Risk of Ruin Calculator
How To Use This Calculator
Describe the bets you place and how you size them. The calculator plays out thousands of possible sequences of wins and losses and shows how often your bankroll falls below the level you would call ruin, where it usually ends up, and how deep the worst drawdowns get.
Use the odds you typically bet at. The bankroll only scales the money figures; the percentages do not depend on it.
Type your expected value per bet, or switch to the true win probability if that is how you think about it.
Pick a Kelly fraction, a custom share of Kelly, or a fixed percentage of your current bankroll.
Choose how many bets to simulate and what loss counts as ruin. Losing 50% of the starting bankroll is the default.
What Is Risk of Ruin?
Risk of ruin is the probability that a run of losses takes your bankroll below a level you cannot or will not accept. A bet can be +EV and the bankroll can still fail, because the edge only shows up over many bets while a bad streak can arrive at any time.
If every stake is a percentage of your current bankroll, the bankroll never quite hits zero: stakes shrink as it falls. Total ruin is the wrong question for that kind of staking. What matters is the chance of a drawdown deep enough to make you quit or start over, which is why this calculator lets you set the threshold yourself.
Worked Example: Full Kelly vs Smaller Stakes
You bet at odds of 2.00 and your true win probability is 52.5%, an edge of 5% per bet. Full Kelly says to stake 5% of your bankroll. We simulated 100,000 runs of 1,000 bets at each stake size and counted how often the bankroll fell below half its starting value at any point.
| Stake per bet | Chance of losing 50% | Median bankroll after 1,000 bets |
|---|---|---|
| Full Kelly (5%) | About 41% | 3.5x start |
| Half Kelly (2.5%) | About 9% | 2.6x start |
| Quarter Kelly (1.25%) | About 0.4% | 1.7x start |
| 1% of bankroll | About 0.06% | 1.6x start |
Full Kelly has the highest median, but four runs in ten lose half the bankroll along the way. Cutting the stake to a fifth of that makes halving the bankroll rare, and the median still grows by more than half.
How to Read the Results
The share of simulated runs that dropped below your threshold at any point, even if they recovered later. A drawdown you would have quit at counts, whatever happened next.
Half the runs finish above this figure and half below. It is a better guide to a typical outcome than the average, which a few lucky runs pull upwards.
Nine runs in ten finish between these two figures. The low end shows a realistic bad outcome, not the worst one possible.
For each run, the biggest fall from a peak to a later low, averaged over all runs. Expect to live through a drop about this size.
What the Simulation Assumes
Every bet is independent and uses the same odds and edge. Correlated bets, such as several markets on one match, swing together and make deep drawdowns more likely than shown here.
The edge you enter is treated as the truth. If your real edge is half what you think, half Kelly on your estimate is full Kelly on reality, with full Kelly's drawdowns. Overestimating your edge is the most common way to oversize.
Stakes are recalculated from the current bankroll before every bet, so they shrink after losses and grow after wins. Real betting also has stake limits and bets that settle at different times.
Common Mistakes to Avoid
Full Kelly maximizes long-run growth, but the swings are severe. Halving the bankroll on the way is common, and most bettors quit before the long run arrives.
Kelly stakes scale with the edge you enter. An edge estimate that is too high quietly turns a conservative fraction into an aggressive one.
Even a solid edge can sit underwater for hundreds of bets. Run the simulation over the number of bets you will really place before you judge results.
Five bets on one match are closer to one big bet than five small ones. Size correlated positions as a group.
Frequently Asked Questions
What is risk of ruin in sports betting?
It is the probability that your bankroll falls below a chosen level before your edge has time to show. Ruin does not have to mean zero: a 50% drawdown is a common definition, because that is where many bettors stop or change course.
How does this risk of ruin calculator work?
It runs a Monte Carlo simulation in your browser. Each run plays out your number of bets with random wins and losses at your win probability, staking the chosen share of the current bankroll each time. The calculator counts how many runs crossed your threshold and summarizes where the rest ended up. The random numbers come from a fixed seed, so the same inputs always give the same result.
What is an acceptable risk of ruin?
There is no universal number. Set the threshold to the loss you would actually stop at, then choose a stake size that keeps the chance of reaching it small enough that you could live with it happening. For most people that means a fraction of Kelly rather than full Kelly.
If full Kelly grows fastest, why bet less?
Full Kelly maximizes long-run growth only if your edge is exactly right and you can stomach the swings. In the example above it loses half the bankroll in about four runs out of ten. Half or quarter Kelly gives up some growth for much smaller drawdowns and protection against an overestimated edge.
What happens if my edge estimate is wrong?
Kelly stakes are roughly proportional to the edge, so an overestimated edge means oversized stakes. If you think you have 5% and really have 2.5%, half Kelly on your estimate is full Kelly on the truth. Lower the edge in the calculator to see how quickly the risk climbs.
Should I bet a fixed percentage or a Kelly fraction?
Both stake a share of your current bankroll. A Kelly fraction adjusts that share to the odds and edge of each bet, while a fixed percentage stays the same. If your edge and odds vary a lot, Kelly sizes each bet more sensibly; if they are steady, a fixed percentage is simpler and behaves similarly.
Does the calculator handle correlated bets?
No. It assumes every bet is independent. Bets that move together, like several markets on the same game, make big drawdowns more likely than this simulation shows, so treat its numbers as a best case for a correlated portfolio.
Why does the chance of ruin show "<" a small number?
None of the simulated runs reached your threshold. That does not make the risk zero, only smaller than one run in the number simulated. Add more bets or lower the threshold to see where it starts to appear.
Stake size is the part you control
This calculator shows what an edge does to a bankroll at different stake sizes. Finding the edge is the hard part. Bet Hero compares bookmaker odds against devigged sharp lines across more than 400 books to find +EV bets, and suggests a stake using the Kelly fraction you choose in Settings.