Are the signals earning trust?
Performance
Market
⚾ Batter
⚾ Pitcher
Points carries the strongest edge (+8.6 pts over base).
What's working
Every market's measured edge against how much of it we actually serve, ranked by lift. The tables further down read served predictions only, so a market that is good but never clears the floor is invisible there — and looks the same as one that does not work.
MarketAll predictionsServed (70+)Share served
Points+8.6 pp n=458+21.6 pp n=25756%
“Starved” means the market beats its base rate by 5+ points across everything it predicted, yet under a tenth of those ever clear the floor — the scale cannot reach, so the edge is never offered. Judged on the engine running today: a superseded scorer’s serving share is not a fact about the current one. The columns stay pooled across versions.
2026 season · Featured11–378.6%n=14
All qualifying165–9264.2%n=257
Featured11–378.6%n=14
Other qualifying154–8963.4%n=243
Deeper diagnostics
Market pulse
Recent slate hit rates compared with each market’s own average. The right edge shows the selected period and the latest three slates versus the prior three.
MarketPeriodAug28Aug27Aug24Aug23Jul15
🏀Points79%n=14building sample100%n=2100%n=10%n=3100%n=3100%n=5
Above own averageNear own averageBelow own averageFaded cells have fewer than 5 decisions
Does a higher score perform better?
Not enough graded props yet to judge whether higher scores perform better. The 99–100 band remains small.
Current scoring engine only — 5 predictions from superseded versions excluded. Pooling engines once made the top band look anti-predictive when it was not; per-version records are in the model table below.
Score bandRecordHit rateSampleVoidsvs base
99–100Small sample6–366.7%n=91+44.3 pp base 22%
Over vs Under
Over11–378.6%n=14
Under0–0Not gradedn=0
MarketOverUnder
Points78.6% n=14Not graded n=0
Edge finder
SegmentHit ratevs baseRecord30-day trend
Below minimum sample
PointsSmall78.6% n=14+49.0 pp base 30%11–379% 30d
Consistency
Last 7
66.7%
6–3 n=9
Last 30
66.7%
6–3 n=9
Prev 30
100.0%
5–0 n=5
Season
78.6%
11–3 n=14
All time
78.6%
11–3 n=14
By month
MonthRecordHit ratevs base
2026-075–0100.0% n=5+56.0 pp base 44%
2026-086–366.7% n=9+44.3 pp base 22%
Model versions
VersionActiveRecordHit ratevs base
WNBA Points
wnba-pra-v42026-08-23→2026-08-296–366.7% n=9+44.3 pp base 22%
1 earlier version2026-07-15→2026-07-155–0100.0% n=5+56.0 pp base 44%
All qualifying includes every published prediction scoring 70+. Featured is the main-page eight. Observed hit rates are not probabilities; prediction counts are correlated within the displayed slate count.