Are the signals earning trust?

Performance

Points carries the strongest edge (+14.1 pts over base). Down 2.7 points vs previous 7 days.

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+14.1 pp n=103+18.9 pp n=8280%

“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 Judged on the engine running today: a superseded scorer’s serving share is not a fact about the current one.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.

Last 7 days · All qualifying50–3261.0%n=8210 void · 0 pending · avg score 93 · 7 slates

All qualifying50–3261.0%n=82

Featured6–366.7%n=9

Other qualifying44–2960.3%n=73

61.0% hit rate, down 2.7 percentage points vs previous 7 days

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.

MarketPeriodAug29Aug28Aug27Aug26Aug25Aug24Aug23

Points61%n=82→ -1 pp29%n=774%n=1950%n=1060%n=1064%n=1155%n=1171%n=14

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 70–74, 75–79, 80–84, 85–89, 90–94, 95–98 bands remain small.

Score bandRecordHit rateSampleVoidsvs base

70–74Small sample3–442.9%n=70-2.2 pp base 45%

75–79Small sample1–0100.0%n=10+37.3 pp base 63%

80–84Small sample8–0100.0%n=86+40.8 pp base 59%

85–89Small sample1–0100.0%n=11+49.7 pp base 50%

90–94Small sample5–550.0%n=100+7.1 pp base 43%

95–98Small sample3–260.0%n=50+2.2 pp base 58%

99–10029–2158.0%n=503+22.1 pp base 36%

Over vs Under

Over50–3261.0%n=82

Under0–0Not gradedn=0

MarketOverUnder

Points61.0% n=82Not graded n=0

Edge finder

SegmentHit ratevs baseRecord30-day trend

Points61.0% n=82+18.3 pp base 43%50–3261% 30d

Consistency

Last 7

61.0%

50–32 n=82

Last 30

63.5%

160–92 n=252

Prev 30

100.0%

5–0 n=5

Season

64.2%

165–92 n=257

All time

64.2%

165–92 n=257

By month

MonthRecordHit ratevs base

2026-075–0100.0% n=5+56.0 pp base 44%

2026-08160–9263.5% n=252+20.5 pp base 43%

Model versions

VersionActiveRecordHit ratevs base

WNBA Points

wnba-pra-v42026-08-21→2026-08-2964–4658.2% n=110+16.4 pp base 42%

4 earlier versions2026-07-15→2026-08-21101–4668.7% n=147+24.8 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.