Stops control the loss on a trade. They do nothing about the accumulation of losses across many trades, which is what a drawdown is. Portfolio-level controls act on the account as a whole: they reduce risk as losses mount and stop trading before the account reaches a point from which recovery is impractical.
The mechanisms
A reduction ladder cuts position size by a set fraction, often half, once the account has fallen a set percentage from its peak, and restores full size once it recovers. A halt stops new trades entirely beyond a deeper level, while existing positions run to their stops or targets. A daily loss limit flattens and pauses for the day when losses reach a threshold, which matters on funded accounts. A total open risk cap limits the combined risk of all open positions at any moment, so that a bad day cannot exceed a known figure.
The trade-offs
Every control that reduces losses also slows recovery. Halving size at a 10 per cent drawdown means the trades that recover the account are half the size of the trades that lost it, and the time under water lengthens. A control that triggers and releases quickly whipsaws, cutting size just before the good run. A halt set too shallow stops a working strategy during its ordinary drawdown. Thresholds have to be set against the strategy's own drawdown distribution, deep enough to be rare and shallow enough to matter.
Why it is worth it
The arithmetic of recovery is the reason. A 20 per cent drawdown needs 25 per cent to recover; 40 needs 67; 60 needs 150. Controls that hold the drawdown at 20 rather than 40 do more for the account's long-run result than any improvement to the entries, because the loss they prevent is the loss that compounds.
A strategy at full size would have had a 34 per cent maximum drawdown over four years. With sizing halved beyond 10 per cent from peak and new trades halted beyond 20, the same trades produce a 17 per cent maximum drawdown, a recovery that takes seven months rather than four, and a total return about 15 per cent lower over the period. The owner of the first version left in month two of the drawdown at 28 per cent down. The owner of the second sat through 17 and was there for the recovery.
An EA that describes a reduction ladder, a halt and an open-risk cap, with their thresholds, is describing survivability, and it is the part of the design that decides whether its owner is still running it in three years. Ask for the thresholds, check them against the record's drawdowns, and expect a control that is used to have cost some return. An EA with no portfolio-level control is relying on the owner to be the control, and the pages on trading plans explain how that goes.