Betting Benchmarks

The record · 2026

How the model is doing

This season our plays are 20–19 (-0.9 units) and our side in every game is 52–37 (58.4% against the spread), through 6 graded weeks. Before this season, the same model went 164–111–6 (59.6%) on plays over 2023–2025, tested as if live. Every pick is published before kickoff and never edited.

Regenerated 2026-10-21 · one unit per play at −110 · not betting advice

This season · 2026

Plays 20–19 (-0.9u) · every game 52–37

Two records, both settled on the final score against the line we published. Plays are the sides that cleared our 53% bar at the opening line, one unit each at −110. Every game is the side our number favors in all of the week’s games, the pick’em view. Nothing is edited after publish.

Plays · this season20–1951.3% · -0.9u

Sides that cleared the bar, one unit each. Break-even at −110 is 52.4%.

Every game · this season52–3758.4% ATS

Our side in every game, the confidence-ranked pick'em list.

Weeks6graded · 1 pending

Week 7 publishes Tuesday and is graded after Monday night.

WeekPublishedPlaysW–LUnitsEvery gameATSStatus
Week 7Oct 21, 6:57 PM ET7————pending
Week 6Oct 13, 7:16 PM ET51–3-2.1u5–838.5%graded
Week 5Oct 6, 11:21 PM ET63–3-0.3u10–471.4%graded
Week 4Sep 29, 11:26 PM ET63–3-0.3u9–660.0%graded
Week 3Sep 22, 8:24 PM ET116–4+1.5u9–660.0%graded
Week 2Sep 15, 6:42 PM ET84–4-0.4u9–756.3%graded
Week 1Sep 4, 11:17 AM ET53–2+0.7u10–662.5%graded
▸Week 7· 7 plays · 14 gamespublished Oct 21, 6:57 PM ET · pending
  • NE +3.0at CHIplay57.0% to coverpending
  • MIN -6.5vs INDplay54.8% to coverpending
  • MIA +4.5at NYJplay54.5% to coverpending
  • PHI +4.5vs DALplay54.3% to coverpending
  • NO +3.5vs PITplay54.0% to coverpending
  • CLE -1.5at TENplay53.6% to coverpending
  • LV +4.0vs LAplay53.1% to coverpending
  • ATL +3.0vs SF52.4% to coverpending
  • DEN -4.0at ARI52.3% to coverpending
  • KC +2.5at SEA51.0% to coverpending
  • NYG +10.5at HOU51.0% to coverpending
  • TB +3.0at CAR50.4% to coverpending
  • CIN -1.0at BAL50.2% to coverpending
  • GB +3.0at DET50.1% to coverpending
▸Week 6· 5 plays · 14 gamesplays 1–3 · -2.1u · every game 5–8
  • NE -7.0vs NYJplay57.9% to coverpush
  • ATL +3.0vs CHIplay56.3% to coverloss
  • CLE +4.5vs BALplay54.4% to coverwin
  • JAX -2.5vs HOUplay54.3% to coverloss
  • LA -12.5vs ARIplay53.1% to coverloss
  • KC -8.5vs LAC52.6% to coverwin
  • DEN +3.0vs SEA52.3% to coverwin
  • PHI -1.0vs CAR52.2% to coverloss
  • TB +2.5vs PIT51.9% to coverloss
  • NYG +1.0vs NO51.3% to coverloss
  • SF -7.5vs WAS51.2% to coverloss
  • GB -1.0vs DAL50.5% to coverwin
  • BUF -5.5at LV50.4% to coverloss
  • TEN +7.0at IND50.3% to coverwin
▸Week 5· 6 plays · 15 gamesplays 3–3 · -0.3u · every game 10–4
  • BUF +3.0at LAplay55.3% to coverwin
  • PIT -2.5vs INDplay54.7% to coverwin
  • LAC +3.5vs DENplay54.4% to coverloss
  • TEN +7.5vs HOUplay53.9% to coverloss
  • CLE +2.0at NYJplay53.8% to coverloss
  • NE -3.5vs LVplay53.7% to coverwin
  • TB +8.0at DAL52.7% to coverwin
  • MIA +7.0vs CIN52.4% to coverpush
  • PHI +7.5at JAX52.3% to coverwin
  • MIN -2.0at NO52.3% to coverwin
