What Is Expected Value (EV) in Betting?

Expected Value (EV) is the cornerstone concept of rational, profitable betting. It represents the average return you can expect per bet over an infinitely large sample, calculated by weighting each possible outcome by its probability. In essence, EV answers the question: "On average, how much profit (or loss) should I expect from this bet in the long run?"

A positive EV (+EV) bet is one where you expect to profit over time. A negative EV (-EV) bet is one where you expect to lose. This distinction separates professional bettors from casual gamblers. While casual bettors hope to pick winners, sharp bettors systematically hunt for bets with positive expected value — regardless of whether any individual wager wins or loses.

How Expected Value Works: The Coin Toss Analogy

Imagine a friend offers you a bet on a fair coin flip. If heads lands, you win £1. If tails lands, you lose £1. Since the coin is fair, heads and tails each have a 50% probability.

The expected value of this bet is:

• EV = (0.50 × £1) - (0.50 × £1) = £0.50 - £0.50 = £0

This is a break-even bet. Over 1,000 flips, you'd expect to neither profit nor lose.

Now imagine the same friend offers you better odds: if heads lands, you win £1.20. If tails lands, you still lose £1.

Now the expected value is:

• EV = (0.50 × £1.20) - (0.50 × £1) = £0.60 - £0.50 = +£0.10

Over 1,000 flips, you'd expect to profit £100. This is positive expected value. Over enough repetitions, the mathematical edge compounds into real profit.

Positive EV vs. Negative EV: The Fundamental Distinction

The difference between +EV and -EV is the difference between a sustainable betting career and guaranteed long-term loss.

How Do You Calculate Expected Value?

The Expected Value Formula for Binary Bets

The most common betting scenario is a binary outcome: you either win or lose. The formula for expected value is:

EV = (P(Win) × Profit) - (P(Loss) × Stake)

Where:

• P(Win) = Your estimated probability of winning (as a decimal, e.g., 0.55 for 55%)
• Profit = The amount you win if successful
• P(Loss) = Your estimated probability of losing (= 1 - P(Win))
• Stake = The amount you're risking

Worked Example: Calculating EV for a Real Bet

Let's say you believe a boxer has a 60% chance of winning a fight. The bookmaker prices the boxer at 2.00 (decimal odds), which implies a 50% probability.

Step 1: Identify your inputs

• Your estimated probability: 60% (0.60)
• Bookmaker's odds: 2.00
• Stake: £1

Step 2: Calculate profit if you win

• At 2.00 odds, a £1 bet returns £2 total
• Profit = £2 - £1 = £1

Step 3: Calculate loss if you lose

• You lose your stake = £1

Step 4: Apply the formula

• EV = (0.60 × £1) - (0.40 × £1)
• EV = £0.60 - £0.40
• EV = +£0.20

This means you expect to earn £0.20 per £1 staked, or a 20% return on investment on this single bet. If you placed 100 identical bets (same edge, different events), your expected profit would be £20 per £1 unit, regardless of which individual bets won or lost.

Expected Value for Multi-Leg Bets and Parlays

Calculating EV for multi-leg bets is more complex because probabilities multiply.

Example: A 2-Leg Parlay

Suppose you bet on:

• Bet A: 60% win probability at 2.00 odds (£0.20 EV as calculated above)
• Bet B: 70% win probability at 2.00 odds

For a parlay, you need BOTH bets to win:

• Combined win probability: 0.60 × 0.70 = 0.42 (42%)
• Parlay payout at 2.00 × 2.00: 4.00 (your £1 becomes £4)
• Profit if both win: £3

The EV of the parlay:

• EV = (0.42 × £3) - (0.58 × £1)
• EV = £1.26 - £0.58
• EV = +£0.68

Interestingly, this parlay has HIGHER EV than the individual bets combined (+£0.40). However, this is misleading. The parlay has lower win probability (42% vs. 65% average on individual bets), so it's riskier. This illustrates why most professional bettors avoid parlays — the variance is much higher.

