Model Health loading…
Window
loading…
What's working, what's broken
analyzing…
Daily hit rate trend
Market × Side If "Hits OVER" hits 55%, that's the bucket to lean into. If "HR OVER" hits 20%, kill it.
| Bucket | W–L | Hit % | ROI |
|---|---|---|---|
| loading… | |||
By Tier
| Tier | W–L | Hit % |
|---|---|---|
| — | ||
By Market
| Market | W–L | Hit % |
|---|---|---|
| — | ||
By Side
| Side | W–L | Hit % |
|---|---|---|
| — | ||
Calibration When the model says X% probability, does it actually hit X%?
loading…
By Park Which stadiums does the model handle best/worst?
| Matchup | W–L | Hit % |
|---|---|---|
| — | ||
Auto-refreshes daily as the grader settles new picks. Reads from player_props_graded
in Supabase. The whole point of this page: see what's working in less than 60s.