  • ARI +5.0vs DET51.8% to coverloss
  • NYG +3.5at WAS51.8% to coverwin
  • SF +3.0at SEA51.6% to coverwin
  • GB +3.0vs CHI51.3% to coverwin
  • BAL +3.5at ATL51.0% to coverwin
▸Week 4· 6 plays · 16 gamesplays 3–3 · -0.3u · every game 9–6
  • TB +4.0vs GBplay59.2% to coverwin
  • NYG +1.5vs ARIplay58.2% to coverwin
  • WAS +3.5vs INDplay57.2% to coverloss
  • PHI +3.0vs LAplay55.3% to coverloss
  • CHI -3.0vs NYJplay54.7% to coverwin
  • MIN -11.0vs MIAplay53.1% to coverloss
  • CAR +3.5vs DET52.8% to coverwin
  • NO -2.5vs ATL52.5% to coverloss
  • SF -3.0vs DEN52.3% to coverwin
  • HOU -2.5vs DAL52.3% to coverloss
  • LAC +7.0at SEA51.8% to coverpush
  • BAL -11.5vs TEN51.8% to coverloss
  • NE +6.5at BUF51.6% to coverwin
  • CLE +2.5vs PIT51.4% to coverwin
  • LV +4.5vs KC51.2% to coverwin
  • JAX +2.5at CIN50.9% to coverwin
▸Week 3· 11 plays · 16 gamesplays 6–4 · +1.5u · every game 9–6
  • CLE +3.0vs CARplay56.7% to coverwin
  • NYG -3.0vs TENplay55.0% to coverwin
  • PIT +3.5vs CINplay54.7% to coverwin
  • WAS +7.0vs SEAplay54.7% to coverwin
  • SF -8.5vs ARIplay54.6% to coverloss
  • NO -3.0vs LVplay54.5% to coverloss
  • CHI +4.5vs PHIplay54.3% to coverwin
  • BUF -7.0vs LACplay54.0% to coverwin
  • HOU -2.5at INDplay53.4% to coverloss
  • DAL +3.0vs BALplay53.0% to coverpush
  • MIA +11.5vs KCplay53.0% to coverloss
  • ATL +6.0at GB52.1% to coverwin
  • DET -6.5vs NYJ51.9% to coverwin
  • DEN +2.5vs LA51.8% to coverwin
  • TB +1.5vs MIN51.7% to coverloss
  • NE +3.0at JAX51.6% to coverloss
▸Week 2· 8 plays · 16 gamesplays 4–4 · -0.4u · every game 9–7
  • SF -13.5vs MIAplay57.6% to coverwin
  • NE -5.0vs PITplay55.1% to coverwin
  • KC -6.5vs INDplay54.9% to coverloss
  • ATL +2.5vs CARplay54.6% to coverloss
  • WAS +3.5at DALplay54.3% to coverloss
  • MIN +5.5at CHIplay53.5% to coverwin
  • SEA -4.0at ARIplay53.4% to coverwin
  • HOU -2.5vs CINplay53.4% to coverloss
  • CLE +8.5at TB52.7% to coverwin
  • NYJ +3.5vs GB52.5% to coverwin
  • BUF -4.5vs DET52.5% to coverwin
  • NO +8.5at BAL52.1% to coverwin
  • JAX +2.5at DEN51.6% to coverloss
  • NYG +7.0at LA51.6% to coverloss
  • TEN +7.0vs PHI51.5% to coverwin
  • LAC -6.5vs LV50.1% to coverloss
▸Week 1· 5 plays · 16 gamesplays 3–2 · +0.7u · every game 10–6
  • NYG +3.0vs DALplay58.0% to coverwin
  • TEN -1.5vs NYJplay55.3% to coverloss
  • MIN -1.5vs GBplay54.8% to coverwin
  • MIA +3.5at LVplay54.4% to coverloss
  • NO +7.0at DETplay53.9% to coverwin
  • JAX -8.0vs CLE52.1% to coverwin
  • SF +3.5at LA51.9% to coverwin
  • NE +3.5at SEA51.8% to coverwin
  • IND +3.5vs BAL51.6% to coverloss
  • PHI -5.0vs WAS51.5% to coverloss
  • LAC -10.0vs ARI51.3% to coverloss
  • PIT -3.0vs ATL50.6% to coverwin
  • CHI -2.5at CAR50.4% to coverwin
  • KC -2.5vs DEN50.3% to coverwin
  • CIN -4.0vs TB50.2% to coverwin
  • HOU +1.0vs BUF50.0% to coverloss

Before this season · 2023–2025

59.6% on plays, 53.8% on every game

The same model and the same bar, run over 2023–2025 as if live: each week’s picks use only information available before kickoff, and the probabilities were fit only on earlier seasons. 275 plays went 164–111–6 for +41.9u.

