Are the signals earning trust?

Performance

Assists carries the strongest edge (+23.1 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

Assists+23.1 pp n=274+32.5 pp n=16460%

“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 Judged on the engine running today: a superseded scorer’s serving share is not a fact about the current one.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 30 days · Featured17–289.5%n=190 void · 0 pending · avg score 100 · 7 slates

All qualifying108–5665.9%n=164

Featured17–289.5%n=19

Other qualifying91–5462.8%n=145

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

Assists89%n=19building sample100%n=2100%n=375%n=4100%n=1100%n=450%n=2100%n=3

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.

Score bandRecordHit rateSampleVoidsvs base

99–100Small sample17–289.5%n=190+71.2 pp base 18%

Over vs Under

Over17–289.5%n=19

Under0–0Not gradedn=0

MarketOverUnder

Assists89.5% n=19Not graded n=0

Edge finder

SegmentHit ratevs baseRecord30-day trend

Below minimum sample

AssistsSmall89.5% n=19+71.2 pp base 18%17–289% 30d

Consistency

Last 7

89.5%

17–2 n=19

Last 30

89.5%

17–2 n=19

Prev 30

100.0%

1–0 n=1

Season

90.0%

18–2 n=20

All time

90.0%

18–2 n=20

By month

MonthRecordHit ratevs base

2026-071–0100.0% n=1+56.7 pp base 43%

2026-0817–289.5% n=19+71.2 pp base 18%

Model versions

VersionActiveRecordHit ratevs base

WNBA Assists

wnba-pra-v42026-08-23→2026-08-2917–289.5% n=19+71.2 pp base 18%

1 earlier version2026-07-15→2026-07-161–0100.0% n=1+56.7 pp base 43%

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.