TRANSACTION · MLBGiants: Claimed 1B Buddy Kennedy off waivers from Seattle (Aug 2)
TRANSACTION · MLBBrewers: Sent RHP Joel Kuhnel on a rehab assignment to Nashville (IL) (Aug 2)
TRANSACTION · MLBDodgers: Acquired CF Zyhir Hope and RHPs River Ryan and Brady Smith from Detroit in excha…
Free during beta —to track favorites + alerts
MLB · NBA · WNBA · NHL · NFL · NCAAF · NCAABPublic track record

What the model has actually called

Every completed game across MLB, NBA, WNBA, NHL, NFL, and NCAA. Each prediction is logged before the game finishes; the result is whatever happened. No cherry-picking, no hindsight edits. Refreshed every 5 minutes.

59.4%30-day accuracy
80.8%Lock tier · all-time
21505Games called

Live model health · 30-day window

Drift fires when the last 30 days' Brier score exceeds the prior 90-day baseline by more than 0.025. When a market drifts, picks pause automatically until the next snapshot clears it. Snapshots refresh on an hourly cron. r363 — this strip mirrors the same data the auto-pause banner reads, surfaced here as a public-facing receipt.

RELIABILITY · PREDICTED vs OBSERVED

Does it win as often as it claims?

X · MODEL SAID Y · ACTUALLY WON n=6,990

6,990 picks · 9 buckets

Model calibration: predicted win probability vs actual win rate by bucket, with 95% Wilson confidence intervals, over 6,990 graded picks.
PredictedActual win rateSample95% CIVerdict
50 to 55%52.8%1381 of 26155155%calibrated
55 to 60%55.3%1233 of 22305357%overconfident
60 to 65%61.6%773 of 12555964%calibrated
65 to 70%68.1%356 of 5236472%calibrated
70 to 75%73.7%165 of 2246879%calibrated
75 to 80%78.8%63 of 806986%calibrated
80 to 85%90.9%40 of 447996%calibrated
85 to 90%94.4%17 of 187499%calibrated
90% or higher0.0%0 of 1079%overconfident

When the model said ~53%, those picks won 52.8% (1381/2615) — inside the 95% noise band.

Accuracy windows · cross-sport

Last 7 days

60.0%

66 / 110 games

Last 30 days

59.4%

256 / 431 games

Last 90 days

61.2%

882 / 1441 games

All-time

+1.0 pts

1575/21505 priced · raw 66.7%

By league · last 30 days

Spread receipts · NCAA

Posted spreads only. No price invented.

50.0%

1,074 decided · 52.4% breaks even at -110

NCAAF

52-46-3 ATS

53.1% cover rate

sharp 54% 34-29

30D reliability

n=98 · ECE 0.083 · Brier 0.249

NCAAB

485-491 ATS

49.7% cover rate

sharp 52% 313-287high 46% 68-79medium 49% 52-54low 42% 52-71

30D reliability

n=976 · ECE 0.086 · Brier 0.259

By model confidence · all-time · cross-sport

Locks

Highest model conviction

81%5400/6684

Edges

Strong lean

68%4125/6088

Leans

Modest lean

57%2764/4809

Tossups

Near coin-flip

53%2065/3924

Higher-conviction tiers should hit at higher rates. If they don't, the model is poorly calibrated — the table tells the truth either way. Tier thresholds tuned per league (NHL tighter than NBA / MLB / WNBA).

What this model is — and isn't

It's accurate and well-calibrated. It wins more than 6 in 10 outright, and when it says 70% those picks land near 70% — that's our Calibration Grade, the receipt we lead with.

It does not consistently beat the closing line. Our closing-line value runs slightly negative — by kickoff the market, on average, prices these games a hair sharper than the model does. Calibration and market-beating are different things, and we're straight about which one we've earned.

Most services bury this number. We publish it, in full, right below. Receipts, not picks.

