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%.

Fade every over

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

Loses2025 held out

−3.3% ROI

13,680 bets, −3.3% ROI.

“Loses” is set by a fixed rule, not a judgement: the ROI interval lies wholly under zero. Re-checked on this page from the file’s own interval.

Hit rate52.5% [51.7%, 53.4%]

53.7% break-even

ROI−3.3% [−5.0%, −1.8%]

Cleared 52.5% [51.7, 53.4] of 13,656 graded, against a break-even of 53.7% at the average price paid (−116): the interval sits wholly under it. The ROI interval runs from −5.0% to −1.8%, and sits wholly under zero.

Bets: 13,680; 36 weeks · 24 void
Cleared: 7,174; 6,482 missed · 0 push
Hit rate: 52.5%; 95% [51.7, 53.4] · of 13,656
Break-even: 53.7%; at the average price paid, −116
ROI: −3.3%; 95% [−5.0, −1.8] · by week
Units: −454.6u; 95% [−671.6, −254.2] · 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/fade_every_over.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

1 of 2 seasons' ROI intervals include zero; 1 sits wholly under it; 0 sit wholly over it.

Season	Bets	Cleared	95%	ROI	95%	Against zero
2023	6,005	50.1%	[48.8, 51.3]	−5.6%	[−8.2, −3.1]
2024	7,675	54.5%	[53.3, 55.6]	−1.5%	[−3.2, +0.1]

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

32 segments in 5 cuts; 6 under 100 bets, greyed.

Market

Segment	Bets	Cleared	95%	ROI	95%
Receiving yards	6,378	51.1%	[49.9, 52.4]	−3.3%	[−5.5, −1.1]
Receptionsunder 100 bets	94	44.7%	[35.0, 54.7]	−17.7%	[−36.5, +1.1]
Rush attempts	974	51.9%	[48.8, 55.0]	−3.1%	[−8.7, +2.4]
Sacks	1,166	65.8%	[63.0, 68.4]	−2.3%	[−6.8, +2.3]
Tackles + assists	5,068	51.5%	[50.1, 52.9]	−3.4%	[−6.1, −1.0]

Home or away

Segment	Bets	Cleared	95%	ROI	95%	Against zero
Away	6,753	53.2%	[52.0, 54.4]	−2.1%	[−4.4, +0.2]
Home	6,706	51.8%	[50.6, 53.0]	−4.8%	[−6.8, −2.9]

Roof

Segment	Bets	Cleared	95%	ROI	95%	Against zero
Open air	9,178	53.0%	[52.0, 54.0]	−2.5%	[−4.6, −0.3]
Dome	4,281	51.4%	[49.9, 52.9]	−5.3%	[−7.7, −3.1]

Position

Segment	Bets	Cleared	95%	ROI	95%
CB	1,348	49.3%	[46.6, 51.9]	−8.5%	[−14.3, −2.7]
DB	188	47.9%	[40.8, 55.0]	−10.4%	[−28.8, +5.9]
DE	664	61.1%	[57.4, 64.8]	−2.0%	[−7.4, +3.8]
DLunder 100 bets	10	70.0%	[39.7, 89.2]	ROI and units withheld: they would average 7 cleared bet prices, fewer than 10, and could be read back to a book price
DT	418	67.9%	[63.3, 72.2]	+1.8%	[−5.0, +8.3]
FBunder 100 bets	20	55.0%	[34.2, 74.2]	+3.8%	[−34.5, +42.0]
FS	362	53.9%	[48.7, 59.0]	+1.8%	[−5.6, +9.7]
ILB	115	52.2%	[43.1, 61.1]	−2.1%	[−19.4, +14.5]
LB	1,483	54.2%	[51.6, 56.7]	−2.6%	[−6.2, +0.9]
MLB	191	53.4%	[46.3, 60.3]	−0.7%	[−14.3, +12.2]
NTunder 100 bets	39	79.5%	[64.5, 89.2]	+27.2%	[+2.0, +48.7]
OLB	284	53.5%	[47.7, 59.2]	−1.9%	[−13.1, +9.2]
QB	801	52.4%	[49.0, 55.9]	−2.0%	[−7.9, +3.7]
RB	1,690	51.4%	[49.0, 53.8]	−3.1%	[−7.1, +1.2]
S	143	51.0%	[42.9, 59.1]	−3.4%	[−17.3, +10.2]
S	967	51.1%	[48.0, 54.3]	−3.5%	[−9.9, +2.9]
SSunder 100 bets	13	69.2%	[42.4, 87.3]	ROI and units withheld: they would average 9 cleared bet prices, fewer than 10, and could be read back to a book price
TE	1,475	49.6%	[47.1, 52.2]	−6.1%	[−12.0, −0.4]
WR	3,418	51.6%	[49.9, 53.2]	−2.5%	[−5.5, +0.6]

