How It Works
A transparent look at the math, methodology, and data behind every pick.
Expected Value (+EV) Identification
Our model calculates the true probability of each outcome using machine learning trained on historical data, lineup information, weather, travel fatigue, and dozens of other factors. We only place a bet when our model probability exceeds the implied probability baked into the posted odds — creating a positive expected value (+EV) situation.
If a team has a true 55% chance of winning but the sportsbook prices them at -108 (implying 51.9%), there's a +3.1% edge. Over hundreds of bets, this compounds into real profit.
Fractional Kelly Criterion Sizing
Bankroll sizing is where most bettors fail. We use a fractional Kelly Criterion (0.25x Kelly) to size each bet proportionally to the edge. Bigger edge = bigger bet. No flat-betting, no gut-feel sizing.
Full Kelly maximizes long-run wealth but comes with extreme variance. At 1/4 Kelly, we preserve 75% of the theoretical growth rate while massively reducing drawdown risk.
Multi-Book Line Shopping
Getting the best price on every bet is non-negotiable. Even 5 extra cents in odds (-110 vs -115) dramatically impacts long-term ROI. We check DraftKings, FanDuel, BetMGM, Bet365, and Caesars for each pick.
Over 500 bets at an average of 3 cents better per bet = roughly 15-20 extra units of profit per year. Line shopping is free money.
Closing Line Value (CLV) Tracking
CLV is the gold standard metric for evaluating if your betting process is actually sound. If you consistently beat the closing line, you are almost certainly generating real edge — not just running lucky.
A bettor who beats CLV by +2 or more points is demonstrating genuine skill. Our current CLV average is +2.4, confirming the model is identifying mispriced lines before the market corrects.
Data Sources
The model is only as good as its inputs. We use best-in-class data providers.
Our Transparency Commitment
Every pick is posted before the game starts with a timestamp. No backdating, no retroactive record massaging.
We track every single bet — wins, losses, and pushes — in our public record. Nothing is hidden or omitted.
When the model is wrong, we say so. When a bet loses for reasons outside the model, we document it. Learning from losses is as important as capitalizing on wins.
Closing Line Value is tracked and reported because it is the most honest measure of whether our edge is real. If we lose CLV consistently, the model is broken — and we will say so.
Betting involves risk. Past results do not guarantee future performance. Please gamble responsibly.