Lesson 15 of 153 min read

Risk Management and the Limits of Probability

Even a well-tested method loses, sometimes many times in a row. Risk management is what makes those streaks survivable, and honesty about probability is what stops you trusting numbers that were never measured.

For educational purposes only. This playbook explains a rule-based way to study price action. It is not financial advice, not a trading signal and not a promise of any result. No setup in it is guaranteed to work; practise it on a paper trading account first.

All lessons

What a probability claim needs

A statement like "this setup works 83% of the time" needs exact rules, a large documented sample, realistic costs and out-of-sample testing. Even then it describes the past, with uncertainty that is wide on small samples. A probability written next to a chart in someone's notes is an opinion, and the playbook never displays one.

Losing streaks are normal

Streaks happen even when nothing is wrong. As pure arithmetic: if a method won exactly half its trades, the chance of at least one run of five losses somewhere in 100 trades is about 81%. Plan for streaks before they happen.

Drawdowns also hurt more than they look: after a 20% drawdown you need a 25% gain to get back, and after 50% you need 100%.

Rules that protect the account

These rules are not optional extras; they are what keep any method alive long enough to be tested.

  • Risk a small, fixed fraction of equity per trade (often 0.5–1%).
  • Never increase risk after a loss, and never use martingale.
  • Set a daily or weekly loss limit and stop when it is hit.
  • Avoid oversized leverage; size from the stop, not from the margin available.
  • Do not trade a size you cannot calculate.
Risk per tradeAfter 5 losses in a rowAfter 10 in a rowAfter 20 in a row
0.5%−2.5%−4.9%−9.5%
1%−4.9%−9.6%−18.2%
2%−9.6%−18.3%−33.2%
5%−22.6%−40.1%−64.2%

Bullish example

  1. 1A bullish setup passes the checklist with a 2R target.
  2. 2You risk 1% of equity, sized from the stop distance.
  3. 3It loses. Your next trade is still 1%, not 2%.

Bearish example

  1. 1A bearish setup passes the checklist.
  2. 2You have already lost 3% today against a 3% daily limit.
  3. 3You skip it, however clean it looks.

What confirms it?

  • Risk per trade is fixed before entry.
  • Size is calculated from the stop distance and the contract specification.
  • Loss limits are written down and respected.

What invalidates it?

  • Risk is increased to recover a loss.
  • Size is chosen from available leverage instead of the stop.
  • A probability figure without a documented backtest is treated as fact.

Common mistakes

  • Believing a high claimed win rate means losses will be rare.
  • Risking 5% because "this one is a sure thing".
  • Martingale: doubling after each loss.
  • Removing the stop to avoid taking a loss.

Replay exercise

  1. In your paper trading account, write down your risk per trade, daily loss limit and weekly loss limit.
  2. Replay twenty sessions and apply them strictly, including stopping when a limit is hit.
  3. Review your largest losing streak and the drawdown it caused.

Quick quiz

Pick an answer to see the explanation. Nothing is saved.

A note says a setup "wins 89% of the time". What should you treat that as?

1. A note says a setup "wins 89% of the time". What should you treat that as?

At 2% risk per trade, roughly how much is lost after ten losses in a row?

2. At 2% risk per trade, roughly how much is lost after ten losses in a row?

After a loss, what should happen to risk on the next trade?

3. After a loss, what should happen to risk on the next trade?