True odds are normalized to probabilities across the two outcomes. EV is computed as p × bet_odds − 1. If a Kelly multiplier is provided, stake = min(max % per bet, (Kelly multiplier × EV decimal) / (bet odds − 1)) × current bankroll. Otherwise, a flat stake equal to (max % × bankroll) is used on every bet.
The EV simulator shows how a betting edge plays out over many bets. Enter the true odds of a two-outcome market, the odds you are getting, your starting bankroll and a staking rule, and it simulates the run of bets and charts your bankroll, with the final bankroll, profit, ROI, total staked and maximum drawdown.
The simulator turns the true odds into probabilities, works out the EV of your bet and sizes each stake from your current bankroll:
True probability (p) = (1 ÷ true odds) ÷ (1 ÷ true odds A + 1 ÷ true odds B)
EV per bet = p × bet odds − 1
Kelly stake = min(max % per bet, Kelly multiplier × EV ÷ (bet odds − 1)) × bankroll
Without a Kelly multiplier: stake = max % per bet × bankrollEach bet wins with probability p. A win adds stake × (bet odds − 1) to the bankroll and a loss removes the stake, so stakes rise and fall with the bankroll. Enter a seed to repeat exactly the same sequence of results.
A positive EV bet can still lose many times in a row. Running the same edge with different seeds shows how wide the range of outcomes is, and comparing full, half and quarter Kelly shows how much drawdown each staking level brings. Use it to choose a stake size you can stick with before you bet real money.
We ran one set of inputs 20,000 times with the simulator's formula: true odds 2 and 2, betting odds 2.10 (+5% EV per bet), a $1,000 starting bankroll, a 10% maximum per bet and 1,000 bets. Only the Kelly multiplier changes.
| Kelly multiplier | Stake (% of bankroll) | Median final bankroll | Runs that lost money | Median max drawdown |
|---|---|---|---|---|
| 0.25 | 1.14% | $1,644 | 9.6% | 24.8% |
| 0.5 | 2.27% | $2,343 | 13.4% | 45.0% |
| 1 (full Kelly) | 4.55% | $3,111 | 23.6% | 73.5% |
| 2 | 9.09% | $1,000 | 49.2% | 96.8% |
Full Kelly triples the median bankroll, but 97% of its runs fall by half at some point. At double Kelly the same +5% edge ends where it started and half of all runs lose money. Quarter Kelly holds the median drawdown to about a quarter of the bankroll. The bankroll management lesson compares these with flat and proportional staking.
Each run draws new random results. Enter a seed to get the same sequence every time, which makes it easier to compare staking settings.
Many bettors use a fraction of full Kelly, such as 0.25 or 0.5, to reduce drawdowns. Run several multipliers with the same seed and compare the final bankroll and maximum drawdown.
Yes. EV is an average, so a positive edge can finish behind over a short run, especially with large stakes or long odds. The more bets you simulate, the closer the results tend to get to the expected value.
Remove the margin from a sharp bookmaker's prices with the vig calculator, or use the fair odds behind the bets in the positive EV finder.