De-vigged price

Segment	Bets	Cleared	95%	ROI	95%
0.40–0.50	3,674	47.1%	[45.5, 48.8]	−4.8%	[−7.7, −2.0]
0.50–0.60	9,195	53.3%	[52.2, 54.3]	−2.7%	[−4.7, −0.6]
<0.40under 100 bets	47	44.7%	[31.4, 58.8]	+9.4%	[−23.4, +40.0]
>=0.60	764	70.2%	[66.9, 73.4]	−5.3%	[−9.2, −1.4]

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 receptions/receiving_yards/rush_attempts/tackles_assists/sacks, 2023-2025 REG weeks 1-18, 0 condition(s), consensus_close at close

Bet type: prop
Markets: Receptions, Receiving yards, Rush attempts, Tackles + assists, Sacks
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 204883ce9388d6c1614c6ac733dc235cae92e1a897948c4bbcffb2dc11df0698 · 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 13,680 bets, newest first; page 195 of 274. Filters narrow the list; nothing is removed.

Date · wk	Bet	De-vig	Result	Actual
2023-11-262023 · wk 12	Josh AllenRush attempts 5.5 underBUF v PHI	46.8%	missed	9
2023-11-262023 · wk 12	Kevin ByardTackles + assists 7.5 underPHI v BUF	52.0%	missed	13
2023-11-262023 · wk 12	Kenny GainwellReceiving yards 6.5 underPHI v BUF	48.0%	missed	10
2023-11-262023 · wk 12	00-0034720Tackles + assists 8.5 underLV v KC	48.0%	missed	11
2023-11-262023 · wk 12	D'Andre SwiftReceiving yards 17.5 underPHI v BUF	48.5%	cleared	4
2023-11-262023 · wk 12	Noah GrayReceiving yards 10.5 underKC v LV	51.0%	missed	30
2023-11-262023 · wk 12	Latavius MurrayReceiving yards 5.5 underBUF v PHI	51.0%	missed	18
2023-11-262023 · wk 12	Julio JonesReceiving yards 12.5 underPHI v BUF	51.0%	cleared	0
2023-11-262023 · wk 12	Jack StollReceiving yards 10.5 underPHI v BUF	49.0%	missed	14
2023-11-262023 · wk 12	Jalen HurtsRush attempts 9.5 underPHI v BUF	49.0%	missed	14
2023-11-262023 · wk 12	00-0035262Tackles + assists 7.5 underKC v LV	49.0%	missed	8
2023-11-262023 · wk 12	00-0033547Tackles + assists 7.5 underPHI v BUF	49.0%	missed	10
2023-11-262023 · wk 12	Rashee RiceReceiving yards 41.5 underKC v LV	50.0%	missed	107
2023-11-262023 · wk 12	Michael MayerReceiving yards 20.5 underLV v KC	50.0%	missed	27
2023-11-262023 · wk 12	Dalton KincaidReceiving yards 48.5 underBUF v PHI	50.0%	cleared	38
2023-11-262023 · wk 12	Tre TuckerReceiving yards 9.5 underLV v KC	50.0%	cleared	0
2023-11-262023 · wk 12	Skyy MooreReceiving yards 15.5 underKC v LV	50.0%	missed	34
2023-11-262023 · wk 12	Khalil ShakirReceiving yards 31.5 underBUF v PHI	50.0%	missed	47
2023-11-262023 · wk 12	James CookReceiving yards 16.5 underBUF v PHI	50.0%	missed	57
2023-11-262023 · wk 12	Isiah PachecoReceiving yards 20.5 underKC v LV	50.0%	missed	34
2023-11-262023 · wk 12	DeVonta SmithReceiving yards 56.5 underPHI v BUF	50.0%	missed	106
2023-11-262023 · wk 12	Gabe DavisReceiving yards 33.5 underBUF v PHI	50.0%	missed	105
