Today · Oct 3 · 18 events Games under way Settled· Sep 30

Regular season · week 4 · Sunday, Oct 4, 1:00 PM ET

New England PatriotsatBuffalo BillsEXPERIMENTAL

Highmark Stadium

New England PatriotsBuffalo BillsSaved on this device

66°F · 6 mph wind from the SW · 5% chance of rain · mostly sunny — not used by the model; no weather term enters the numbers below.

Simulation story

5 chapters · read from the published artifact

Skip to the report ↓

› Loading the matchup model› Reading the outcome distribution from 10,000 simulated games› Preparing the score scenariosPress Play to step through what the model expects, chapter by chapter, or read the report below.

NE @ BUF

New England Patriots at Buffalo Bills at Highmark Stadium.

Simulation

What our model expects

10,000 simulated games · model nfl-regular-season-public-v1

Projected score

NE 20 — 27 BUF

the middle of our simulated outcomes, not a call on the exact final

Win chance

NE 33.9% · BUF 63.4%

how often each side won in 10,000 simulations · ties 2.6%

Total points (both teams)

8 in 10 simulations landed between 30 and 64

BUF winning margin

a minus means NE wins by that much · 8 in 10 between -11 and 22

Live game state

Live beta

The left side is what the live source reports right now. The right side is the GameTime forecast made before kickoff — it is frozen and does not change while the game is played.

Live now

Checking the live feed…

Pregame GameTime · frozen

NE: 20 (34% pregame win chance)
BUF: 27 (63% pregame win chance)

Forecast frozen at Oct 3, 2:37 PM ET. It does not change during the game.

Model details

From the margin-of-victory Elo win head, published exactly as evaluated on held-out 2006–2021 — no shrink toward 50% is applied, and no claim to out-predict the sportsbook market is made. Generated 2026-10-03T18:37:22Z under model nfl-regular-season-public-v1 and frozen pre-kickoff; every forecast is settled against the official result.

Projected scorecard

expected statistical summaries · not one simulated game

NE 20 — 27 BUFwin chance NE 33.9% · BUF 63.4%total 47 (30–64)margin +6

Likely TD scorers

Rhamondre Stevenson 31.2%

TreVeyon Henderson 31.1%

Mack Hollins 19.6%

Passing

Drake Maye 211 pass yds (113–355)

Rushing — withheld for NE: the players' modelled shares add up to 110% of the team's carries, more than exists — shown again once the shares are reconciled

Receiving leaders

Mack Hollins 4 rec · 41 yds

Rhamondre Stevenson 2 rec · 21 yds

Romeo Doubs 2 rec · 28 yds

BUF

Likely TD scorers

James Cook III 64.7%

Josh Allen 62.4%

DJ Moore 34.8%

Passing

Josh Allen 237 pass yds (128–395)

Rushing leaders

James Cook III 79 rush yds (25–186)

Josh Allen 25 rush yds (4–82)

Receiving leaders

DJ Moore 4 rec · 41 yds

Dalton Kincaid 3 rec · 42 yds

Khalil Shakir 3 rec · 31 yds

pass interceptions · First/last/2+ touchdown — withheld, with the exact bar each failed

pass interceptions: RESEARCH_ONLY
First/last/2+ touchdown: DISABLED — no ordering model and no calibration receipt of their own; never derived from anytime probabilities

Model detail — who this forecast leaves out (3)

Rushing NE — Withheld for this game: the players' modelled shares add up to more of the team's opportunity than exists; these numbers are not renormalised to fit.
Lan Larison NE — No receiving projection: the receiving model's usage pool does not include his 2026 role here yet.
A.J. Brown NE — Not playing: Listed Injured Reserve.

Range

How wide the outcomes are

the 10th to 90th percentile of each team's simulated score

Team	Low (10th)	Projected	High (90th)
New England Patriots	8	20	33
Buffalo Bills	14	27	38

Exact score

The likeliest final scores

from an event-based simulation of touchdowns and field goals, solved onto the median margin and total above

Score (NE – BUF)	Chance
17–20	0.72%
24–27	0.71%
20–23	0.63%
20–17	0.60%
17–24	0.58%
20–27	0.58%

No single scoreline is likely — these are the most common of hundreds. The projected score above is derived from the medians and answers a different question: the middle of the distribution, not its most common point.

