A cutter velocity 89.0 mph → 90.7 mph +1.7 mph 2025 → 2026
- ▼ Jared Triolo MLB bat speed 71.0 mph → 69.9 mph −1.1 mph 2025 → 2026
- ▼ Adolfo Oviedo Bradenton whiff rate 26% → 19% −10.0 ptsnet 2025 → 2026
- ▼ Michael Darrell-Hicks Triple-A Stuff 93 → 87 −6 2025 → 2026
- ▲ David Matoma Bradenton sinker velocity (primary fastball) 96.0 mph → 96.9 mph +0.9 mph 2025 → 2026
- ▼ Luke Weaver MLB cutter velocity 91.5 mph → 90.6 mph −1.0 mph 2025 → 2026
- ▼ Edgleen Perez Bradenton chase rate 15% → 22% +7.8 ptsnet 2025 → 2026
- ▼ Richard Ramirez Bradenton chase rate 26% → 37% +12.1 ptsnet 2025 → 2026
- ▼ Oneil Cruz MLB bat speed 77.2 mph → 75.1 mph −2.1 mph since Aug 17
- ▼ Oneil Cruz MLB bat speed 77.8 mph → 76.6 mph −1.2 mph 2025 → 2026
- ▼ Thomas Harrington Triple-A cutter velocity 88.4 mph → 86.9 mph −1.5 mph 2025 → 2026
Development plans
Goals come only from model outputs. Velocity and shape targets come from his own history; a new pitch appears only when it holds up under every model, and it’s an experiment; rate targets beat what regression to the mean would deliver with no plan. Sorted by 2027 role, then runs at stake where a goal has one (Stuff declines, pitch shapes, new pitches and the Hitting Lab’s swing decisions and contact; command and prospect goals carry none). Open a player for trends and the full plan.
Paul Skenes MLB 2027: rotation · ≈9.4 runs in play
- Velocity Recover the four-seamer velocity he showed in 2025
- Pitch shape Get the splitter back to its 2025 shape (2″ more drop)
- Diagnose Find the cause of the drop in bat speed
- Approach Attack hittable pitches
- Approach Swing less at pitches out of the zone
- Pitch shape Get the four-seamer back to its 2024 shape (2″ more ride)
- Health Medical clearance before any development work
- Contact Make more contact on strikes
- Diagnose Find the cause of the drop in bat speed
- Health Medical clearance before any development work
- Health Medical clearance before any development work
- Bat speed Get bat speed back to his best season
- Position Re-establish second base (the Winter Plan moves him from third base)
- Health check Velocity check before the throwing program
- Approach Swing less at pitches out of the zone
- Velocity Recover the four-seamer velocity he showed in 2025
- Pitch shape Get the sweeper back to its 2025 shape (7″ less drop and 4″ more arm-side run)
- Diagnose Find the cause of the drop in cutter velocity
- New pitch Try a sweeper (~83 mph, 3″ IVB, −15″ HB)
- Command Command first: start with the sweeper
- Bat speed Get bat speed back to his best season
- Command Command first: start with the curveball
- Health Medical clearance before any development work
- Diagnose Find the cause of the drop in Stuff
- Diagnose Find the cause of the drop in Stuff
- Diagnose Find the cause of the drop in cutter velocity
How sure
What the feed and the plans can and can’t tell, measured on past seasons.
- A clear late-season change is only partly real: about half of a velocity change holds into the next season, under a quarter of a Stuff change. Running this feed’s in-season rule on every big-league pitcher, a flagged change kept a median 52% (2024, 12 flags) and 48% (2025, 12) of its size into the next season for fastball velocity, and 34% and 6% for Stuff (each season on its own league scale). Read a late flag as a question, not a verdict.
- Targets beat regression to the mean. A player far from the median drifts back part of the way with no plan at all. Each rate target is his expected next season plus the step a quarter of similar players beat. Tested forward on big-leaguers (the rule fit on 2024 → 2025, scored on 2025 → 2026), 25% of 323 targeted players reached their target with no plan (walks 21%, chase 35%, zone contact 13%, hittable pitches taken 28%, platoon gap 28%). Prospect targets can only be set on 2025 → 2026 itself (minor-league tracking starts in 2025), so their quarter is by construction, not a test. Meeting a target is evidence the work took only against that baseline.
- Checks are timed to the sample. A goal is checked at the 2027 All-Star break only when the sample he had by the 2026 break would put the target two standard errors from today; otherwise at season’s end, or after the sample it needs (relievers: batters faced, not a date). A goal that would need more than half again his 2026 sample gets no check: it is a lean (14 of 32 rate goals). Each window needs 8+ games.
- Year-over-year changes in rates the strike zone moves (zone rate, chase, whiffs, contact, walks, strikeouts, xwOBA) are net of the league’s median change at that level (“net” on the list); velocity, bat speed, swing shape and exit velocity are raw changes; Stuff and Location are already centered on each season’s league.