Daily Loss Limits and Drawdown Protection

Daily loss limits are the least glamorous risk control and one of the few that reliably works, because the failure it prevents is a human one.

What it actually protects against

Not a bad strategy. A strategy that loses money will lose it across sessions regardless of when each one stops.

What it protects against is the specific pattern where a losing session becomes a catastrophic one — increasing size to recover, taking setups outside the plan, overriding exits, or simply continuing to trade a strategy that is not working in the current regime. Those decisions get made by someone who is down, and being down is a poor state for judgment.

Automation does not remove this. It relocates it. The operator still decides whether to leave automation running, whether to intervene, and whether to adjust settings mid-session. A loss limit removes that discretion at the moment it is least reliable.

Sizing it against variance, not against a round number

The most common mistake is choosing a number that sounds disciplined.

Every strategy has a distribution of daily outcomes, and losing days inside that distribution are normal rather than evidence of a problem. If your strategy routinely has sessions down some amount without anything being wrong, a limit inside that range will halt you on ordinary days. You will stop trading a functioning strategy at the point it most often recovers, and you will do it repeatedly.

Too loose and it never engages, which means it is decoration.

The useful reference points are your strategy's typical losing day and its worst historically observed day. A limit meaningfully beyond the typical and inside the worst gives you a control that fires when something is genuinely unusual rather than when the day is merely bad. Where exactly it sits between those is a judgment about how much of an outlier you are willing to sit through.

One caution: the worst observed day in your data is not the worst possible day. It is the worst that happened in the sample you have. Sizing to it is not a ceiling.

Fixed dollar or percentage

A percentage of account scales as the account grows, which sounds appropriate and has a specific drawback: it also scales after losses, shrinking the limit exactly when you are drawing down. That produces a tightening feedback loop where each bad stretch reduces the room the next one has.

A fixed dollar amount is unresponsive to growth and predictable. You know today what stops trading tomorrow.

The middle path is a fixed amount reviewed on a schedule rather than continuously — recalculated monthly or quarterly against current capital, so it tracks the account without reacting to individual sessions.

Realised, unrealised, or both

A design decision that changes behaviour substantially and is often left implicit.

Realised only counts closed trades. Clean and countable, and it ignores an open position that is deeply underwater — you can be well past your limit in economic terms while the counter says you are fine.

Including unrealised counts open positions at current marks. More honest about actual exposure, and it means normal intraday fluctuation in an open position can trip the limit on a day that would have recovered.

Most systems count realised losses for the halt decision while monitoring total exposure separately. Whichever you choose, be explicit, because a system that is ambiguous about this behaves unpredictably at exactly the wrong moment.

What happens at the limit

Stopping new entries is the easy half. The important question is what happens to open positions, and there is no universally right answer.

Close everything is decisive and forces liquidation at a moment that may be the worst available price, since you are exiting because things went badly.

Stop new entries, manage existing positions to their exits respects the plan each position was opened under and leaves you exposed while already at your limit.

The second is usually more defensible, because the exits were chosen deliberately and the limit was about not opening new risk. But it requires that your existing exits are broker-resident, or you are relying on software that you have just told to stop trading.

Whatever you choose, decide it before the day it matters. A rule invented mid-drawdown is a decision made by someone who is down.

Implementation notes

Enforce it server-side, in the component that places orders, rather than in whatever generates signals. A limit that lives in a signal source can be bypassed by a second signal source, a manual order, or a bug. One enforcement point, applied to every order regardless of origin.

State must survive a restart. A system that reloads mid-session and recomputes from an empty ledger will happily resume trading past a limit it already hit. Persist the daily accumulator outside the process.

Define the reset boundary explicitly, including what happens across sessions, holidays, and time zones.

And make the halt loud. A system that silently stops trading is indistinguishable from a system that broke. On a self-hosted deployment, alerting is yours to configure, which means it happens only if you configure it.

The honest limits

A loss limit caps a session, not a loss. A gap or a halt can take you past it in a single move before any counter updates, and the limit only prevents further trading after the fact.

It does not improve a strategy. Stopping earlier on bad days changes the distribution of daily outcomes without changing whether the edge exists.

It can stop you on days that would have recovered. That is the cost of the control, not a defect in it.

And it is downstream of the control that actually bounds risk. Position sizing determines what any single trade costs and therefore how fast a limit can be reached — capital divided by twenty as the ceiling per position, under the divide-by-20 rule. A loss limit on top of oversized positions is a smoke alarm in a house made of matches.

Frequently asked questions

What is a daily loss limit? A session-level halt that stops trading once cumulative losses reach a defined amount.

How should I size it? Relative to your strategy's normal daily variance — beyond a typical losing day, inside the worst observed one — rather than to a round number.

Should it count unrealised losses? Either can work, but be explicit. Realised-only ignores an underwater open position; including unrealised can trip on normal intraday fluctuation.

What should happen to open positions at the limit? Usually stop new entries and let existing positions reach their planned exits, provided those exits rest at the broker rather than in the software you just halted.

Does a loss limit reduce risk? It caps session damage from continued trading. It does not cap a single adverse move, and it does not fix a strategy.

Disclaimer: This article is educational content about trading mechanics and software engineering. It is not investment advice, financial advice, tax advice, legal advice, or a recommendation to buy or sell any security, nor a recommendation of any strategy, order type, or configuration. Any instruments, settings, or figures named are used solely to illustrate mechanics. Options trading involves substantial risk of loss and is not suitable for all investors. Please read Characteristics and Risks of Standardized Options before trading options. Automated trading systems carry additional risks including software defects, connectivity failures, broker API changes, and outages that may prevent orders from being placed, modified, or cancelled. Stop orders do not guarantee an execution price and stop-limit orders may not execute at all. Past performance does not indicate future results, and no configuration, order type, position-sizing rule, or risk setting can guarantee a profit or prevent a loss.

StaxInvesting LLC sells self-hosted trading software. It is not a broker-dealer, investment adviser, or financial institution, and it does not manage accounts, hold member funds, place trades on behalf of members, or access member brokerage accounts. Members run the software in their own cloud environment, connect their own brokerage accounts under their own credentials, and are solely responsible for their configuration, their credential security, and every trade executed in their account. Third-party platform and broker details described here reflect publicly available information as of publication and are subject to change without notice; always verify against current official documentation. Consult a qualified financial adviser and tax professional regarding your individual circumstances.