2023-11-262023 · wk 12	Josh JacobsReceiving yards 16.5 underLV v KC	50.0%	cleared	15
2023-11-262023 · wk 12	A.J. BrownReceiving yards 81.5 underPHI v BUF	50.0%	cleared	37
2023-11-262023 · wk 12	Hunter RenfrowReceiving yards 17.5 underLV v KC	50.0%	missed	38
2023-11-262023 · wk 12	Jakobi MeyersReceiving yards 39.5 underLV v KC	50.0%	missed	79
2023-11-262023 · wk 12	Justin WatsonReceiving yards 31.5 underKC v LV	50.0%	cleared	3
2023-11-262023 · wk 12	Marquez Valdes-ScantlingReceiving yards 15.5 underKC v LV	50.0%	cleared	−1
2023-11-262023 · wk 12	Stefon DiggsReceiving yards 75.5 underBUF v PHI	50.0%	cleared	74
2023-11-262023 · wk 12	Davante AdamsReceiving yards 62.5 underLV v KC	50.0%	missed	73
2023-11-262023 · wk 12	Travis KelceReceiving yards 72.5 underKC v LV	50.0%	missed	91
2023-11-262023 · wk 12	00-0031576Tackles + assists 3.5 underCLE v DEN	65.0%	cleared	1
2023-11-262023 · wk 12	00-0038623Tackles + assists 4.5 underARI v LA	59.7%	cleared	4
2023-11-262023 · wk 12	Dorian Thompson-RobinsonRush attempts 4.5 underCLE v DEN	59.0%	missed	5
2023-11-262023 · wk 12	00-0036884Tackles + assists 3.5 underARI v LA	41.1%	void	—
2023-11-262023 · wk 12	00-0033868Tackles + assists 3.5 underCLE v DEN	41.1%	cleared	2
2023-11-262023 · wk 12	00-0033918Tackles + assists 4.5 underDEN v CLE	58.0%	cleared	4
2023-11-262023 · wk 12	00-0031388Tackles + assists 4.5 underLA v ARI	56.9%	cleared	0
2023-11-262023 · wk 12	00-0039137Tackles + assists 3.5 underLA v ARI	44.1%	cleared	2
2023-11-262023 · wk 12	00-0036282Tackles + assists 6.5 underCLE v DEN	55.1%	cleared	4
2023-11-262023 · wk 12	00-0036996Tackles + assists 3.5 underCLE v DEN	54.1%	missed	7
2023-11-262023 · wk 12	00-0036874Tackles + assists 3.5 underDEN v CLE	54.1%	missed	4
2023-11-262023 · wk 12	Kyler MurrayRush attempts 6.5 underARI v LA	46.8%	cleared	1
2023-11-262023 · wk 12	00-0036568Tackles + assists 3.5 underARI v LA	46.9%	cleared	3
2023-11-262023 · wk 12	00-0036431Tackles + assists 5.5 underLA v ARI	46.9%	cleared	3
2023-11-262023 · wk 12	00-0035933Tackles + assists 3.5 underLA v ARI	46.9%	missed	5
2023-11-262023 · wk 12	00-0033890Tackles + assists 6.5 underARI v LA	46.9%	missed	9
2023-11-262023 · wk 12	00-0031898Tackles + assists 10.5 underDEN v CLE	46.9%	cleared	10
2023-11-262023 · wk 12	00-0034368Tackles + assists 7.5 underDEN v CLE	52.0%	cleared	3
2023-11-262023 · wk 12	00-0036994Tackles + assists 9.5 underLA v ARI	48.0%	cleared	6

All 13,680 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/fade_every_over.json, written Sun, Sep 27, 2:14 AM ET · universe built Sun, Sep 27, 2:14 AM ET · strategy 204883ce9388

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