Daily loss limit: set it from the average daily range and backtest it

Contents

1. What does a daily loss limit do, and does it help a systematic strategy?
2. What percentage is a reasonable daily loss limit?
3. What does a daily loss limit change in drawdown and tail days?
4. What does stopping early cost you?
5. How does position size interact with the daily loss limit?
6. What should you do the day after the daily loss limit triggers?
7. How do prop firm daily loss rules differ from a research limit?
8. Frequently asked questions
9. References

What does a daily loss limit do, and does it help a systematic strategy?

A daily loss limit stops trading for the rest of the session once the day's losses reach a set amount. Size it from the market's average daily range, point value and contract count, so it sits above a normal losing day and fires only on tail days. Then backtest it, because it also removes the trades after the trigger.

A daily loss limit is a rule that cancels all orders and takes no new trades for the rest of the session once the day's closed and open losses reach a set dollar amount. It works at the level of one session. A per-trade stop loss works at the level of one position: it is a price distance from the entry, and how a stop loss order works on futures explains how it fills. An account drawdown limit works at the level of the whole account: it measures the fall from the highest equity and pauses trading until you review the strategy. The three rules answer three different questions, so one does not replace another.

	Per-trade stop	Daily loss limit	Drawdown limit
What is measured	One trade	One session	Whole account
When it triggers	Price distance from entry	Dollars lost today	Fall from peak equity
What it stops	One position	New entries for the day	Trading until review
When it resets	Next trade	Next session	After review

Whether the limit helps a systematic strategy is a trade-off, not a default to switch on. It removes the worst part of the worst days. It also removes every trade the strategy would have taken after the trigger, and some of those trades would have won. Kaminski and Lo (2014) make this point for stop-loss rules in general: a stop adds value when losses tend to continue after the trigger and subtracts value when they tend to reverse. The same test decides a daily limit, and it is the same test that picks between fixed, ATR, time or trailing stops for a futures strategy. The sections below show how to set the amount and how to measure both sides of the trade-off.

What percentage is a reasonable daily loss limit?

A reasonable daily loss limit is set in dollars from the market's average daily range, the point value and the contract count, and then checked against two bounds: above the strategy's typical losing day, and below the share of the account you accept losing in one day. The percentage comes out at the end; it is not the input.

The average daily range is the average of each session's high minus its low over a recent stretch of sessions, measured in points. It tells you how far the market normally travels in a day, so it tells you how much one contract can normally gain or lose in a day. To turn points into dollars you need the point value. For the E-mini Nasdaq-100 future (NQ) the contract is $20 times the index, so one index point is $20 and the minimum tick of 0.25 points is $5 (CME Group contract specifications). Exchanges can change these values, so confirm them on the exchange's contract page before you set a limit; the free futures contract calculator shows tick size, tick value and point value for the contract you trade, and the NQ and MNQ specs, sessions and backtest costs article covers the rest of the setup for that market.

1. Measure the average daily range over the last 20 to 60 sessions of the session you actually trade (day session only, or full session).
2. Convert to dollars per contract: range in points × point value.
3. Decide how many losing trades the day may hold. Look at the strategy's typical loss per trade as a share of the daily range. Two or three typical losses is a common choice, so the limit does not fire on an ordinary losing day.
4. Multiply by the contract count and check the account share. The one formula in this article: limit in dollars = points allowed for the day × point value × contracts, where points allowed for the day is the sum of the losing trades you accept before stopping. Divide the result by the account size to read the percentage.

Where the limit lands	What happens	What to do
Below the typical losing day	Fires several times a month; the strategy is cut on ordinary days	Raise it, or reduce size
Two to three typical losses, a few percent of equity	Fires on tail days only	Test it as described below
Above the largest losing day in the backtest	Never fires in the test; it protects only against the unseen	Keep it as a last safety line, not as a tested rule

Why a few percent of equity is the upper check: a day that costs 10 percent of the account means five such days in a bad month remove half of it. A limit that only fires on tail days and still leaves most of the account intact is one you can live through and keep trading at the planned size.

What does a daily loss limit change in drawdown and tail days?

A daily loss limit shortens the left tail of the daily results but barely moves the long drawdown, because a drawdown is built from many losing days in a row, not from one bad day. A tail day is a losing day far larger than the strategy's typical losing day. Picture the daily results as a pile of coins sorted by size: the limit caps the few largest losing coins, but the drawdown is the total of a long row of ordinary losing coins, and that row is untouched. Magdon-Ismail and Atiya (2004) show that the expected maximum drawdown grows with the length of the period, which is another way of saying that drawdowns are accumulated over time rather than caused by a single day.

