Module 12: Risk Management and Trading Psychology
Drawdowns, Losing Streaks, and Risk of Ruin
Chapter 6 of 14 · about 5 minutes
6.1 Seven in a row
Mint had been following a tested strategy that won about half the time. Then Mint lost seven trades in a row.
"The strategy is broken," Mint said. "Seven losses can't be bad luck."
"Let's check," said Hash. "With a 50% win rate over 100 trades, the chance of hitting at least one streak of seven losses is about one in three. It's not rare at all. The question is whether you sized your trades so that seven losses didn't hurt you badly."
6.2 Losing streaks are normal
People underestimate how often long losing streaks happen. Here are simulated probabilities of experiencing at least one losing streak of a given length during 100 trades:
| Win rate | Streak of 5 or more losses | Streak of 7 or more | Streak of 10 or more |
|---|---|---|---|
| 50% | ~81% | ~32% | ~5% |
| 40% | ~98% | ~69% | ~21% |
A trader with a 40% win rate, which can be perfectly profitable with good risk-reward (Module 4, Chapter 16), will almost certainly face five losses in a row, and is more likely than not to face seven.
"So a long losing streak doesn't mean my strategy stopped working," said Mint.
"Not necessarily," said Hash. "It might. But you need to know what normal looks like before you can tell the difference."
6.3 Drawdowns
A drawdown is the fall from a peak in your account value to a subsequent low (Module 1, Chapter 13). Maximum drawdown is the largest such fall over a period.
| Value | |
|---|---|
| Account peak | $25,000 |
| Lowest point afterwards | $20,000 |
| Drawdown | $5,000, or 20% |
Drawdowns matter for two reasons: the maths of recovery (Chapter 1), and psychology. Many traders abandon good strategies during drawdowns, often right before they recover.
6.4 How streaks become drawdowns
The same losing streak creates very different drawdowns depending on risk per trade.
| Consecutive losses | Drawdown at 1% risk | At 2% risk | At 5% risk | At 10% risk |
|---|---|---|---|---|
| 5 | ~4.9% | ~9.6% | ~22.6% | ~41.0% |
| 7 | ~6.8% | ~13.2% | ~30.2% | ~52.2% |
| 10 | ~9.6% | ~18.3% | ~40.1% | ~65.1% |
At 1% risk, ten losses in a row is uncomfortable but recoverable. At 10% risk, it's devastating.
Chapter 6 · Drawdowns and losing streaks
The same ten-loss streak produces very different drawdowns depending on risk per trade. At 1% risk the account falls only modestly, while at 5% or 10% risk the same streak carves a deep drawdown that is hard to recover from, because losses compound.
6.5 Risk of ruin
Risk of ruin is the probability that losses reach a level you can't recover from, or won't tolerate. Traders often define "ruin" as a large drawdown, such as 50%, rather than total loss.
Chapter 3 showed simulated results for a profitable strategy that wins 45% of the time with winners twice the size of losers:
| Risk per trade | Approximate chance of a 50% drawdown over 200 trades |
|---|---|
| 1% | Close to 0% |
| 2% | Close to 0% |
| 3% | Under 1% |
| 5% | ~19% |
| 10% | ~94% |
The lesson: even a strategy with a genuine edge has a high risk of ruin if bets are too large.
6.6 The Kelly criterion, and why traders use less
The Kelly criterion is a formula for the bet size that maximises long-term growth, given a known win rate and payoff.
Kelly fraction = win rate − (loss rate ÷ win-to-loss ratio)
For the strategy above: 0.45 − (0.55 ÷ 2) = 0.175, or 17.5% of the account per trade.
That sounds very different from the 1% to 2% rule. Why?
- Kelly assumes you know your exact win rate and payoff. In reality, you're estimating them from limited data, and you're often overestimating.
- Full Kelly produces enormous drawdowns. Even when correct, it's an extremely bumpy ride.
- Markets change. A strategy's edge can shrink or disappear.
Professional traders who use Kelly typically use a small fraction of it. Many simply use fixed risk of 1% to 2%, which is far below Kelly and much more robust to errors.
6.7 Planning for drawdowns
Hash suggested writing a drawdown plan in advance:
| Drawdown from peak | Action (example) |
|---|---|
| 10% | Review recent trades in the journal; check whether you're following your rules |
| 15% | Reduce risk per trade by half |
| 20% | Stop trading for at least a week; full review of strategy and behaviour |
| 30% | Stop active trading; reassess whether the strategy has an edge |
For long-term investors, a drawdown plan might instead be: "During a 50% decline, I won't sell, I'll continue planned buying, and I'll review my thesis, not the price."
"So I decide in advance what to do when it hurts," said Mint.
"Because when it hurts, you won't think clearly," said Hash.
6.8 Try it yourself
Using your own trading journal, or a hypothetical one, find your longest losing streak and your maximum drawdown. Then use the table in section 6.4 to estimate what a ten-trade losing streak would do to your account at your current risk per trade. Write your own drawdown plan with at least three levels and specific actions.
Key takeaways from this chapter
- Long losing streaks are normal: at a 40% win rate, a streak of seven losses in 100 trades is more likely than not.
- A drawdown is the fall from a peak to a low; deep drawdowns are hard to recover from and lead many to abandon good strategies.
- Risk per trade determines how much a losing streak hurts.
- Risk of ruin rises sharply with bet size, even for profitable strategies.
- The Kelly criterion suggests much larger bets than most traders use, because real-world estimates are uncertain and full Kelly is extremely volatile.
- Write a drawdown plan in advance, with specific actions at each level.