Replay · NFL

Test a rule against real closing prices.

Each rule here was replayed against every settled line it would have bet, at the closing price, with how sure the result is. Choose one to read it.

Rules replayed

4 more listed that the engine cannot run

Seasons replayed

1999–2024 · 2025 held out

Priced at

all 4 replayed rules · consensus close · a choice in the builder

01Rules

4 rules replayed · 4 more listed that the engine cannot run · Film Room

Loses01Fade every over13,680 bets · −3.3% ROI · open in the builder
Unclear02Rush attempts unders2,256 bets · −1.8% ROI · open in the builder
Loses03Chase the streak1,150 bets · −13.2% ROI · open in the builder
Cannot run04Follow the model's leansWaiting for walk-forward model predictions persisted for the historical seasons.
Cannot run05Buy the middleWaiting for a rule language with two legs: lab.strategy/1 bets one side at one price.
Cannot run06Arbitrage between booksWaiting for a rule language with two legs: lab.strategy/1 bets one side at one price.
Loses07Home underdogs of 3 to 71,400 bets · −5.6% ROI · open in the builder
Cannot run08Unders in the windWaiting for forecast weather joined to the replayed games.

4 rules replayed on this site so far. The more rules are tried, the more likely one clears zero by chance alone: with 4 tried, a 95% interval that allows for it spans about ±2.50 standard errors (Bonferroni), not ±1.96. The intervals on this page are the plain 95%.

Rush attempts unders

prop under on rush_attempts, 2023-2025 REG weeks 1-18, 0 condition(s), consensus_close at close

Unclear2025 held out

−1.8% ROI

2,256 bets, −1.8% ROI.

“Unclear” is set by a fixed rule, not a judgement: the ROI interval spans zero and the luck check is outside the 5th to 95th percentile, or could not be run. Re-checked on this page from the file’s own interval.

Hit rate52.4% [50.4%, 54.5%]

53.0% break-even

ROI−1.8% [−5.0%, +1.0%]

Cleared 52.4% [50.4, 54.5] of 2,253 graded, against a break-even of 53.0% at the average price paid (−113): the interval contains it. The ROI interval runs from −5.0% to +1.0%, and includes zero.

Bets: 2,256; 36 weeks · 3 void
Cleared: 1,181; 1,072 missed · 0 push
Hit rate: 52.4%; 95% [50.4, 54.5] · of 2,253
Break-even: 53.0%; at the average price paid, −113
ROI: −1.8%; 95% [−5.0, +1.0] · by week
Units: −41.3u; 95% [−111.0, +22.3] · 1u flat

Worst drawdown: coming once the rule’s file carries its drawdown key (a-44).
CLV: none to publish. Every bet here is taken at the close, which is the price CLV is measured against.

Returns over time

units · by week block

No luck check: the rule takes every bet the universe offers at this side, so a random draw of the same size IS the rule.

Returns over time: coming once this rule’s file, lab/nfl/presets/rush_attempts_unders.json, carries its week-block series (the returns key, a-44). Nothing is drawn from anything else in the meantime.

How this is computed

The line is cumulative units, staked 1 unit flat, counted in date order. It moves once per block of weeks, not once per bet: each block is whole weeks of one season holding at least the file's floor of cleared bets, because consecutive points of a bet-by-bet curve would give each bet’s book price back, and the file withholds those (“Deliberately restricted: no bet carries its book price, its book, its stake or its profit, and the file has no equity curve. The prices are The Odds API's, whose terms (read 2026-09-24, updated 31 Aug 2026) say nothing either way about displaying historical odds, so they stay on the server and only aggregates leave it. A winning flat bet's profit is its price minus one, and consecutive points of a cumulative curve difference back to it, so those are withheld for the same reason. ROI and units appear only where they average at least 10 cleared bets' prices. Do not add these fields back without a decision that the terms allow it.”). The band is where 90% of random rules land at the same block, the 5th to the 95th percentile of rules drawing the same number of bets, at random, from the same pool of eligible lines on the same side. Inside it, the rule did nothing a coin could not.