Why Does Expected Value Matter in Betting?

The Edge Over Bookmakers

Every bet placed with a bookmaker carries a built-in margin, often called the "vig" or "hold." This margin is how bookmakers guarantee profit regardless of the outcome.

For example, a typical moneyline bet might be priced at -110 on both sides (American odds). This -110 on both sides means the implied probabilities add up to approximately 104.76%, not 100%. The extra 4.76% is the bookmaker's margin.

To break even on -110 odds, you need to win 52.38% of your bets. Any bet below this win rate is negative EV by default.

Finding positive EV means finding bets where the bookmaker's odds are more generous than the true probability warrants. This happens when:

1. You have superior analysis — Your model is better than the bookmaker's
2. Market inefficiency — The public's money has moved the line away from fair value
3. Slow line movement — You bet early before sharp money adjusts the line
4. Promotional offers — Free bets or enhanced odds create temporary +EV opportunities

Long-Term Profitability vs. Short-Term Variance

This is the most critical distinction in betting: Expected value is a long-run concept.

You can place 20 consecutive +EV bets and lose all of them. This is not evidence that your EV calculation was wrong — it's evidence that you haven't bet enough times yet. In any probabilistic system, short-term results deviate from expectation. This deviation is called variance.

Consider a simple +EV scenario:

• Bet on a 55% probability at 2.00 odds (EV = +£0.10 per £1)
• You place 10 bets of £1 each
• Possible outcomes: Win 4, lose 6 (a -£2 result despite +EV)
• Probability of this outcome: ~20%

This is entirely normal. The expected value only reveals itself across larger samples.

This is why bankroll management is crucial. You need enough capital to survive variance while your edge compounds.

Why Most Casual Bettors Ignore EV (And Why They Lose)

Most casual bettors approach betting like lottery players: they hope to pick the day's winners. They might research a game, feel confident about an outcome, and place a bet. If they win, they feel smart. If they lose, they feel unlucky.

Professional bettors approach betting like investors: they systematically hunt for mispriced lines. They don't care if they're right about the outcome of any individual game. They care about whether the odds offered exceed the true probability.

The difference is profound:

• Casual bettor: "I think Team A will win" → Places bet → Hopes for best
• Sharp bettor: "I estimate Team A at 55% probability, but odds imply 50%" → Places bet → Expects profit over 1,000 bets

Casual bettors lose because they're fighting against the bookmaker's margin with no edge. Sharp bettors win because they're finding +EV and letting variance work in their favour over time.

How Is Expected Value Different From Other Betting Metrics?

Expected Value vs. Implied Probability

Implied probability is the probability that the bookmaker's odds suggest. For example, 2.00 decimal odds imply a 50% probability (1 ÷ 2.00 = 0.50).

Expected value is the difference between your estimate and the implied probability.

Example:

• Bookmaker odds: 2.00 (implies 50% probability)
• Your estimate: 55% probability
• Difference: 5 percentage points
• EV: Positive (you're getting better odds than the true probability warrants)

If you estimated 45% probability instead, the EV would be negative, and you should avoid the bet.

Expected Value vs. Return on Investment (ROI)

Expected Value = Average profit per individual bet

Return on Investment (ROI) = Total profit ÷ Total amount wagered (across all bets)

These are related but different:

• A single +EV bet has positive expected value
• A portfolio of +EV bets has positive expected ROI (in the long run)
• But in the short term, you could have +EV bets and negative ROI due to variance

Example:

• You place 10 bets, each with +£0.10 EV (expected total profit: £1)
• Due to variance, you lose 7 and win 3 (actual result: -£0.70)
• Expected ROI: +10% (£1 profit ÷ £10 wagered)
• Actual ROI: -7% (-£0.70 ÷ £10 wagered)

Both metrics matter. EV tells you if you're making sound decisions. ROI tells you if those decisions are paying off. Over enough bets, positive EV becomes positive ROI.

