Validation
We measure whether the evaluation stayed true to the mission.
Project 80 tracks decision quality by category so we can improve without overreacting to one good or bad day.
What we measure.
Official Play Accuracy
Did official Playable calls deserve action and meet the limits?
Too Hard Accuracy
Did Too Hard correctly identify games that were noisy, volatile, or unclear?
Price Gate Performance
Did the gate save us, cost us, or reveal stale/conflicting markets?
Pass Discipline
Did we avoid bad favorite prices, thin edges, or unsupported markets?
Best Expression
Did we route the edge to the right market: ML, F5, total, team total, run line, or pass?
Rule Status
Is a rule validated, provisional, or still only a watchlist idea?