Plays · 2023–202559.6%164–111–6 · +41.9u

275 plays at one unit each. +13.9% return on the money risked.

Every game · 2023–202553.8%451–388–14 · +24.2u

Our side in every game, no bar. Above 52.4% beats the −110 price.

The market followed our plays77%of the time

When a line moved between open and kickoff, it moved toward our side 77% of the time, by 1.5 points on average. A sign the picks find real value, not noise.

Cumulative units, week by weekOur plays, one unit each
0+10+20+30+40202320242025+41.9u

Running total of units won or lost, one unit per play, week by week. Flat stretches are weeks with few or no plays.

SeasonPlaysW–L–PATSUnitsEvery gameATSUnits
20235036–14–172.0%+20.6u28054.3%+11.2u
202410657–49–253.8%+3.1u28052.5%+0.7u
202511971–48–359.7%+18.2u27954.5%+12.3u
▸Fine printmethodology, the closing line, every season back to 2021, and where the bar sits

Probabilities are walk-forward: fit only on earlier seasons, never on the games being graded. Plays are anchored and graded at the opening line because that is where the edge lives; the same plays re-graded at the closing line, and a control that makes its picks at the close, show how much of it the market takes back by kickoff. Seasons marked pre are walk-forward but predate the window the bar was set on.

ATS · plays at the open59.6%164–111–6 · 275 plays
Every game, flat (no bar)
53.8%
Break-even at −110
52.4%

A play is a side whose calibrated cover probability clears the bar at the opening line.

Units · 1u per play, −110+41.9u+13.9% ROI
Flat-betting every game
+24.2u · +2.6%

ROI is return on total risk (1.1u per play). Pushes return the stake.

Closing line value77%of moved lines went our way
Average move toward the pick
+1.5 pts
Lines that moved at all
84%

The market tends to follow the plays between open and close — the cleanest sign the edge is real, not noise.

Same plays, graded at the close56.0%+20.9u · +6.9%

What the identical picks would have returned at the closing number. The gap to the open is the price of waiting.

Picks made at the close53.6%97 plays · +2.3%

The control: the same model and bar, but anchored on the closing line. The ranking signal is largely gone by then — which is why the board publishes at the open.

Projected margin · MAE9.97pts
Our projection
9.97
Opening line
10.06
Closing line
9.77

Straight-up winner 65.6% over 853 games. The margin matches the market; the edge is in the calibrated probability, not the point estimate.

Cumulative units, week by weekPlays at the openPicks made at the close
0+10+20+30+40202320242025+41.9u+2.5u

Every week of the walk-forward history, in order. Flat weeks are bye-heavy or play-free. The blue line is the record the board is built on; the orange line is the same bar applied to picks made at the close.

Season by season

Left: the model. Middle: plays at the bar, anchored and graded at the opening line, re-graded at the close, and how often the close moved our way. Right, greyed: the control that makes its picks at the close.

SeasonGamesSUMAE model · open · closePlaysW–L–PATSUnitsROIAt closeCLVClose playsATSUnits
2021pre28465.0%10.85 · 10.78 · 10.70189101–88–553.4%+4.2u+2.0%52.1%70%17550.9%-5.6u
2022pre28364.4%9.14 · 9.02 · 8.8015975–84–447.2%-17.4u-9.9%46.3%56%17149.7%-9.6u
202328564.6%10.26 · 10.38 · 9.985036–14–172.0%+20.6u+37.5%64.6%84%2948.3%-2.5u
202428469.0%9.83 · 9.90 · 9.6710657–49–253.8%+3.1u+2.7%55.2%71%3554.3%+1.4u
202528463.2%9.82 · 9.89 · 9.6611971–48–359.7%+18.2u+13.9%53.3%78%3357.6%+3.6u

Where the bar sits

2023–2025, plays at the open. Higher bars win more often on fewer plays; the published bar is the one the board is built on.

BarPlaysATSUnitsROI
Every game83953.8%+24.2u+2.6%
p ≥ 0.5240656.9%+38.5u+8.6%
p ≥ 0.523835457.6%+39.0u+10.0%
p ≥ 0.53the bar27559.6%+41.9u+13.9%
p ≥ 0.5418362.3%+38.1u+18.9%
p ≥ 0.5511658.6%+15.2u+11.9%