Risk of Ruin Explained

At a glance

Definition: The probability of losing a defined portion of capital
Main driver: Risk per trade, far more than win rate
Practical target: Under 1 percent probability of a 50 percent loss
Key insight: Losing streaks are longer and more common than people expect

Key takeaways

A positive edge does not guarantee survival. Sizing too large can produce ruin even when every trade has positive expected value.
Losing streaks follow simple probability: with a 40 percent win rate, ten consecutive losses occurs roughly once every 165 trades.
Risk of ruin falls steeply as risk per trade falls, which is why the difference between 1 percent and 3 percent per trade is not a matter of degree.
Correlated positions compound the problem, because several positions can lose simultaneously and act as a single much larger trade.
Simulating your own strategy is more useful than any formula, because it captures the actual distribution of your trades.

Losing streaks are normal

Before discussing ruin, understand how common long losing runs are. Independent trades with a given win rate produce streaks that feel like evidence of failure and are simply arithmetic.

Probability of a streak of length k = (1 - win rate)^k

Win rate 40%, loss rate 60%:
   5 in a row   =  7.8%   -> happens roughly every 13 sequences
   8 in a row   =  1.7%   -> roughly every 60
  10 in a row   =  0.60%  -> roughly every 165
  12 in a row   =  0.22%  -> roughly every 460

Over 500 trades per year, a 10-trade losing streak is
not a warning sign.  It is the expected experience.

Expected longest streak in N trades is approximately:
   log(N) / -log(loss rate)
   For N = 500 and 60% loss rate: about 12 consecutive losses.

Expected longest losing streak in a series of trades.

Calculating risk of ruin

The classical formula assumes fixed bet sizes and binary outcomes, which approximates a trading strategy with a consistent risk percentage and a consistent payoff ratio. More useful in practice is a simulation, but the formula shows the shape of the relationship.

Risk per trade	Probability of 50% drawdown	Assessment
0.5%	Below 1%	Comfortable
1%	A few percent	Standard practice
2%	Roughly 10 to 20%	Aggressive
3%	Roughly 30 to 40%	Likely to end badly
5%	Over 60%	Ruin is the base case
10%	Near certain	Not a strategy

Approximate probability of a 50 percent account loss, for a strategy with a modest positive edge (45 percent win rate, 1.5R average win).

The nonlinearity is the point. Moving from 1 percent to 2 percent risk does not double the danger; it increases it by a much larger factor, because ruin depends on surviving compound sequences rather than on any single outcome.

Simulating your own risk of ruin

1
Collect your trade results in R terms
Use backtested and, where available, live results. At least 100 trades, expressed as multiples of the risk taken.
2
Resample with replacement
Draw a random sequence of the same length from your historical trades, thousands of times. This is a bootstrap Monte Carlo and it captures your actual distribution, including fat tails.
3
Apply your sizing rules
Compound the equity path using fixed fractional sizing, so position size changes as equity changes, exactly as it would live.
4
Record the distribution of outcomes
The proportion of paths that hit your ruin threshold is your risk of ruin. The 95th percentile drawdown is your realistic worst case.
5
Adjust size until the numbers are acceptable
Target under 1 percent probability of the drawdown level at which you would stop trading. Reduce risk per trade until the simulation agrees.
6
Repeat with correlated clustering
If your strategy takes several positions at once, resample by day rather than by trade, so simultaneous losses stay together. This produces a materially worse and more realistic picture.

What the formula misses

Correlated positions. Several trades losing on the same day behave like one large trade, which the independent-trade assumption ignores.
Fat tails. Real trade distributions contain occasional losses far larger than the planned risk, from gaps, slippage, and halts.
Edge decay. The formula assumes a constant edge. A strategy whose edge has disappeared has a much higher ruin probability than its historical statistics suggest.
Behavioural ruin. Many accounts are not lost through a mathematical sequence but through a single undisciplined decision after a drawdown.
Leverage interaction. Leverage can produce liquidation before the account reaches the ruin threshold, which the formula does not model.
Adding capital. Topping up an account during a drawdown changes the calculation entirely and usually increases total losses.

Frequently asked questions

What is an acceptable risk of ruin?

Under 1 percent probability of reaching the drawdown level at which you would stop trading. Some practitioners target under 0.1 percent. The exact threshold matters less than the discipline of computing it, because the calculation almost always shows that common risk levels of 2 to 3 percent per trade are far more dangerous than they feel.

Can I have a positive edge and still go broke?

Yes, and this is the central point. Sizing too large means a normal losing streak reduces capital enough that the remaining edge cannot recover it. This is why professional risk management focuses on survival first: an edge only compounds if you are still trading when it works.

How long should I expect my worst losing streak to be?

Approximately the logarithm of your trade count divided by the negative logarithm of your loss rate. For 500 trades a year at a 60 percent loss rate, that is around twelve consecutive losses. Knowing this number in advance turns a distressing experience into an anticipated one.

Does diversification reduce risk of ruin?

Genuine diversification does, by reducing the probability that many positions lose simultaneously. Apparent diversification across correlated positions does not, and can increase it by creating a false sense of safety that justifies larger total exposure. Measure the correlation before counting the benefit.

Should I add money to my account after a drawdown?

Only on a predetermined schedule, never in response to losses. Topping up after a drawdown increases the total amount at risk in a strategy that has just demonstrated it might not work, and it removes the natural limit that a bounded account provides. Scheduled contributions are fine; reactive ones are how bounded experiments become unbounded ones.

Test this idea before you trade it

Describe the rules in plain language and AlgoTrader AI turns them into a structured strategy blueprint with a configurable historical backtest, cost assumptions, and exportable code.

Build a backtest

Keep reading

Referenced by

Educational use only. This guide explains how a strategy works. It is not investment advice, not a recommendation, and no result described here is a forecast. Test any approach on historical and out-of-sample data, size positions conservatively, and never risk money you cannot afford to lose.