Key numbers

How often it lands on 3, 7, 10 or 14

football's margins pile up on those four — this is the one thing a normal-draw model cannot show you

Margin	This game	Last 3 regular seasons
Decided by exactly 3	9.5%	14.2%
Decided by exactly 7	6.3%	8.3%
Decided by exactly 10	5.5%	4.2%
Decided by exactly 14	4.7%	4.4%
Any of the four	26.1%	31.1%

The right-hand column is what actually happened across 816 regular-season games (2023, 2024, 2025) — the same finals this engine was fitted to. It gets the shape right, with 3 the most common margin by a distance, and it is not calibrated to the number: it puts less weight on 3 and more on 10 than those seasons did.

NE touchdowns

2.3

BUF touchdowns

2.9

NE field goals

1.6

BUF field goals

1.9

Tied after regulation

3.4%

Comparison

Us versus the sportsbooks

two independent reads, shown side by side

Our win chance

63.4%

BUF to win

Sportsbook win chance

73.5%

BUF to win, read from 8 sportsbooks' odds with their built-in margin removed

Difference

-10.1 pp

ours minus the sportsbooks', in percentage points — not a recommendation

Total points: ours vs sportsbooks

47 vs 50

our projected total against the sportsbooks' over/under line

The sportsbook numbers are the books' own, shown for context. A difference is a difference — this model is not validated to out-predict the sportsbook market.

Prices captured 2026-10-03T18:08:15Z — before kickoff.

Player projections · 21 modelled

The player board

Every modelled player, with his availability. Players listed out carry no volume projection.

James Cook IIIMatchupBUF vs NEStartSun 1:00 PM ETMarketDraftKingsYes -140Captured Sat 12:52 PM ETModelGameTimePicks64.7%
Josh AllenMatchupBUF vs NEStartSun 1:00 PM ETMarketDraftKingsYes -140Captured Sat 12:52 PM ETModelGameTimePicks62.4%
DJ MooreMatchupBUF vs NEStartSun 1:00 PM ETMarketDraftKingsYes +155Captured Sat 12:52 PM ETModelGameTimePicks34.8%
Rhamondre StevensonMatchupNE vs BUFStartSun 1:00 PM ETMarketDraftKingsYes +135Captured Sat 12:52 PM ETModelGameTimePicks31.2%
TreVeyon HendersonMatchupNE vs BUFStartSun 1:00 PM ETMarketDraftKingsYes +215Captured Sat 12:52 PM ETModelGameTimePicks31.1%
Dalton KincaidMatchupBUF vs NEStartSun 1:00 PM ETMarketDraftKingsYes +160Captured Sat 12:52 PM ETModelGameTimePicks28.6%
Ty JohnsonMatchupBUF vs NEStartSun 1:00 PM ETMarketDraftKingsYes +500Captured Sat 12:52 PM ETModelGameTimePicks25.1%
Khalil ShakirMatchupBUF vs NEStartSun 1:00 PM ETMarketDraftKingsYes +240Captured Sat 12:52 PM ETModelGameTimePicks23.2%
Mack HollinsMatchupNE vs BUFStartSun 1:00 PM ETMarketDraftKingsYes +260Captured Sat 12:52 PM ETModelGameTimePicks19.6%
Romeo DoubsMatchupNE vs BUFStartSun 1:00 PM ETMarketDraftKingsYes +265Captured Sat 12:52 PM ETModelGameTimePicks16.6%
Keon ColemanMatchupBUF vs NEStartSun 1:00 PM ETMarketDraftKingsYes +370Captured Sat 12:52 PM ETModelGameTimePicks16.3%
Hunter HenryMatchupNE vs BUFStartSun 1:00 PM ETMarketDraftKingsYes +235Captured Sat 12:52 PM ETModelGameTimePicks16.3%
DeMario DouglasMatchupNE vs BUFStartSun 1:00 PM ETMarketDraftKingsYes +370Captured Sat 12:52 PM ETModelGameTimePicks15.8%
Dawson KnoxMatchupBUF vs NEStartSun 1:00 PM ETMarketDraftKingsYes +600Captured Sat 12:52 PM ETModelGameTimePicks13.3%
Joshua PalmerMatchupBUF vs NEStartSun 1:00 PM ETMarketDraftKingsYes +550Captured Sat 12:52 PM ETModelGameTimePicks12.7%
Drake MayeMatchupNE vs BUFStartSun 1:00 PM ETMarketDraftKingsYes +475Captured Sat 12:52 PM ETModelGameTimePicks12.3%
Corey KinerMatchupNE vs BUFStartSun 1:00 PM ETMarketDraftKingsYes +1900Captured Sat 12:52 PM ETModelGameTimePicks12.0%
Kyle WilliamsMatchupNE vs BUFStartSun 1:00 PM ETMarketDraftKingsYes +700Captured Sat 12:52 PM ETModelGameTimePicks8.4%
Lan LarisonMatchupNE vs BUFStartSun 1:00 PM ETMarketNot offeredModelGameTimePicks8.2%