By season

ROI · 95% interval

2 of 2 seasons' ROI intervals include zero; 0 sit wholly under it; 0 sit wholly over it.

Season	Bets	Cleared	95%	ROI	95%	Against zero
2023	1,020	51.6%	[48.5, 54.6]	−3.0%	[−7.4, +1.4]
2024	1,236	53.1%	[50.3, 55.9]	−0.8%	[−5.0, +3.1]

How this is computed

A real effect shows up every year; one good season inside a run of intervals that include zero is what chance looks like. The ROI interval resamples whole weeks, so bets in the same week move together. Seasons under 100 bets are greyed: their intervals are wide enough that the row should not be read on its own.

2025 is held out

The file: “held out: no figure in this result includes the 2025 season”. The season is kept back so the rule can be tested once, on games it was never shaped by. Looking is done once, when the rule has stopped changing: the job that replays it reveals the season and writes that into the file, and from then on 2025 is marked as seen on this rule.

Not revealed · no 2025 figure anywhere on this page

Where it won and lost

95% intervals · under 100 bets greyed

13 segments in 5 cuts; 4 under 100 bets, greyed.

Market

Segment	Bets	Cleared	95%	ROI	95%	Against zero
Rush attempts	2,256	52.4%	[50.4, 54.5]	−1.8%	[−5.0, +1.0]

Home or away

Segment	Bets	Cleared	95%	ROI	95%	Against zero
Away	1,096	53.1%	[50.1, 56.0]	−0.6%	[−6.1, +4.3]
Home	1,120	51.9%	[49.0, 54.8]	−2.8%	[−7.7, +2.0]

Roof

Segment	Bets	Cleared	95%	ROI	95%	Against zero
Open air	1,542	53.2%	[50.7, 55.7]	−0.4%	[−4.3, +3.7]
Dome	674	50.9%	[47.1, 54.7]	−4.8%	[−11.4, +1.8]

Position

Segment	Bets	Cleared	95%	ROI	95%
QB	800	51.9%	[48.4, 55.3]	−3.2%	[−9.2, +2.7]
RB	1,441	52.6%	[50.0, 55.1]	−1.3%	[−4.9, +2.1]
TEunder 100 bets	4	100.0%	[51.0, 100.0]	ROI and units withheld: they would average 4 cleared bet prices, fewer than 10, and could be read back to a book price
WRunder 100 bets	2	50.0%	[9.5, 90.5]	ROI and units withheld: they would average 1 cleared bet price, fewer than 10, and could be read back to a book price

De-vigged price

Segment	Bets	Cleared	95%	ROI	95%
0.40–0.50	1,004	48.5%	[45.4, 51.5]	−3.2%	[−9.1, +1.9]
0.50–0.60	1,219	55.3%	[52.5, 58.1]	−1.3%	[−6.2, +3.7]
<0.40under 100 bets	15	46.7%	[24.8, 69.9]	ROI and units withheld: they would average 7 cleared bet prices, fewer than 10, and could be read back to a book price
>=0.60under 100 bets	18	83.3%	[60.8, 94.2]	+27.7%	[−0.6, +53.0]

How this is computed

The same bets, cut by what was known before kickoff. A segment is not a new rule: with this many cuts, some will sit away from the rest by chance, so a single segment’s interval is not evidence on its own. ROI and units are withheld where they would average fewer than 10 cleared bets’ prices, and the file says so on the row.

