Expected Value Calculator
Calculate if a prediction market bet is +EV based on your estimated true probability
You believe this outcome has a 65% chance, but the market prices it at 52.0%. If you are right, betting $100 has an expected profit of $25 (+25.0% ROI). You need to be right at least 52.0% of the time to break even.
Edge = Your Probability - Market Probability
Max Profit = (Stake / Market Price) - Stake
Expected Profit = P(win) x Max Profit - P(lose) x Stake
Expected ROI = (Expected Profit - Fees) / Stake x 100
Break-Even = Market Prob x (1 + Fee%)
What is Expected Value?
Expected Value (EV) is the average amount you can expect to win or lose per bet if you were to place the same bet many times. A positive expected value (+EV) means you expect to profit over the long run, while a negative expected value (-EV) means you expect to lose. In prediction markets, EV is calculated by comparing the market's implied probability (the price you pay) with your own estimate of the true probability. If you believe an outcome is more likely than the market suggests, the bet is +EV. Professional traders and bettors focus exclusively on making +EV bets, knowing that while individual outcomes are uncertain, the math works in their favor over many bets.
How to Estimate True Probability
The key to finding +EV bets is accurately estimating the true probability of an outcome. This is both the hardest and most important skill in prediction market trading. Start by looking at base rates: how often has this type of event happened historically? Then adjust for specific factors that make this situation different. Consider multiple information sources, expert opinions, and statistical models. Be honest about your uncertainty. A common mistake is being overconfident in your estimates. If you are not sure, it is better to assume the market is efficient and pass on the bet. Your edge should come from genuine information advantages, not wishful thinking.
Understanding Edge in Prediction Markets
Edge is the difference between your estimated true probability and the market's implied probability. If the market prices an outcome at 50% but you believe the true probability is 60%, your edge is +10 percentage points. The larger your edge, the more profitable the bet is in expectation. However, edge alone does not tell the full story. A 2% edge on a 50/50 market is very different from a 2% edge on a 95/5 market. The EV calculation combines edge with the potential payouts to give you the complete picture. On platforms like Polymarket, the price you pay IS the implied probability (52 cents = 52% implied probability), making it straightforward to calculate your edge.
Why +EV Does Not Mean Guaranteed Profit
A +EV bet is one you should make, but it does not guarantee you will win. Even with a large edge, any individual bet can lose. If you believe an outcome has a 70% chance and buy at 50 cents, this is a great +EV bet but you will still lose 30% of the time. The key insight is that over many +EV bets, the law of large numbers ensures your results converge to the expected value. This is why bankroll management is critical. You need to survive the short-term variance to realize your long-term edge. A single bad outcome should not wipe out your ability to keep betting. Think in terms of hundreds or thousands of bets, not individual outcomes.
Bankroll Management for Prediction Markets
Even with +EV bets, poor bankroll management can lead to ruin. The Kelly Criterion is a widely used formula that suggests betting a fraction of your bankroll proportional to your edge. For a simple binary bet, the Kelly fraction is: (edge / odds) or more precisely (bp - q) / b, where b is the odds received, p is your estimated probability of winning, and q is (1 - p). Many professional bettors use fractional Kelly (half or quarter Kelly) to reduce variance at the cost of slightly lower expected growth. Never bet more than 5% of your total bankroll on a single prediction market, regardless of how confident you are. Diversify across uncorrelated markets to smooth out variance and protect against model errors in any single estimate.