Drawdown: the four things the word means

Maximum peak-to-trough drawdown

The largest fall from any equity high to any subsequent low, over the whole history. This is the figure a journal reports by default, and it is order-dependent: the same trades in a different sequence produce a different maximum.

It is a single number describing the worst stretch that happened. It says nothing about how long the account spent in it, which is why time-in-drawdown is reported beside it.

Closed equity versus including open positions

A curve built from closed trades only moves when a position is closed. A curve including unrealised P&L moves continuously, and it can register a drawdown that the closed-equity curve never shows — a position held through a large adverse move and then closed at break-even is invisible in one and severe in the other.

Neither is wrong; they answer different questions. What causes trouble is comparing a closed-equity figure from a journal against an intraday figure from a broker or a prop platform and expecting them to agree.

Daily loss, which is not a drawdown at all

A daily loss limit measures one day against that day's own starting balance, and it resets. It is a floor, not a fall from a peak, and a book can breach a daily limit on a day when the account is at an all-time high.

Mixing the two is a common source of confusion when a rule set uses both — which most prop firm rule sets do.

Trailing drawdown, and why it is the hardest to reproduce

A trailing drawdown moves its floor up as the account makes new highs, and often does not move back down. Where the high-water mark tracks the intraday unrealised peak rather than the closed balance, the floor can be raised by a level the account touched for a second and never held.

That version cannot be reconstructed from trade rows at all: the rows record fills, and the peak happened between them. Any figure derived from a journal is therefore an estimate of it, and should say so.

Recovering from one is not symmetrical

The gain needed to return to the previous peak is larger than the fall that caused it, because it is measured against a smaller balance. A 20% fall needs a 25% gain; a 50% fall needs 100%. This is arithmetic rather than a claim about markets.