Expected Value vs. Bookmaker Margin (The Vig)

The bookmaker margin (or "vig") is the built-in profit the bookmaker takes from the market. It's expressed as a percentage of total wagered.

For -110/-110 odds (common in American sports):

• Implied probabilities: 52.38% + 52.38% = 104.76%
• Margin: 4.76% (the excess over 100%)

Expected value is relative to this margin. A +EV bet is one where your edge exceeds the margin.

Example:

• Bookmaker margin: 4.76%
• Your edge: 5% (you think the true probability is 55%, but odds imply 50%)
• Result: Slight +EV after accounting for margin

If your edge is smaller than the margin, the bet is still -EV overall.

Can You Lose Money on Positive EV Bets?

Understanding Variance in Betting

Yes, absolutely. This is the most misunderstood aspect of expected value.

Variance is the natural fluctuation of results around the expected value. In a fair coin flip, you expect 50% heads and 50% tails. But in 10 flips, you might get 7 heads and 3 tails. This deviation from expectation is variance.

In betting, variance is even more pronounced because:

1. Probabilities are estimates, not certainties — You might be wrong about the true probability
2. Sample sizes are often small — A bettor might only place 50-100 bets per year
3. Odds vary — Different bets have different variance profiles

A real scenario:

• You identify 50 +EV bets, each with a 55% win probability and +£0.10 EV
• Expected profit: £5
• Possible actual outcomes:
  • Best case: Win 35 bets, lose 15 (profit: +£8.50)
  • Worst case: Win 20 bets, lose 30 (loss: -£5)
  • Probability of losing all £5: ~2%

The 2% chance of loss despite +EV is entirely normal. This is why bankroll management matters.

How Many Bets Until EV "Materialises"?

There's no fixed number, but it depends on your edge size:

These are rough estimates. Variance can extend or compress these timelines significantly. This is why professional bettors:

1. Maintain large bankrolls — To survive variance
2. Bet frequently — To accumulate samples quickly
3. Seek large edges — To reduce the sample size needed
4. Track everything — To verify EV is actually positive

The Gambler's Fallacy and Expected Value

A common mistake is thinking that past losses guarantee future wins. This is the gambler's fallacy.

If you've lost 10 consecutive +EV bets, the expected value of your next bet hasn't changed. The odds haven't shifted. Your probability estimate hasn't improved. The fact that you've lost 10 in a row doesn't make the 11th bet more likely to win.

Each bet is independent. This is fundamental to understanding variance. Losing streaks are normal and don't indicate a change in your edge.

How to Find Positive EV Bets in Practice

The Three-Step Process to Identify +EV

Step 1: Estimate True Probability

You need a method to estimate the true probability of an outcome. Methods include:

• Statistical models — Historical data, team strength ratings, player metrics
• Expert analysis — Injury reports, matchup analysis, situational factors
• Market comparison — Compare odds across sharp and soft bookmakers

Step 2: Convert Bookmaker Odds to Implied Probability

For decimal odds, implied probability = 1 ÷ Odds

Examples:

• 2.00 odds = 50% implied probability
• 1.91 odds = 52.36% implied probability
• 3.00 odds = 33.33% implied probability

Step 3: Compare and Calculate EV

If your estimate > implied probability, the bet is +EV.

EV = (Your Estimate × Profit) - ((1 - Your Estimate) × Stake)

Line Shopping Across Multiple Bookmakers

Odds vary across bookmakers. A 0.05 difference in decimal odds might seem small, but it significantly impacts EV.

Example:

• Bookmaker A: 2.00 odds (50% implied)
• Bookmaker B: 2.05 odds (48.78% implied)
• Your estimate: 52% probability

At Bookmaker A:

• EV = (0.52 × £1) - (0.48 × £1) = +£0.04

At Bookmaker B:

• EV = (0.52 × £1.05) - (0.48 × £1) = +£0.06

The difference is 50% higher EV for the same bet. This is why sharp bettors maintain accounts at multiple bookmakers.

