Understanding EV% — The Only Number That Matters Long-Term
Most bettors track wins and losses. Sharp bettors track expected value. Here's why EV% is the only metric that tells you if you're actually betting correctly — and how PropPrizm uses it to surface the best opportunities every day.
The Core Problem With Tracking Wins and Losses
Imagine you flip a coin that's slightly weighted — it lands heads 55% of the time. You bet on heads every flip. After 20 flips, you might be down because you hit a rough patch of tails. Does that mean your strategy was wrong? No. You were right every single time you bet. The outcomes just didn't cooperate yet.
Prop betting works exactly the same way. A bet can be correct — meaning it has a positive expected value — and still lose. A bet can be wrong — meaning the book has the better of you — and still win. Over the short run, variance rules. Over hundreds of bets, math takes over. The only thing you can actually control is whether you're consistently betting with positive EV.
What Is Expected Value?
Expected value (EV) is the average amount you'd win or lose per dollar bet if you made the same bet an infinite number of times. A positive EV bet makes money in the long run. A negative EV bet loses money in the long run. It's that simple — and that important.
A prop is priced at -110 (standard juice). The book's implied probability is 52.4%.
Your model projects the true probability at 58%. The edge is 58% − 52.4% = 5.6%.
EV = (0.58 × $100) − (0.42 × $110) = $58 − $46.20 = +$11.80 per $110 risked
That's a +10.7% EV bet. Over 100 bets at $110 each, you'd expect to profit roughly $1,180.
How EV% Is Calculated in PropPrizm
PropPrizm's Edge Scanner shows EV% for every prop it tracks. Here's how it works:
- Model probability — PropPrizm generates a projection (e.g., 58% chance a pitcher exceeds his K line) using the matchup model
- Book implied probability — the book's line is converted to an implied probability, then de-vigged to remove the house edge
- EV% = model probability − book implied probability
When EV% is positive, the model believes the bet is underpriced by the book. When it's negative, the book has the better of you. PropPrizm only surfaces positive EV plays — it won't highlight a bet where the math is against you.
Important: EV% is calculated against the book's no-vig (fair) line, not the raw line. Stripping the vig gives a cleaner comparison between what the book believes and what the model believes.
Why a Positive EV Bet Can Still Lose
This is the part that trips up most bettors. Even a 70% favorite loses 30% of the time. A bet with +8% EV will have losing streaks. The key insight is that the outcome of any individual bet is not a signal about whether you made a good decision.
The math doesn't lie — it just takes time to manifest. This is why professional sports bettors think in terms of sample size and unit profitability, not individual bet outcomes. A bad week doesn't mean the model is broken. A bad week with consistently negative EV bets means something is wrong.
The Difference Between Positive EV and Picking Winners
Here's a paradox that confuses casual bettors: you can have a 45% win rate and be profitable. You can have a 60% win rate and be losing money.
How? It depends on the odds. If you're consistently betting +150 underdogs and winning 45% of the time, you're printing money. If you're consistently betting -200 favorites and winning 60% of the time, you're losing — because -200 requires a 66.7% win rate to break even.
This is why win rate alone is a misleading metric. EV% accounts for the price. A bet at +180 with a 42% true probability is a great bet. A bet at -130 with a 48% true probability is a bad bet. PropPrizm's Edge Scanner surfaces the former and filters out the latter.
How to Use EV% in Practice
Set a minimum EV threshold
Not all positive EV bets are worth taking. Model uncertainty, line movement risk, and the time cost of researching each play mean you should focus on bets where the edge is meaningful — generally 3% or higher. PropPrizm highlights bets in the 3–10% range as the most actionable.
Bet consistently sized units
Don't vary your bet size based on how confident you "feel." Confidence is not a reliable signal. Bet consistent units so the math can work in your favor over time without a single bad beat wiping you out.
Track EV, not just results
Log every bet with the EV% at the time you placed it. Over time, compare your actual results to your expected results. If you're consistently losing more than variance explains, the model may be wrong on that prop type. If you're close to expected value, the strategy is working even if you had a bad stretch.
Don't over-bet illiquid markets
Some props have low limits and wide bid-ask spreads. The EV% looks great on paper but you can only get a small amount down before the line moves against you. Focus on markets with enough liquidity to get your bets placed at the price you see.
Bottom line: EV% is the only number that tells you if your betting decisions are good or bad, independent of outcomes. If you consistently bet with positive EV, the long run takes care of itself. PropPrizm exists to find those opportunities every day.
What EV% Doesn't Tell You
EV% is powerful but not everything. It doesn't account for:
- Bankroll management — even a high-EV bet can be too large relative to your bankroll
- Correlation — multiple bets on the same game that all lose together can amplify downswings
- Model accuracy — EV% is only as good as the model generating it. Treat it as a signal, not a guarantee
- Line movement — if the line moves before you bet, the EV changes. Always check live prices before placing
Used alongside the matchup dashboard and a disciplined unit system, EV% is the closest thing to an objective compass for prop betting. Everything else is noise.
PropPrizm is a statistical tool and does not guarantee outcomes. Bet responsibly.