Performance

The scoreboard
doesn't lie.

We track every prediction we make. No cherry-picking. No revisionist history. Just a running record of how our models perform against the market — updated continuously.

NFL Simulator — Live Performance
ATS Accuracy65%
Break-Even Threshold52.4%
Edge Above Market+12.6pts
Evaluation Period2025–26
Prediction TypeATS

Predictions evaluated against closing line at primary sharp market. All results are out-of-sample.

ATS by Edge Threshold
3+ Point Edge (864 games)66.8%
5+ Point Edge (578 games)69.6%
8+ Point Edge (264 games)70.5%
3+ Edge Taking the Dog (458 games)68.6%
Over/Under ATS
Overall (1,411 games, p<0.001)58.6%
3+ Point Edge (781 games)63.1%
5+ Point Edge (411 games)66.7%
3+ Edge Over Picks (173 games)71.7%

All numbers are walk-forward out-of-sample — no in-sample contamination. Vegas baseline: ~52.4%.

Rolling ATS Win Rate
Week 152.4% break-evenWeek 16
CBB Simulator — 2025–26 Season

1,283 games
simulated.

Simulation Accuracy
Straight Up66.5%
Against the Spread66.6%
Over/Under49.4%
NCAA Tournament — 78.8%
First Four75.0%
Round of 6477.4%
Round of 3281.2%
Sweet 1675.0%
Elite 875.0%
Final Four100%
Championship100%
Championship Pick

Michigan over UConn (69-63) — simulator had Michigan -2.9

Season accuracy: 853/1,283 straight up, 469/704 ATS, 348/704 O/U. NCAA Tournament: 52/66 correct predictions across all rounds.

UFC Simulator — Notable Picks

Where the
edge showed up.

+500
Tresean Gore vs Azamat Bekoev
Biggest edge of the season at +28.8% — won by SUB R3
Apr 4
+260
Tofiq Musayev vs Ignacio Bahamondes
LOW confidence underdog — won by unanimous decision
Mar 28
+124
Joe Pyfer vs Israel Adesanya
HIGH confidence pick against a former champ — won by KO R2
Mar 28
-117
Casey O'Neill vs Gabriella Fernandes
HIGH confidence — won by KO R1
Mar 28
-180
Terrance McKinney vs Kyle Nelson
HIGH confidence — won by KO R1
Mar 28
-122
Dione Barbosa vs Melissa Gatto
MED confidence — won by majority decision
Apr 4
-183
Michael Page vs Sam Patterson
MED confidence — won by unanimous decision
Mar 21
Biggest Miss

Maycee Barber -183 vs Alexa Grasso (Mar 28) — HIGH confidence pick that lost by KO R1.

Led to post-mortem fixes: KO rate floor, quality-of-loss weighting, and skill-usage gap modeling.

NBA Simulator — 2 Seasons Live

1,176 games.
28.9M simulations.

Performance
Winner Prediction64%
High-Confidence Picks68.7%
Last 2 Months76.6%
Longest Win Streak18
ROI — Season 1+3.1%
ROI — Season 2+2.4%
Infrastructure
Simulations Run28.9M+
Games Analyzed1,176
Player Box Scores53,000+
Odds Data Points55,000+
Sportsbooks Tracked12
Players Tracked628
Transparency Note

During the 2025–2026 season, we discovered and corrected a model bug that had inflated our ATS accuracy. The corrected figure — 65% — reflects our actual, validated performance.

We publish this correction because it's the right thing to do, and because it's the kind of transparency that separates serious analytics operations from black-box pick services. The model is actively improving through continued experimentation.

Evaluation Methodology

All performance metrics are evaluated against closing lines — the sharpest available market signal — rather than opening lines. This is the honest test.

Closing Line Value (CLV) is the gold standard for measuring predictive edge because it captures the final consensus of all market information. A model that consistently beats the closing line is generating real alpha, not exploiting stale openers.

Why we
publish this.

Most analytics companies don't show you their track record. We do, because edge is verifiable — and because the clients we want to work with know the difference between a real model and a marketing claim.

Request Full Performance Documentation