Using EV Tools and Calculators

Modern betting platforms offer EV calculators that automate these calculations:

• OddsJam — Compares odds across books; identifies +EV
• Outlier — AI-powered +EV detection; Kelly Criterion integration
• OddsShopper — Portfolio EV tracking; market inefficiency detection

These tools are valuable, but they don't replace judgment. A calculator can tell you that odds imply 50% probability, but it can't tell you the true probability. That requires analysis and experience.

Expected Value and Bankroll Management

The Kelly Criterion and Expected Value

The Kelly Criterion is a formula that uses your expected value to calculate the optimal bet size:

Kelly % = (EV% + 1) / Odds in Decimal

Or more simply:

Kelly % = (Probability × Odds - 1) / (Odds - 1)

Example:

• Probability: 55% (0.55)
• Decimal odds: 2.00
• Kelly % = (0.55 × 2.00 - 1) / (2.00 - 1) = 0.10 / 1 = 10%

This means you should bet 10% of your bankroll on this wager.

Why Kelly matters: It maximizes long-term bankroll growth while accounting for variance. Bet too much, and a losing streak can bust you. Bet too little, and you're not capitalizing on your edge.

Most professional bettors use fractional Kelly (25-50% of Kelly) for safety, because:

1. Your EV estimate might be wrong
2. Variance can be severe with full Kelly
3. Fractional Kelly still compounds profits while reducing ruin risk

Bankroll Preservation When EV Isn't Certain

Not all +EV is created equal. A 1% edge is much riskier than a 5% edge.

Conservative bettors adjust their Kelly fraction based on confidence:

• High confidence (5%+ edge): Use 50% Kelly
• Moderate confidence (2-5% edge): Use 25% Kelly
• Low confidence (1-2% edge): Use 12.5% Kelly or avoid

This approach balances profit potential with ruin risk.

How EV Shapes Long-Term Bankroll Growth

Small edges compound dramatically over time.

Scenario: 2% EV on 500 bets per year, using 25% Kelly (average bet: 0.5% of bankroll)

The 2% edge compounds into significant long-term growth. This is why finding even small +EV is valuable.

Common Misconceptions About Expected Value

Misconception 1: "Positive EV Means I Will Win This Bet"

The myth: If a bet has +EV, it's more likely to win.

The reality: +EV means the odds are favourable relative to true probability, but the bet can still lose. A bet with 55% true probability and 2.00 odds is +EV, but it loses 45% of the time.

Why this matters: Bettors often confuse +EV with "likely to win." They then get frustrated when +EV bets lose and abandon the strategy. In reality, +EV bets are supposed to lose sometimes — that's variance.

Misconception 2: "If I Lose, My EV Calculation Was Wrong"

The myth: Losses prove the EV was miscalculated.

The reality: A single loss tells you nothing about EV. You need hundreds of bets to evaluate if your EV estimate was accurate.

Why this matters: This misconception causes bettors to second-guess sound decisions. They might abandon a +EV strategy after a few losses, missing the long-term profit.

Misconception 3: "All Bookmaker Bets Are the Same EV"

The myth: Every bet at a sportsbook has identical -EV because of the margin.

The reality: EV varies by bet type, odds offered, and market inefficiencies. A 2.00 favourite might be -EV, while a 2.00 underdog in an inefficient market might be +EV.

Why this matters: This misconception discourages bettors from looking for +EV. In reality, opportunities exist constantly — you just have to look.

Misconception 4: "EV Betting Requires Predicting the Future"

The myth: You need to predict game outcomes to find +EV.

The reality: You only need to estimate probability better than the odds imply. You don't need to be right about the outcome.

Why this matters: This misconception makes +EV betting seem impossible. In reality, it's about finding market inefficiencies, not predicting futures.