Vs market · cross-sport · 2,853 games on closing lines

Agreed with Vegas

Model picked the closing favorite

59.9%1218 / 2035 games

Disagreed (the edge bets)

Model picked the closing underdog

54.2%443 / 818 games
MLB1,582 games

Agree

59.5% 654/1099

Disagree

57.1% 276/483

NBA55 games

Agree

71.4% 25/35

Disagree

60.0% 12/20

WNBA201 games

Agree

74.2% 118/159

Disagree

52.4% 22/42

NHL1,015 games

Agree

56.7% 421/742

Disagree

48.7% 133/273

Disagreement bucket = bets where the model picked the underdog at the closing line. If that row hits above ~46%, the model would be +EV against the closing price assuming standard −110-equivalent juice. WNBA absent — ESPN's WNBA pickcenter doesn't ship closing spreads consistently. Pure CLV (cents won vs closing moneyline) is below — see the “Pure CLV” panel.

Pure CLV · cross-sport · 1,942 graded picks

Avg CLV

+383

basis points

Median CLV

+336

basis points

Positive %

6802.0%

picks beat the close

Per 100u

-1262285¢

cents · normalized

Sharp

301 picks

+1276

10000% positive

High

149 picks

+473

9866% positive

Medium

752 picks

+796

9907% positive

Low

740 picks

-417

1730% positive

CLV = the price we issued vs the closing line at game start. +100 bps means we got +1c better than the close on average. A sustained positive CLV is the cleanest forward-signal that the model is finding real edge — it's independent of whether any individual pick cashed. r298 ingest writes one row per settled pick; this panel is the all-time cross-sport view.

Edge honesty · 1,942 graded picks · 1077 price-less excluded

Did the edge cash?

FLAT 1u STAKE · WIN% INCLUDES PUSHES (DISCLOSED) · THIN BUCKETS (n < 30) DIMMED, NOT HIDDEN

By closing-line value

clv_bps vs the close

BucketnWin%Avg priceROI
< −100 bps54657.9%-181-7.9%
−100 – 07572%-145+25%
0 – 1008361.4%-122+9%
100 – 25017152%-112-8.5%
≥ 250 bps106752.5%-66-3%

By modeled edge

edge_bps at log time

BucketnWin%Avg priceROI
< 200 bps74058.9%-168-3.5%
200 – 40017556.6%-101+0.9%
400 – 70039852.8%-91-6.1%
≥ 700 bps62951.7%-50-2.4%

By confidence tier

producer-assigned

BucketnWin%Avg priceROI
sharp30153.8%-17+6%
high14943.6%-71-20.6%
medium75254.1%-93-3.4%
low74058.9%-168-3.5%

Unsmoothed, straight off the picks ledger. As of this build the CLV→cash gradient is NOT clean — the deep-negative bucket is heavy favorites at negative ROI and the middle buckets are thin. That result is published, not curated.

Edge honesty · 1,942 graded picks · 1077 missing a price, dropped from both series

Was our timing worth anything?

FLAT 1u PER PICK · LOSSES −1 AT BOTH PRICES (ONLY WINNING PAYOUTS DIFFER) · SAME PICK SET, TWO PRICES

Timing alpha

-20.2u

At our price

-64.0u

At the close

-43.8u

Max drawdown

−84.5u

Streaks

W16 / L9

at our logged price· at the close

The exact same picks, graded at the price we logged vs the closing price. Positive alpha would mean betting at publish beat betting at close — as of this build it is negative, and that ships too. Drawdown and streaks are on the logged series.

Bet-history simulator · cross-sport

Bankroll simulator · $100 per pick

Ending bankroll

$6796

Net profit

+$5796

ROI

13.4%

Picks

431

Simulates $100 per pick at standard −110 juice (+$91 win, −$100 loss) on every completed cross-sport model pick in the window. Real betting layers in book vig variance, line shopping, and bet-sizing — this is the "follow every pick flat" baseline.

Last 30 days · daily hit rate · cross-sport

431 games · 256 correct · 59.4%

50% baseline ←→ 100%

Download all picksCSV · MLB + NBA + WNBA + NHL + NFLMLB onlyNBA onlyWNBA onlyNHL onlyNFL only

Where to play

21+ · Bet responsibly · 1-800-GAMBLER

Predictions logged at game start · results compared to actual outcomes · no editing after the fact. Bet responsibly · 21+ · 1-800-GAMBLER

Get the graded slate in your inbox

Last night's record — wins and losses alike — plus tonight's slate. Free, double opt-in, one click to leave.

Support this work →