RiskKit field guide / Sequence risk

How many losses in a row should your trading plan survive?

A positive average does not prevent a rough sequence. Plan for losing streaks by separating the probability of one specified run from the chance that a run appears somewhere in your trading horizon.

The short version

With a constant 45% win rate and independent trades, the model gives about a 72.9% chance of at least one six-loss run somewhere in 100 trades. The chance that the next six trades all lose is only 2.8%. These answer different questions; neither predicts when a real losing streak will occur.

Two streak questions that sound alike

Suppose the probability of a losing trade is 55%. A particular block of six trades all losing has probability 0.556, or about 2.8%. That is the right answer only when you name the block in advance—for example, your next six trades.

In 100 trades, a six-trade run can begin at many different positions. Those candidate runs overlap, so multiplying 2.8% by the number of possible starting positions is wrong. The Losing Streak Calculator instead tracks the probability of being on a current run of zero, one, two and more losses after each trade. Once a path reaches six losses, it belongs to the “streak occurred” group. Under the stated assumptions, that yields 72.9% for at least one such run in 100 trades.

Same win rate · 100 trades

At least one four-loss run: 99.4%. At least one six-loss run: 72.9%. At least one eight-loss run: 30.7%. A longer streak is less likely, but still possible.

Turn the streak into a balance stress test

Probability is only half the planning question. The other half is whether your account and trading rules can tolerate the losses. If you risk a fixed percentage of current equity and every planned stop is filled exactly, the balance after k full losses is starting balance × (1 − risk per trade)k.

On a $10,000 account, six losses at 1% of then-current equity leave about $9,414.80, a 5.9% decline. At 2% per trade, the same streak leaves about $8,858.42, an 11.4% decline. The percentage risk is recalculated after each loss; these are not six identical dollar losses.

CME Group's controlling-risk lesson uses the same fixed-percentage principle to show how position risk shrinks with the account during a losing streak. In practice, spreads, slippage, gaps and commissions can make realised losses differ from a neat percentage.

Know which assumptions can fail

  • Constant win rate: a backtested win rate is an estimate, not a property guaranteed to persist. Market conditions and execution can change it.
  • Independent trades: this model treats each win or loss as unrelated to the previous result. Trades driven by the same market regime can be correlated.
  • Binary outcomes: a trade is classified as a win or loss for this calculation. The size of each win does not affect streak probability, while the size of a loss matters greatly for capital survival.
  • Exact planned loss: the balance example assumes every loss equals the specified fraction of current equity. It does not model gaps, liquidation, funding costs or a broker's margin rules.

Try a lower win rate and a longer horizon to see how sensitive the streak probability is. If the account result would force you to abandon the strategy, reduce planned risk or revisit the rules before trading. A simulation is a stress test for decisions, not proof that a strategy is safe.

Use the tools together

Start with the Trading Expectancy Calculator to examine the average R from your win rate and realised payoff. Then use the Losing Streak Calculator for a specific loss run. Finally, test changing account size, drawdowns and a stop-trading threshold in the Fixed Risk Simulator. The Monte Carlo guide explains why sampled paths are conditional scenarios rather than forecasts.