The Season Simulator
Your edge decides whether you survive in the long run. Your bet sizing decides how long it takes to find out. That's a plain-English claim; this tool turns it into something you can actually watch happen — a real Monte Carlo simulation, not a canned animation, run a thousand times in your browser right now, so you can see what "risk of ruin" looks like as a spread of outcomes rather than a single number.
Type in an edge and a bet size — both percentages, nothing else — and watch a season's worth of bets play out a thousand different ways.
Every thin line is one simulated season. The shaded band shows where the middle 80% of all simulated seasons land at each point in the season.
This season (Monte Carlo)
Probability of hitting zero bankroll within the season length set above.
If sustained indefinitely, flat betting
The closed-form long-run risk of ruin at this edge and bet size, under flat betting — a different, longer horizon than the season above, shown for comparison, not conflated with it. Proportional betting has no equivalent closed form, so this figure is always the flat-betting benchmark, even when Proportional mode is selected above.
Median ending bankroll
As a multiple of the bankroll this season started with.
Flat betting risks a fixed percentage of your starting bankroll on every bet — the fixed-unit model the closed-form formula is built from, and the one that can reach an exact, absorbing zero. Proportional betting risks a fixed percentage of your current bankroll each time, the Kelly-style approach — mathematically it can approach zero but never exactly reach it in a finite number of bets. This tool treats a proportional bankroll that falls below 10% of its starting size as "functionally ruined," a convention chosen for this illustration rather than a rigorously derived figure.
Model: edge e is translated into a per-bet win probability p = (1+e)/2 on an even-money-equivalent wager — the standard derivation behind the gambler's-ruin closed form, RoR ≈ ((1−e)/(1+e))B/u. The Monte Carlo engine powering this chart was independently checked against that closed form, and against basic random-walk mean/variance identities, in a standalone verification harness before this tool shipped.