Risk of Ruin: How Likely Is Your Trading Strategy to Blow Up?

Risk of ruin is the probability that a run of losses shrinks your account to a point you cannot recover from, or will not keep trading through. It depends on three things: your edge (win rate and payoff), how much you risk per trade, and where you draw the ruin line. Even a genuinely profitable strategy has a high risk of ruin if it bets too much per trade, so this lesson puts numbers on that danger.

Defining ruin

Losing 100% is the obvious ruin, but it is not the practical one. Most traders stop, change strategy or lose confidence long before zero. A more useful definition is a drawdown limit: for example, a 50% fall from the account's peak. At that point the account needs a 100% gain to recover, and few people keep following their rules.

Losing streaks are normal

If the chance of losing a trade is q, the chance of N losses in a row is q to the power of N.

So a perfectly healthy trend-following system will, more likely than not, deliver eight losses in a row at some point in 200 trades. Your risk per trade must be small enough to make that boring rather than frightening.

The classic formula

For the simple case of fixed-size bets that win or lose the same amount, the gambler's ruin formula gives the chance of eventually losing your whole stake:

Risk of ruin = (q / p) ^ U

where p is the win probability, q = 1 - p and U is the number of betting units in the account.

Real trading is more complex, with different win and loss sizes and position sizes that change with the account, so simulation is more useful.

A simulation

DanioX Research ran 100,000 simulated sequences of 200 trades for a hypothetical strategy that wins 40% of the time at +2R and loses -1R: an expectancy of +0.2R per trade, which is a strong edge. Each trade risked a fixed percentage of current equity. The table shows the share of sequences that suffered a drawdown of 30% or 50% from a peak, and the median ending balance.

Risk per trade	30% drawdown	50% drawdown	Median ending balance
1%	0.3% of runs	under 0.1%	1.46x
2%	19%	0.5%	2.04x
5%	99%	50%	4.35x
10%	100%	99%	6.98x

Higher risk grew the median account faster, but at 5% half of all runs fell 50% from a peak, and at 10% almost every run did. Few people can sit through a halving of their account and keep following the rules.

Now the more important test: what if the true edge is smaller than you think? The same simulation with a 35% win rate (expectancy +0.05R):

Risk per trade	30% drawdown	50% drawdown	Median ending balance
1%	6.5%	under 0.1%	1.08x
2%	58%	8.8%	1.13x
5%	99.9%	85%	1.00x
10%	100%	100%	0.39x

At 10% risk, a strategy that still has a positive edge loses most of the money in the typical run. Live edges are usually smaller than backtests suggest, because of the biases in How to Backtest a Crypto Trading Strategy, so this second table is the one to plan for.

The Kelly criterion

The Kelly criterion gives the fraction of capital that maximises long-run growth, if your edge is known exactly:

Kelly fraction = p - q / b

where b is the ratio of average win to average loss.

Notice how sensitive it is: a five-point drop in win rate cuts the "optimal" size by three-quarters. Betting more than Kelly reduces growth and raises ruin risk; that is what happened at 5% and 10% risk in the second table. Because nobody knows their true edge, practitioners who use Kelly usually bet a small fraction of it, and 0.5% to 1% risk per trade sits comfortably below Kelly for most realistic strategies.

Rules that keep risk of ruin low

1. Risk a small fixed fraction of current equity, 0.5% to 1% per trade. As the account falls, positions shrink automatically.
2. Cap correlated exposure. Five altcoin longs at 1% each can behave like one 5% bet, as shown in Crypto Portfolio Construction.
3. Cut size in drawdowns, for example halve risk after a 10% fall from peak.
4. Avoid leverage that can skip your stop, as in Liquidation Price.
5. Plan for a smaller edge than your backtest, and for longer losing streaks.
6. Write a stop rule for the strategy itself: a drawdown or live-performance level at which you stop trading it and review.

Everything is now in place for the final lesson: Capstone: Build and Test Your Own Crypto Trading Plan.