The History and Evolution of Expected Value

Origins in Probability Theory

Expected value originates in 17th-century probability theory. The concept emerged from a famous problem: the "Problem of Points."

Two gamblers are playing a game when interrupted. How should they divide the pot fairly? Blaise Pascal and Pierre de Fermat solved this by calculating the expected value of each player's position — the probability-weighted outcome.

This mathematical framework evolved into the formal definition of expected value used today in statistics, finance, and betting.

EV in Professional Gambling

For centuries, gambling was intuition-based. In the 20th century, professional gamblers and mathematicians began applying expected value systematically.

Key figures:

• Edward Thorp — Developed card-counting systems for blackjack using EV
• Sharp bettors — Applied EV to sports betting in the 1980s-1990s
• Modern quants — Use machine learning and EV to identify market inefficiencies

The transition from intuition to mathematics transformed betting from a game of luck into a game of edge.

The Modern Era: EV Tools and Accessibility

The internet democratized EV analysis. Today, bettors can:

• Access real-time odds across dozens of bookmakers
• Use automated EV calculators
• Compare sharp books vs. soft books instantly
• Build statistical models with public data

This accessibility has made +EV betting more competitive. The edge is smaller and harder to find, but it still exists for those who seek it systematically.

Practical Scenarios: EV in Different Betting Markets

EV in Moneyline Betting

Scenario: NFL moneyline, Kansas City Chiefs vs. Denver Broncos

• Sharp book (Pinnacle): Chiefs -105, Broncos -105 (both imply ~50%)
• Soft book (FanDuel): Chiefs -110, Broncos -110
• Your estimate: Chiefs 52%, Broncos 48%

At FanDuel Chiefs -110:

• Implied probability: 52.38%
• Your estimate: 52%
• EV: Slightly negative (you're slightly underestimating the Chiefs)

At Pinnacle Chiefs -105:

• Implied probability: 51.28%
• Your estimate: 52%
• EV: Slightly positive (you have a small edge)

This illustrates why line shopping matters. The 5-point difference in odds creates a meaningful EV difference.

EV in Point Spread Betting

Scenario: College basketball, Team A -5.5 (-110)

• Implied probability to cover: 52.38%
• Your model: Team A 54% to cover
• EV = (0.54 × £1) - (0.46 × £1) = +£0.08

To break even at -110, you need 52.38% win rate. Your 54% estimate gives you a small edge. Over 100 bets, you'd expect +£8 profit.

EV in Totals (Over/Under) Betting

Scenario: NFL game, Total 47.5 points, Over -110

• Implied probability: Over 47.5 points hits 52.38% of the time
• Your model: 54% probability of over
• EV: +£0.08 per £1 wagered

Totals markets are often less efficient than moneylines because:

1. Fewer bettors focus on totals
2. Public betting is more skewed (public loves overs)
3. Sharp money takes longer to adjust

This creates more +EV opportunities in totals than in moneylines.

Conclusion: Expected Value as Your Betting Foundation

Expected Value is not a guarantee. It's a mathematical framework that, applied consistently over hundreds of bets, separates profitable bettors from losing ones.

The key principles:

1. +EV is the only path to long-term profit — Betting without EV is gambling, not investing
2. Variance is normal — Short-term losses don't invalidate +EV
3. Sample size matters — You need enough bets to overcome variance
4. Bankroll management is critical — Use Kelly Criterion or fractional Kelly to size bets
5. Line shopping is essential — Odds vary; small differences compound
6. Discipline beats intuition — Systematic +EV hunting beats hope-based betting

If you understand expected value and apply it consistently, you've mastered the foundation of profitable betting. Everything else — statistical models, market analysis, bankroll growth — builds on this single concept.

The bookmakers know this. That's why they employ teams of mathematicians and statisticians. But they also know that most bettors ignore EV. If you don't, you've already gained an edge over 95% of the betting public.