The rule, as the engine approved it

prop under on rush_attempts, 2023-2025 REG weeks 1-18, 0 condition(s), consensus_close at close

Bet type: prop
Markets: Rush attempts
Side: under
Seasons: 2023–2025 · REG · weeks 1–18
Conditions: none
Line: main
Price: consensus_close · draftkings, fanduel, betmgm · de-vig multiplicative · at the book's posted juice · -110 where none is posted
Timing: close
Staking: flat · 1u
Limits: per game no limit · per player week 1 · per week no limit
Settlement: push: refund · no snap: void

Where the prices came from

2023–2025 · The Odds API historical close, draftkings, fanduel, betmgm · a close · last snapshot at or before kickoff

strategy 1c99c42f3b02ef27be10f90cb8cd444023742300995c75982621917d918468ec · AUDIT-2026-09-24, the Lab library launch set

How this is computed

The strategy is stored exactly as the engine approved it, defaults filled, and hashed; the same rule always hashes the same. Every condition is a figure stored as of before kickoff, so nothing known after the game can decide a bet.

Every bet it made

50 shown of 2,256 bets, newest first; page 17 of 46. Filters narrow the list; nothing is removed.

Date · wk	Bet	De-vig	Result	Actual
2024-10-132024 · wk 6	Tyler AllgeierRush attempts 9.5 underATL v CAR	52.8%	missed	18
2024-10-132024 · wk 6	Dak PrescottRush attempts 2.5 underDAL v DET	52.5%	cleared	1
2024-10-132024 · wk 6	Rico DowdleRush attempts 11.5 underDAL v DET	48.0%	cleared	5
2024-10-132024 · wk 6	Chuba HubbardRush attempts 15.5 underCAR v ATL	48.8%	missed	18
2024-10-132024 · wk 6	Jahmyr GibbsRush attempts 12.5 underDET v DAL	49.0%	cleared	12
2024-10-132024 · wk 6	Bijan RobinsonRush attempts 14.5 underATL v CAR	49.0%	missed	15
2024-10-132024 · wk 6	Aidan O'ConnellRush attempts 2.5 underLV v PIT	61.4%	cleared	0
2024-10-132024 · wk 6	Alexander MattisonRush attempts 13.5 underLV v PIT	46.7%	missed	14
2024-10-132024 · wk 6	Justin FieldsRush attempts 8.5 underPIT v LV	48.0%	missed	11
2024-10-132024 · wk 6	Bo NixRush attempts 5.5 underDEN v LAC	49.0%	missed	6
2024-10-132024 · wk 6	Javonte WilliamsRush attempts 12.5 underDEN v LAC	49.0%	cleared	6
2024-10-132024 · wk 6	Najee HarrisRush attempts 15.5 underPIT v LV	50.0%	cleared	14
2024-10-132024 · wk 6	J.K. DobbinsRush attempts 16.5 underLAC v DEN	50.0%	missed	25
2024-10-132024 · wk 6	Will LevisRush attempts 2.5 underTEN v IND	41.6%	missed	3
2024-10-132024 · wk 6	C.J. StroudRush attempts 2.5 underHOU v NE	43.3%	missed	3
2024-10-132024 · wk 6	Jayden DanielsRush attempts 8.5 underWAS v BAL	43.7%	cleared	6
2024-10-132024 · wk 6	Justice HillRush attempts 3.5 underBAL v WAS	45.1%	cleared	1
2024-10-132024 · wk 6	Deshaun WatsonRush attempts 4.5 underCLE v PHI	54.1%	cleared	4
2024-10-132024 · wk 6	Drake MayeRush attempts 5.5 underNE v HOU	53.5%	cleared	5
2024-10-132024 · wk 6	Lamar JacksonRush attempts 9.5 underBAL v WAS	53.5%	missed	11
2024-10-132024 · wk 6	Trey SermonRush attempts 13.5 underIND v TEN	53.3%	missed	18
2024-10-132024 · wk 6	Derrick HenryRush attempts 18.5 underBAL v WAS	52.5%	missed	24
2024-10-132024 · wk 6	Tony PollardRush attempts 15.5 underTEN v IND	52.3%	missed	17
