Sports betting simulator
At −110 you have to win 52.38% of your bets just to break even: win half of them and you lose 4.55% of everything you stake. Enter your odds, your real win rate and how you stake, and see thousands of ways the same bets can turn out. Nothing here places a bet or picks a game.
Winning 50% of bets at −110, you finished ahead in 15.2% of simulations and lost 229 of your 1,000 on average.
These odds break even at a 52.38% win rate. At 50% the expected result is −4.55 per 100 staked; these simulations came to −4.59. Along the way, 14.6% of simulations were down to half the bankroll or less and 0% went broke.
Six sample simulations
Bankroll after each betWhere the simulations ended
Final bankrollHow the final bankrolls spread out across all simulations. The dashed line is the starting bankroll.
The same bets at other win rates
10,000 simulations each, same seed| Win rate | Ahead | Down to half | Average result | Per 100 staked |
|---|---|---|---|---|
| 0.3% | 76.3% | −609 | −12.22 | |
| 2.9% | 42.9% | −420 | −8.40 | |
| 15.2% | 14.6% | −229 | −4.59 | |
| 44.1% | 2.8% | −39 | −0.78 | |
| 51.1% | 1.9% | −3 | −0.05 | |
| 77.6% | 0.3% | +153 | +3.07 |
Only the win rate changes between rows. Below 52.38%, the break-even rate at these odds, no staking plan makes money on average.
The same bets with other staking plans
10,000 simulations each, same seed| Plan | Ahead | Down to half | Broke | Average result |
|---|---|---|---|---|
| 15.2% | 14.6% | 0% | −229 | |
| 11.2% | 3% | 0% | −205 | |
| No bets: no edge to stake on | ||||
| 20.2% | 84.2% | 79.2% | −431 | |
Every plan has the same expected result of −4.55 per 100 staked. They differ in how much they stake and how far the results spread.
Share this result
The link holds every setting and the seed (cgs), so anyone who opens it gets exactly these simulations.
How each staking plan bets
- Flat stakes
- The same stake on every bet. The baseline: what your win rate is worth with nothing else going on.
- Percent of bankroll
- A fixed percent of whatever the bankroll is now. Stakes grow after wins and shrink after losses, which slows a fall but cannot stop it.
- Kelly criterion
- The percent of the bankroll that grows it fastest in the long run: your edge divided by the net odds, here times the multiple you choose (0.5 is half Kelly). It needs your true win rate, and with no edge it stakes nothing.
- Martingale
- After each loss, the stake that wins back the whole losing streak plus the first bet’s profit: double at even money, 2.1 times at −110. Back to the first stake after a win.
What win rate do you need to break even?
At −110 you need to win 52.38% of your bets. The break-even win rate is 1 divided by the decimal odds: the shorter the price, the more often you have to be right.
| American | Decimal | Win rate to break even |
|---|---|---|
| −200 | 1.50 | 66.67% |
| −150 | 1.67 | 60.00% |
| −120 | 1.83 | 54.55% |
| −115 | 1.87 | 53.49% |
| −110 | 1.91 | 52.38% |
| −105 | 1.95 | 51.22% |
| +100 | 2.00 | 50.00% |
| +150 | 2.50 | 40.00% |
| +200 | 3.00 | 33.33% |
The second price in each row is rounded the way sportsbooks quote it.
How many bets before a record proves anything?
More than most bettors ever place. A bettor with no edge at all, picking winners like a coin flip, wins 55 or more of 100 bets 18.4% of the time.
With flat stakes at −110, the chance that the results point the wrong way after a number of bets:
| Bets | A 50% bettor is ahead | A 55% bettor is behind |
|---|---|---|
| 100 | 30.9% | 30.7% |
| 250 | 24.3% | 18.7% |
| 500 | 15.2% | 11.3% |
| 1,000 | 6.9% | 4.6% |
| 2,500 | 0.9% | 0.4% |
These figures are exact, from the binomial distribution. Point spreads and totals are set so that each side wins about half the time, which is why long-run win rates far above 50% are rare.
What the simulator assumes
- It takes your win rate as given. It cannot tell you what your real win rate is, and a few hundred bets cannot either, as the table above shows.
- Every bet is at the same odds and independent of the last, with no pushes. Real bets come at mixed prices, and sportsbooks may limit the stakes of accounts that keep winning.
- Kelly stakes are sized with the win rate you enter. If your true rate is lower, the same stakes are too big, which is why bettors who use Kelly usually stake a half or a quarter of it.