strategy · 6 min read · April 26, 2026

+EV Betting Explained

In short: A bet has positive expected value (+EV) when the offered odds are better than the true probability of the outcome. You find +EV bets by comparing a retail book’s odds to a sharp reference line. If the sharp line says a team has a 55% chance of winning but the retail book offers odds implying only 47%, the retail book is mispriced and your bet there is mathematically profitable. You won’t win every bet — but over hundreds of bets, the math wins.

Arbitrage betting guarantees profit on every bet by covering all outcomes. +EV betting takes a different approach: you place single bets where the math is in your favor and accept that you’ll lose some of them, because over time the wins more than compensate.

Think of it like a casino in reverse. A casino doesn’t win every hand of blackjack, but the house edge means they’re profitable across thousands of hands. +EV betting puts you in the casino’s position — small edge on every bet, volume does the rest.

What Expected Value Means

Expected value is the average amount you win or lose per bet if you could place the same bet thousands of times.

The formula:

EV = (probability of winning × profit if you win) − (probability of losing × amount you lose)

A worked example

Suppose the sharp consensus on a baseball game is essentially a 50/50 — both teams equally likely to win after vig removal. You find a retail sportsbook offering the underdog at +110, and you bet $100.

EV = (50% × $110) − (50% × $100)
EV = $55 − $50
EV = +$5.00 per bet

You won’t make exactly $5 every time — you’ll either win $110 or lose $100 on any individual game. But over hundreds of similar bets, you’ll average $5 profit per bet. The retail book’s +110 implies a 47.6% probability; the actual probability is 50%. That gap is your edge.

When the math runs the other way

Same hypothetical 50/50 game, but the underdog is offered at -110 (you risk $100 to win $90.91):

EV = (50% × $90.91) − (50% × $100)
EV = $45.45 − $50
EV = -$4.55 per bet

At -110 on a true 50/50 event, you lose an average of $4.55 per bet. That’s the vig — the sportsbook’s edge.

The key insight

A bet is +EV when the sportsbook’s odds imply a lower probability than the true probability. If the actual chance of an outcome is 50% but the book offers odds that imply 47.6%, the gap (50% − 47.6% = 2.4%) is your edge.

The challenge is figuring out the “true probability.” You don’t know it for certain. But you can estimate it from sources better than the bookmaker offering you the bet.

Sharp References: Finding the True Probability

You can’t know the exact probability the Lakers beat the Celtics tonight. But some estimates are much better than others, and the best estimates available come from the markets where the sharpest bettors are active.

Sharp books like Pinnacle accept large bets from professional bettors and adjust their lines based on the action. Their closing lines (the final odds when the event starts) have been hammered by the most informed bettors in the world. They represent the collective wisdom of people staking real money on their opinions.

Why trust sharp closing lines as the reference? The empirical answer is calibration: over thousands of games, teams the sharp closing line gives a 60% chance of winning actually win about 60% of the time. This has been documented across multiple sports and across decades of data. The market is genuinely efficient, in the Eugene Fama sense, within the limits of available information.

So the sharp closing line is the best available estimate of true probability. The vig-removed sharp price is what you compare retail prices to in order to identify mispriced lines.

Vig removal

Before you can compare a sharp line to a retail line, you need to strip the vig from the sharp book’s odds. The standard method:

true_prob = implied_prob ÷ combined_implied_prob

For Pinnacle’s -155/+135:

Implied (favorite): 60.78%
Implied (underdog): 42.55%
Combined: 103.33%

Vig-removed:

True (favorite): 60.78% ÷ 103.33% = 58.82%
True (underdog): 42.55% ÷ 103.33% = 41.18%

These vig-removed probabilities are your best estimate of the true outcome likelihood. Compare them to retail book implied probabilities to identify edges.

Important exception: Exchanges don’t charge vig in the traditional sense — they charge a commission on winnings. Their posted odds are already close to fair value. Don’t apply vig removal to exchange odds; you’d be removing vig that isn’t there.

A Real +EV Example

On April 15, 2026: Toronto Maple Leafs at Ottawa Senators. DraftKings offered Senators -1.5 (puck line) at +110.

The sharp consensus (Pinnacle, Circa, Novig, weighted) estimated the true probability of the Senators winning by 2 or more goals at 58.7%. DraftKings’ odds of +110 implied 47.6%.

Edge: 58.7% − 47.6% = 11.1 percentage points

In dollar terms:

EV = (58.7% × $110) − (41.3% × $100)
EV = $64.57 − $41.30
EV = +$23.27 per $100 wagered

A 23.3% edge. That’s enormous and rare — most +EV bets sit in the 3-8% range. But it’s the math at work: when the offered odds are far enough from the true probability, the edge is real.

The Senators won 3-1, covering the puck line. The bet cashed for $110 profit on a $100 stake.

But — and this is the part new +EV bettors find hardest to internalize — the bet would have been correct even if it had lost.

Why +EV Bettors Lose Individual Bets

In the Senators example, the math gave you a 58.7% chance of winning. That means a 41.3% chance of losing. The bet was correct, the price was good, and there was still a decent chance the Senators would lose the game outright (they did win, but only because the actual game played out a certain way).

