Understand your risk of ruin before it costs you
Bankroll is a free, browser-based Monte-Carlo simulator for anyone sizing repeated bets
against an edge: sports bettors, small traders, and poker players moving up in stakes. You
set three numbers (your win probability, the payout ratio, and the fraction of your bankroll
you stake per bet) and it runs 10,000 independent equity paths, then draws them as a fan
chart so you see the full range of outcomes instead of a single average.
The headline figure counts how many of those 10,000 paths fall to zero. That percentage is
the number a plain expected-value calculation hides: a strategy with a genuine positive edge
can still go broke most of the time if you stake too much of your bankroll on each wager.
Bankroll makes that variance visible so you can pick a bet size you can actually survive.
How to use it
- Set the win probability and payout ratio to match the bet you actually make. The payout
ratio is how many units you win per unit staked, so an even-money bet is 1.0 to 1.
- Drag the bet-size slider and watch the risk-of-ruin counter and the fan chart react
live. Small stakes keep almost every path alive; large stakes widen the fan and push the
ruin number up quickly.
- Turn on "Snap to Kelly" to lock the bet size to the Kelly-optimal fraction for your
edge, then compare its risk of ruin against your manual size at the same edge.
- Copy the address bar to save or share a scenario. Every slider value lives in the URL,
so a link reproduces the exact simulation.
Frequently asked questions
- What is risk of ruin?
- Risk of ruin is the probability that a betting or trading strategy loses your whole
bankroll before you stop. Bankroll estimates it by simulation: it runs 10,000 sequences of
bets and reports the share that fall to effectively zero.
- What is the Kelly criterion?
- The Kelly criterion is the bet size that maximizes the long-run growth rate of a bankroll
for a given edge. For a bet paying b units per unit staked with win probability
p, it stakes a fraction (bp − q) / b, where
q = 1 − p. Bet more than Kelly and you take on variance without
extra growth; bet far more and ruin becomes likely.
- Does a positive edge guarantee a profit?
- No. Edge sets the direction of the average outcome, not the spread around it. With a real
edge but an oversized bet, an ordinary losing streak can bankrupt you before the edge pays
off. That gap between the average and the spread is exactly what the fan chart shows.
- Is my data sent anywhere?
- No. Every simulation runs locally in your browser using a Web Worker. There is no
account, no server, and nothing you type is uploaded or tracked.
- How accurate are the numbers?
- Each result is a Monte-Carlo estimate from up to 10,000 paths, so it carries sampling
noise and shifts slightly between runs. Raising the path count tightens the estimate;
lowering it speeds the simulation up on older devices.