2024-10-132024 · wk 6	Antonio GibsonRush attempts 13.5 underNE v HOU	47.7%	cleared	13
2024-10-132024 · wk 6	James ConnerRush attempts 16.5 underARI v GB	47.7%	cleared	7
2024-10-132024 · wk 6	Bucky IrvingRush attempts 15.5 underTB v NO	52.0%	cleared	14
2024-10-132024 · wk 6	Josh JacobsRush attempts 16.5 underGB v ARI	52.0%	missed	18
2024-10-132024 · wk 6	Tyjae SpearsRush attempts 7.5 underTEN v IND	48.0%	cleared	6
2024-10-132024 · wk 6	Jerome FordRush attempts 10.5 underCLE v PHI	48.0%	cleared	2
2024-10-132024 · wk 6	Jordan LoveRush attempts 2.5 underGB v ARI	49.0%	missed	5
2024-10-132024 · wk 6	Spencer RattlerRush attempts 3.5 under2024_06_TB_NO	51.0%	missed	4
2024-10-132024 · wk 6	Emanuel WilsonRush attempts 7.5 underGB v ARI	51.0%	cleared	7
2024-10-132024 · wk 6	Jalen HurtsRush attempts 9.5 underPHI v CLE	49.0%	missed	14
2024-10-132024 · wk 6	Alvin KamaraRush attempts 16.5 underNO v TB	49.0%	cleared	13
2024-10-132024 · wk 6	JaMycal HastyRush attempts 4.5 underNE v HOU	50.5%	cleared	3
2024-10-132024 · wk 6	Joe MixonRush attempts 13.5 underHOU v NE	50.5%	cleared	13
2024-10-132024 · wk 6	Baker MayfieldRush attempts 3.5 underTB v NO	49.5%	cleared	3
2024-10-132024 · wk 6	Kyler MurrayRush attempts 5.5 underARI v GB	50.0%	missed	7
2024-10-132024 · wk 6	Saquon BarkleyRush attempts 18.5 underPHI v CLE	50.0%	cleared	18
2024-10-132024 · wk 6	Austin EkelerRush attempts 10.5 underWAS v BAL	50.0%	cleared	9
2024-10-132024 · wk 6	D'Andre SwiftRush attempts 14.5 underCHI v JAX	55.3%	missed	17
2024-10-132024 · wk 6	Tank BigsbyRush attempts 9.5 underJAX v CHI	53.5%	cleared	7
2024-10-132024 · wk 6	Caleb WilliamsRush attempts 4.5 underCHI v JAX	52.0%	cleared	4
2024-10-132024 · wk 6	Trevor LawrenceRush attempts 3.5 underJAX v CHI	51.5%	missed	4
2024-10-132024 · wk 6	Roschon JohnsonRush attempts 6.5 underCHI v JAX	51.2%	cleared	5
2024-10-132024 · wk 6	Travis EtienneRush attempts 10.5 underJAX v CHI	49.0%	cleared	3
2024-10-102024 · wk 6	Zach CharbonnetRush attempts 3.5 underSEA v SF	44.1%	missed	5
2024-10-102024 · wk 6	Geno SmithRush attempts 3.5 underSEA v SF	55.4%	cleared	1
2024-10-102024 · wk 6	Jordan MasonRush attempts 18.5 underSF v SEA	53.1%	cleared	9
2024-10-102024 · wk 6	Brock PurdyRush attempts 3.5 underSF v SEA	48.5%	missed	4

All 2,256 bets in the rule’s file (JSON) · no price, book, stake or profit per bet, by the file’s restriction

How this is computed

One row per bet as the file lists it: the line, the side, the de-vigged probability of that side at the close, and how it settled. A void is a player with no snap: not a bet, and not a win.

Hypothetical results · nothing here was bet · not advice · lab/nfl/presets/rush_attempts_unders.json, written Sun, Sep 27, 2:14 AM ET · universe built Sun, Sep 27, 2:14 AM ET · strategy 1c99c42f3b02

Test a rule against real closing prices.

00Build a rule

01Rules

Returns over time

By season

2025 is held out

Where it won and lost

Market

Home or away

Roof

Position

De-vigged price

The rule, as the engine approved it

Where the prices came from

Every bet it made