Velo Watch
Every start's fastball velocity against the pitcher's own normal range, so a real drop stands out, and you can see whether it came back.
Builds on Answers for Coach
- Coach questions
- 3
- Asked on air
- 17×
- Top question
- Q-07
Coach questions it could answer
The playbook →- Are injuries rising, and does workload explain it?
Mark injured-list stints on the same timeline to test whether velocity dips came before injuries, and whether heavy recent workloads came before the dips.
- What drives in-season velocity swings, and does velocity come back?
Plot every start's average fastball against the pitcher's own baseline band. Dips outside the band are flagged, and the chart shows how often velocity came back and how long it took.
- How much do starters vary from start to start?
The band itself answers this one. Its width is how much a starter's stuff normally varies from one start to the next.
Data it needs
The data lake →Everything it needs is on hand, so it could be built today.
Could use later: Injured list, Box scores, Rest and travel
The question
When a pitcher’s velocity is down, the hosts want to know whether it’s normal noise, early-season build-up, fatigue, or a warning sign, and whether it comes back.
How it would work
- Each start’s average fastball velocity as a dot, against a band for that pitcher’s normal range from the past year.
- Flags when a start falls outside the band, and a running count of how many in a row.
- Optional layers: pitch count and days of rest before each start, and injured-list stints.
First version
- Every qualified starter this season, one chart each, with the band and flags.
- A league page listing current outliers.
- Season-start ramp-up shown separately so April isn’t flagged as a crisis.
The data
Statcast pitch-level velocity, already cached for 2022–26 by Answers for Coach, MLB Stats API game logs for pitch counts and rest, and public injured-list transactions.
What it needs
- A sensible band. Velocity rises through April and can fade late, so the normal range has to move with the calendar.
- Careful claims about injuries. A dip before an injury is suggestive, not proof; the page should show how often dips happen without one, too.
Status
An idea on the launchpad. Nothing is built yet.