Free tool

Losing streak calculator

A run of losses is not bad luck, it is arithmetic. This says how long a run is normal at your win rate, what it costs at the size you trade, and what size would leave you still trading afterwards.

Over how many trades

At a 45% win rate over 100 trades, a run of 6 losses is more likely than not. At 1% risk that costs 5.9% of the account.

Chance of each losing run over 100 trades, and what it costs
Losing runChance you hit itWhat it costs
1 in a row100.0%1.0%
2 in a row100.0%2.0%
3 in a row100.0%3.0%
4 in a row99.4%4.0%
5 in a row92.0%4.9%
6 in a rowmore likely than not72.9%5.9%
7 in a row49.7%6.8%
8 in a row30.7%7.7%
9 in a row17.9%8.6%
10 in a rowplan for this one10.1%9.6%
11 in a row5.6%10.5%
12 in a row3.1%11.4%

Add an account size for the money column. Every percentage above is the same without it.

Turning it around
Risk per trade that keeps a 6 loss run inside each drawdown
If the most that run can cost you isthen your risk per trade is
10%1.74%
20%3.65%
33%6.45%
50%10.91%

No row here is the right one. Which drawdown you can live with is yours to pick, and this only says what each one costs per trade.

This treats every trade as independent. Real losses cluster, so your true run is at least this long and often longer.

Why the streak is the number that matters

A 45% win rate sounds survivable until you work out what it implies. Over a hundred trades it makes a run of six losses more likely than not, and a run of nine roughly a one-in-ten event. Traders size for the average trade and then meet the run. The run is what closes accounts.

The math, in the open

The chance of a run is computed exactly, by walking every way the wins and losses can fall, not by simulating and not by an approximation that only holds over long horizons. The cost compounds: each loss is a percent of what is left, so ten losses at 2% costs 18.3% and not 20%. Turning it around, the risk that keeps a run inside a drawdown you name is the same formula read backwards.

The full working, and what it assumes

What it assumes, and where that breaks

It treats every trade as independent of the last one. Real trading is not: one bad regime, one tilted afternoon, one revenge entry after a loss, and the losses arrive together. That error runs one way only. Your real worst run is at least as long as this says and usually longer, so read the answer as a floor.

What a calculator cannot see

It used the win rate you typed. It has no way of knowing whether that is your win rate, and no way of telling you the number that beats every estimate on this page: the longest losing run already in your own history. Travise counts both from your closed trades, and says so when the run you have already lived through is longer than the one this page called normal.

Turn a risk percent into a share count with the sizing calculator.

Grade my first 7 trades free$0 today. Cancel before day 7.