The run below gives the baseline such a limit would be measured against. AlphaPrime (Valynox) tested NQ daily bars from 2019-01-02 to 2025-06-30, long and short, one contract, with $20 commission and $10 slippage per trade. The rule is a Tail Heaviness Index rule used as the only entry and exit, with no stop, no target and no filters. It is a stand-in built from standard indicators, not a daily loss limit itself. A daily loss limit is an intraday rule, and daily bars do not show the path inside the day, so this run cannot be rerun with the limit applied; it only supplies the baseline net profit and drawdown. The candidate shown is the one the search kept out of 1,024 it tried, kept for its return against its drawdown, and choosing the kept one after the fact flatters it. These numbers describe what the rule did on this history; they say nothing about what it or a reader will earn.

Read the box this way. The max drawdown of $43,730 is many times the average trade of $1,277.61, so the drawdown was a run of losing trades, not one tail trade. A daily limit that capped the single worst day would leave most of that run in place. What the limit does change is the shape of the left tail: the days beyond the line are collapsed onto the line, while the centre of the distribution, which is where the drawdown lives, stays as it was.

To measure the effect on your own intraday strategy, add up each day's closed trades from the trade list, sort the days, and count how many fall beyond the limit you set in the previous section. The sum of the excess beyond the line is the most the limit could have saved; compare it with the max drawdown. When the saving is a small fraction of the drawdown, the limit is a tail guard and not a drawdown cure. The trade list in AlphaPrime's Performance Report carries the date of every trade, which is the raw material for sorting a backtest into days before you decide where a daily line should sit.

What does stopping early cost you?

Stopping early costs you every trade the strategy would have taken after the trigger, and the MAE and MFE of those trades tell you whether the strategy usually recovered or kept losing from that point. MAE (maximum adverse excursion) is the largest loss a trade showed while it was open; MFE (maximum favorable excursion) is the largest gain it showed while open. How MAE and MFE set stops and targets from excursion explains both measures in full, and how to set profit targets from maximum favorable excursion covers the MFE side.

The cost is measured with this sequence:

1. Find the trigger days. In the backtest trade list, find each day where the running total of closed losses crossed the limit, and note the time of the crossing.
2. Collect the remaining trades. List every trade the strategy opened after that time on that day. These are the trades the limit throws away.
3. Sum their net result. A positive sum is the profit the limit costs; a negative sum is the loss it saves.
4. Read their MAE and MFE. Large MFE with small MAE on the remaining trades means the day usually recovered after the trigger; large MAE with small MFE means the losses usually continued.
5. Compare with the saving from the previous section. The limit helps only when the loss it saves on the trigger days is larger than the profit it removes on the same days.

Remaining trades on trigger days	What it means	Decision
Net negative, large MAE, small MFE	Losses tended to continue after the trigger	Keep the limit at this level
Net near zero	Recovery and further loss roughly cancelled	Keep it only as a safety line; it changes little
Net positive, large MFE	The strategy usually recovered	Raise the limit or remove it

How does position size interact with the daily loss limit?

Set the position size first and convert the daily limit from that size, never the other way round, because the contract count moves the drawdown far more than any daily line does. The second run below uses the same trades as the first box: the entries and exits do not change, only the number of contracts each method assigns, so every difference between rows comes from sizing alone. Each method uses its report defaults, with capital of $100,000 and at most 10 contracts.

Kelly sizing averaged 1.01 contracts and its max drawdown of $43,730 is the same as the single-contract run. Fixed fractional risk averaged 5.54 contracts and reached a max drawdown of $331,520, measured in dollars at the much larger equity that method had built by then. The trades are the same list; the difference is the size. Full Kelly, half Kelly or fixed risk on futures explains why most traders size well below the Kelly figure.

For the daily limit this means one thing: a dollar limit chosen first and then traded at a different size is either far too tight or far too loose. Fix the contract count from your account and the strategy's drawdown, then compute the limit as points allowed for the day × point value × that contract count. When the size changes, recompute the limit the same day.

What should you do the day after the daily loss limit triggers?

Resume the next session at the planned size with the rules unchanged, after a short check that the trigger came from the market and not from a data or execution fault, and change anything only when a pre-written trigger count is reached. A bad day that stays inside the backtest's range of bad days is the strategy working as tested; a change made because of one such day is tuning on a sample of one.

