How the hockey grades have done · 2026-09-29 – 2026-11-16
The read. The bands shipped as cut from the fit window; the one 2025-26 holdout look is spent and its meaning is pre-committed: a band may be called edge-shaped only on a flat-stake return above zero at 30 or more settled leans. This page applies the NFL lane's verdict table to the same four reads — return at the price, return at the close, beat-the-close on movers, the model's mean P(lean side) beside the hit rate — at the same 30-lean first look. The holdout said: moneyline A 75% on 12 (noise), B 68% hits and a 1.6% loss on 142; puck line A lost 15.6% on 177 and B 6.4% on 211.
Columns. RETURN is the flat-stake return per $1 at the price the card showed. AT CLOSE is the same result priced at the captured close. BEAT is how often the lean's side got a better number than the close, counted over the leans whose line moved. MODEL is the model's mean chance on the lean's side, to hold against WIN %.
OVERALL
| SEGMENT | N | W | L | PUSH | WIN % | RETURN | AT CLOSE | BEAT | MODEL | NOTES |
|---|
| Every settled grade | 996 | 481 | 499 | 16 | 49.1% | -2.6% | -3.0% | 47% of 180 | 50.1% | |
| of which graded — a lean | 178 | 98 | 80 | 0 | 55.1% | +10.8% | +10.7% | 66% of 35 | 51.7% | |
| of which no lean — coverage | 818 | 383 | 419 | 16 | 47.8% | -5.5% | -6.0% | 42% of 145 | 49.7% | |
The two indented rows sum to the first. Graded is the pick record — every card where the model claimed an edge. No lean is every card where it agreed with the market; settlement still writes a W or an L against its nominal side, which is bookkeeping, not a call.
BY MARKET AND BAND· the pick record, crossed · the verdict reads the A leans
| SEGMENT | N | W | L | PUSH | WIN % | RETURN | AT CLOSE | BEAT | MODEL | NOTES |
|---|
| Moneyline · A | 32 | 26 | 6 | 0 | 81.2% | +20.1% | +19.6% | 83% of 12 | 70.6% | edge-shaped |
| Moneyline · B | 52 | 36 | 16 | 0 | 69.2% | +5.6% | +5.4% | 62% of 21 | 64.5% | |
| Moneyline · all graded picks | 84 | 62 | 22 | 0 | 73.8% | +11.1% | +10.8% | 70% of 33 | 66.8% | |
| Puck line · A | 39 | 17 | 22 | 0 | 43.6% | +24.2% | +24.1% | — | 38.5% | noise |
| Puck line · B | 55 | 19 | 36 | 0 | 34.5% | +1.0% | +0.9% | 0% of 2 | 38.0% | |
| Puck line · all graded picks | 94 | 36 | 58 | 0 | 38.3% | +10.6% | +10.5% | 0% of 2 | 38.2% | |
Verdicts. Moneyline A: edge-shaped · Puck line A: noise
BY GRADING MODEL· which model made the letter · old letters kept apart
| SEGMENT | N | W | L | PUSH | WIN % | RETURN | AT CLOSE | BEAT | MODEL | NOTES |
|---|
| Moneyline · market blend (the old letter) | 73 | 56 | 17 | 0 | 76.7% | +10.8% | +10.6% | 68% of 25 | 69.2% | |
| Moneyline · lineup driven | 11 | 6 | 5 | 0 | 54.5% | +13.0% | +12.0% | 75% of 8 | 50.8% | insufficient sample |
| Puck line · all factors | 17 | 7 | 10 | 0 | 41.2% | +14.1% | +14.0% | — | 43.6% | insufficient sample |
| Puck line · market blend (the old letter) | 77 | 29 | 48 | 0 | 37.7% | +9.9% | +9.8% | 0% of 2 | 37.0% | |
The puck line has been graded on every factor since 2026-11-03 — the letter is how far the full picture sits from the market, the way the baseball book grades. Letters made before that were the market blend's; they keep their own row. On the moneyline, lineup-driven letters fill games the blend gives no lean.
NO LEAN — COVERAGE· the market's right; not a track record
| SEGMENT | N | W | L | PUSH | WIN % | RETURN | AT CLOSE | BEAT | MODEL | NOTES |
|---|
| Total | 332 | 147 | 169 | 16 | 46.5% | -5.9% | -6.5% | 0% of 18 | 48.9% | |
| Moneyline | 248 | 139 | 109 | 0 | 56.0% | -2.1% | -2.2% | 49% of 122 | 56.3% | |
| Puck line | 238 | 97 | 141 | 0 | 40.8% | -8.5% | -9.4% | 20% of 5 | 43.9% | |
BY DAY· the graded leans per slate · open a day for its games