Model detail — who this forecast leaves out (3)

Rushing NE — Withheld for this game: the players' modelled shares add up to more of the team's opportunity than exists; these numbers are not renormalised to fit.
Lan Larison NE — No receiving projection: the receiving model's usage pool does not include his 2026 role here yet.
A.J. Brown NE — Not playing: Listed Injured Reserve.

Families not shown, and the exact bar each failed (2)

pass interceptions: RESEARCH_ONLY
First/last/2+ touchdown: DISABLED — no ordering model and no calibration receipt of their own; never derived from anytime probabilities

Scoring outlooktop 6 of 19 by TD chance · full list in the board’s TD tab

64.7% James Cook III · BUF
62.4% Josh Allen · BUF
34.8% DJ Moore · BUF
31.2% Rhamondre Stevenson · NE
31.1% TreVeyon Henderson · NE
28.6% Dalton Kincaid · BUF

Anytime-scorer probability means scoring a touchdown — never throwing one. Void if the player does not play; questionable players carry their state above.

Reading key

What these numbers mean

Projected score: The middle outcome across every simulated game — not a prediction of the exact final.
Win chance: How often each side won across the simulations, after the calibration described above.
80% range: Eight in ten simulated games landed inside this band. Real games land outside it too.
pp (percentage points): The plain difference between two percentages. A gap is a difference, not an advantage.
Experimental: Published while its out-of-sample record is still accumulating. Every forecast is frozen before kickoff and graded against the official result.

Provenance

Where this came from

model: nfl-regular-season-public-v1 v2
simulations: 10,000
input hash: 9527a970c0357d02
generated: 2026-10-03T18:37:22Z
kickoff: 2026-10-04T17:00Z
state: UPCOMING

Player-family model provenance (4 published)

Rushing yards: Share-level model: cleared every bar when re-tested on 2014–2021 (14,502 player-games) — a second look at seasons examined once before, not a blind test. Each week's forecast is frozen before the first kickoff and graded; its blind 2026 record is still accumulating (216 graded player-games so far).
Receiving yards: props-v1 evaluation: n=3575, all promotion bars pass
Receptions: props-v1 evaluation: n=3575, all promotion bars pass
Anytime touchdown: Opportunity touchdown model: cleared every bar on a blind test of 2014–2021 (35,128 player-games never used to build it). Probabilities condition on playing — a player who does not play settles void. Each week's forecast is frozen before the first kickoff and graded; its blind 2026 record is still accumulating (524 graded player-games so far).

Our win chances come from a team rating that weights how decisively teams win. Tested on 4,281 past games it had never seen (2006–2021), it scored 0.629 on log loss (lower is better) — better than our previous rating's 0.642 and a coin flip's 0.693, but short of the sportsbooks' own odds at 0.610. It has NOT been shown to beat the sportsbook market. Where our win and margin ratings disagree on the favourite, the previous rating is used for that game. The 2026 season evaluation still runs under bars frozen in August, so every forecast stays experimental. These ratings are built from results, not from who is playing: they do not know that a starting quarterback has been ruled out. Our player projections do remove players who cannot play; the team win chance beside them does not. We tested whether adding that would help, using perfect hindsight about whether each team's main quarterback actually played — on 855 games from 2023 to 2025 it improved the win rating on average but was WORSE in one of the three seasons, so it did not clear the bar we hold these ratings to and we left the rating alone.

Educational and paper-only — not betting advice.