system doctor

← lab · at 2026-10-02T19:15:27.542189+00:00 · failed 2 · ran 10/2/2026

What this is: A one-off study — its numbers below are exactly what the run reported.

checks

check	detail	status
feed:api-football	last fetch 0.3h ago (max 8h)	ok
feed:kalshi	last fetch 5.2h ago (max 13h)	ok
feed:the-odds-api	last fetch 14.5h ago (max 26h)	ok
feed:espn	last fetch 14.7h ago (max 30h)	ok
odds:upcoming-24h	1/2 priced contests fresh (<14h); stale: São Paulo FC v Santos FC	FAIL
derived:metrics	last metric computed 13.3h ago (max 30h)	ok
derived:sims	last sim finished 12.9h ago (max 30h)	ok
results:finished-unscored	3 finished contests missing scores: TBD v Chicago White Sox, Atlanta Braves v Diamondbacks/Phillies, TBD v Chicago White Sox	FAIL
odds:multi-book	all upcoming competitions have >=3 fresh sportsbooks	ok
quota:api-football	30 requests today (plan ~150k/day)	ok
bets:stale-pending	0 pending >24h after all legs finished: none	ok

AUC	predictability: 0.50 = coin flip, ~0.70 = ceiling for sports
Importance	how much the model leans on this factor (permutation importance)
Direction	sign of the raw correlation with the outcome
Survives all eras	effect points the same way in every historical era

Spec · the reproducible recipe

{
  "kind": "doctor"
}