What drawdown is normal at 40 trades a day
Posted: Sat Oct 31, 2026 8:35 am
Common misconception: lots of trades smooth the equity curve, so drawdowns must stay small. Smoother per trade, maybe, but a high trade count also means streaks get many more chances to happen.
Quick maths with a 55 percent win rate and independent trades. The chance of 8 losses in a row from a given start is 0.45 to the power of 8, about 0.17 percent. Looks negligible. At 40 trades a day that is roughly 800 trades a month, and the expected number of 8-loss runs is about 800 x 0.55 x 0.0017, so around 0.7 a month. More likely than not within two months of normal trading with nothing broken.
At 0.5 percent risk per trade, 8 straight losses is 4 percent off the account from that streak alone. Mixed sequences, such as 14 losses in 20 trades, can dig deeper than a pure streak.
Independence is the flattering assumption. Real results cluster by regime, so drawdowns tend to arrive heavier and in clumps than this maths says. Treat the figure as a floor on what is normal, and size the account limits to it.
What maximum drawdown in R have you actually logged across a few thousand trades, and how did it compare with the streak maths?
Quick maths with a 55 percent win rate and independent trades. The chance of 8 losses in a row from a given start is 0.45 to the power of 8, about 0.17 percent. Looks negligible. At 40 trades a day that is roughly 800 trades a month, and the expected number of 8-loss runs is about 800 x 0.55 x 0.0017, so around 0.7 a month. More likely than not within two months of normal trading with nothing broken.
At 0.5 percent risk per trade, 8 straight losses is 4 percent off the account from that streak alone. Mixed sequences, such as 14 losses in 20 trades, can dig deeper than a pure streak.
Independence is the flattering assumption. Real results cluster by regime, so drawdowns tend to arrive heavier and in clumps than this maths says. Treat the figure as a floor on what is normal, and size the account limits to it.
What maximum drawdown in R have you actually logged across a few thousand trades, and how did it compare with the streak maths?