1. The limit triggers and all orders are cancelled. Flatten any open position if your rule counts open losses. Do not re-enter in the same session.
2. Check fills, data and connection. Compare the fills with the signals the strategy should have produced. A wrong bar, a frozen feed or a doubled order is a different problem from a bad market day and is fixed before the next session.
3. Log the day in the journal. Record the trigger time, the loss at the trigger, and the session's final result as if you had kept trading, so the cost of stopping early (previous section) builds up as live evidence.
4. Open the next session at the planned size with unchanged rules. Reducing size after one trigger means trading small on the days that follow the worst days, which the backtest never did.
5. Review only when the pre-written count is reached. For example: if the limit fires 3 times in 20 sessions, stop and compare the live trigger rate with the backtest's. Write the count down before you go live.

The review itself is a comparison of two rates. The backtest told you how often the limit fired per 100 sessions; the live log tells you the same number. A live rate far above the tested one means the market has changed, the size is wrong for the current range, or the execution is worse than the backtest assumed. Each has a different fix, and only the first one touches the strategy rules.

How do prop firm daily loss rules differ from a research limit?

A prop firm's daily loss rule is written by the firm, so a strategy traded under one must be backtested with the firm's exact definition and not with the limit you would choose from the daily range. A prop firm (proprietary trading firm) is a company that lets a trader trade the firm's capital under a set of risk rules, and breaking the daily rule usually ends the account rather than pausing it for a day.

The definitions differ in ways that change the backtest. Many firms count open profit and loss as well as closed trades, so a position that is deep under water at the wrong moment can breach the rule even if it later closes at a gain. Commissions are often included in the figure. The day resets at the firm's clock, often a fixed exchange time, which may not match the session your strategy uses. Some firms add a trailing account drawdown on top, which is the account-level rule from the first section with the firm's own numbers.

	Prop firm rule	Research limit
Who sets the amount	The firm	You, from the daily range and size
What counts	Closed plus open P&L, usually including commissions	Usually closed trades
Reset time	The firm's clock, often a fixed exchange time	The session you chose
Consequence	Account breach	Pause until the next session
Backtest setting	Copy the firm's rule exactly	Free to test several levels

In practice this means two backtests, not one. The first uses your own research limit and answers whether the limit helps the strategy. The second copies the firm's amount, the open-P&L counting, the commission treatment and the reset time, and answers a different question: how many times in the test period would this account have been breached. A strategy that passes the first test and fails the second is a strategy for your own account, not for that firm. Count the breaches in your backtest under the firm's exact definition before you pay for an evaluation, and if the count is above zero, the next test is a smaller contract count under the same rule.

Frequently asked questions

What percentage should a daily loss limit be?

Work it out in dollars first and read the percentage at the end. Multiply the market's average daily range by the point value and your contract count, decide how many typical losing trades the day may hold (two or three is common), and divide the result by the account. The limit should sit above the strategy's typical losing day, so it fires only on tail days, and below the share of the account you accept losing in one day, which for most traders is a few percent.

What should I do the day after the daily loss limit triggers?

Check that the trigger came from the market and not from a data or execution fault, log the day, and open the next session at the planned size with the rules unchanged. Change size or rules only when a count you wrote down in advance is reached, for example three triggers in twenty sessions. One bad day inside the backtest's range of bad days is the strategy behaving as tested.

Does a daily loss limit reduce profit?

It can. Every trigger removes the trades the strategy would have taken for the rest of the day, and on days where the strategy would have recovered those trades were winners. Measure it: list the trades after each trigger in the backtest, sum their results, and read their MAE and MFE. When the sum is positive the limit costs profit; when it is negative the limit saves losses. The sections on stopping early and on MAE and MFE above show the steps.

Should a daily loss limit count open positions?

For your own research limit, choose one definition and backtest with it; counting closed trades only is simpler and does not depend on the price path inside a bar. Under a prop firm rule you have no choice: most firms count open profit and loss as well as closed trades, so the backtest must do the same, or it will miss breaches that happen while a position is still open.

Does a daily loss limit reduce maximum drawdown?

Usually only a little. A maximum drawdown is built from a run of losing days, and the limit caps only the days that fall beyond its line. Sort the backtest's days by result, add up how much of the loss lies beyond the limit, and compare that sum with the max drawdown. If the sum is a small fraction, the limit is a tail guard and the drawdown must be controlled by position size instead.

References

• Kaminski, K. M., & Lo, A. W. (2014). When Do Stop-Loss Rules Stop Losses? Journal of Financial Markets, 18, 234–254.
• Magdon-Ismail, M., & Atiya, A. F. (2004). Maximum Drawdown. Risk, 17(10), 99–102.
• CME Group. E-mini Nasdaq-100 Futures contract specifications (contract size $20 × index, minimum tick 0.25 index points = $5.00). cmegroup.com/markets/equities/nasdaq/e-mini-nasdaq-100.contractSpecs.html