If you placed this exact bet 100 times:

About 59 wins × $110 = $6,490
About 41 losses × $100 = $4,100
Net: $2,390 profit over 100 bets, or $23.90 per bet

That $23.90 average is your edge. But in the short term — 10 bets, 20 bets, even 50 bets — variance can make you look like a losing bettor. You might lose 7 of your first 10 +EV bets and be down $300. This is normal. It’s not evidence the strategy is broken.

The math only delivers reliably over volume. Hundreds of bets, not dozens.

Win rate ≠ profitability

A common confusion: a high win rate doesn’t mean you’re profitable, and a low win rate doesn’t mean you’re losing.

Scenario: betting only on heavy favorites at -200. Your win rate might be 70%. But each win pays only $50 on a $100 stake, while losses cost $100.

(70 × $50) − (30 × $100) = $3,500 − $3,000 = +$500 over 100 bets

Profitable, but barely.

Scenario: betting only on underdogs at +250. Your win rate might be 33%. Two-thirds of bets lose. But each win pays $250 against $100 risk:

(33 × $250) − (67 × $100) = $8,250 − $6,700 = +$1,550 over 100 bets

Three times as profitable, with a much lower win rate.

Both scenarios are +EV. The win rate is different, the emotional experience is different, but the math works in both. Experienced +EV bettors don’t obsess over win rate; they obsess over whether they’re consistently getting prices better than true probability.

How Variance Affects +EV Bettors

Even with a real edge, short-term results vary wildly. A +EV bettor with a 5% edge over hundreds of bets might:

Have stretches of 50 bets where they’re up 30% (better than expected)
Have stretches of 50 bets where they’re down 15% (worse than expected)
Land near the 5% edge over 500-1,000 bets

This is why closing line value (CLV) matters as a leading indicator. CLV tells you whether you’re consistently beating the closing line, which is strong evidence your bets are +EV — even when the win-loss results haven’t smoothed out yet.

A bettor with positive CLV but negative ROI over 200 bets is almost certainly experiencing a variance-driven downturn and will recover. A bettor with negative CLV but positive ROI over 200 bets is probably getting lucky and will eventually regress.

+EV vs Arbitrage: Tradeoffs

Both strategies exploit mispriced odds, but they make very different tradeoffs.

	Arbitrage	+EV
Risk per bet	Near zero	Real (you can lose any individual bet)
Edge per bet	1-5% typical	3-15% typical
Bets needed	Both sides simultaneously	One side only
Books needed	Two (one per side)	One
Detection risk	High	Lower
Variance	None	Significant short-term
Time to profitability	Immediate	Months (need volume)

Many serious bettors use both strategies. Arbing for steady, low-variance income; +EV for higher-margin gains. The skills transfer directly — both require understanding odds, vig, sharp references, and pricing differences across books.

Common Pitfalls

Using a bad reference line. If your “sharp reference” is stale or low-quality, every downstream calculation is wrong. Use the highest-quality sharp source available — Pinnacle’s closing line is the gold standard.

Ignoring exchange commission. Applying vig removal to exchange odds (which don’t have vig) inflates your edge estimate. Use raw implied probability for exchange references.

Drawing conclusions from small samples. Losing 50 +EV bets in a row means almost nothing. Real edge evaluation requires hundreds, ideally thousands, of bets.

Chasing high-edge bets without considering liquidity. A 20% edge on a $25-max prop isn’t worth the time to place. Sort opportunities by absolute expected value (edge × stake), not just edge percentage.

Forgetting about long-term book access. Every +EV bet at a retail book draws from a finite well. If a book limits your account, the bets you placed there had better have been worth the future access you gave up. This doesn’t mean avoiding retail books, but it does mean being thoughtful about where you place which bets.

Frequently Asked Questions

How is +EV different from “picking winners”?

Picking winners is about predicting outcomes. +EV is about finding mispriced odds. You don’t need to be better at predicting games than the sportsbook; you just need to identify when one book’s price is worse than the sharp consensus. The sharp line does the prediction work; you compare prices.

What edge percentage should I look for?

Anything above 2-3% (after accounting for vig) is potentially actionable. The sweet spot for most bettors is 3-10%. Very high stated edges (>15%) should be treated with skepticism — they often indicate a data error, stale line, or calculation mistake rather than a real opportunity.

How much can I expect to make?

Realistic expectations: with a $5,000 bankroll, 5-15 bets per day at 2-5% unit sizing and 5-8% average edge, you might expect $500-$1,500 per month after variance smooths out — over several months, not weeks. Results vary enormously based on bet volume, edge quality, book access, and bankroll management.

Can I do +EV and arbitrage at the same time?

Yes, and many serious bettors do. They use different bankroll allocations: arb capital is deployed for guaranteed small returns, +EV capital targets higher-edge bets with more variance. Same skills, different risk profiles.

Is +EV betting legal?

Yes. You’re placing legal bets at licensed sportsbooks. There’s no law against shopping for the best odds or comparing prices. Some books may limit your account if they identify you as a consistently winning bettor — but limiting is a business decision, not a legal issue.

Parcae is a sports betting intelligence tool currently in development. Learn more about what we’re building →

Sports betting involves risk. If you or someone you know has a gambling problem, contact the National Council on Problem Gambling at 1-800-522-4700.