ES emini scalping strategy: does it survive costs in a backtest?

Contents

1. Does an emini scalping strategy survive commissions and slippage?
2. What does scalping mean on ES emini futures?
3. How many ticks per trade do you need after commission and slippage?
4. How much capital does scalping ES need?
5. Which session is best for an ES scalping test?
6. What short-bar entry rules can you test?
7. How many trades and which checks validate a scalping backtest?
8. Frequently asked questions
9. References

Does an emini scalping strategy survive commissions and slippage?

An emini scalping strategy survives a backtest only when its gross edge per trade is several times the round-trip cost of commission and slippage, measured in ticks. On ES, one tick of slippage per side already costs two ticks, so a rule that aims for three ticks has almost no room. Put costs in the first run, not at the end.

Most ES scalping rules that look good without costs fail when costs are added, because the move they capture is about the same size as the cost of capturing it. Slippage is the difference between the price the backtest assumed and the price a real order would fill at. For a swing trade that aims at 100 ticks, two ticks of slippage is a rounding error. For a scalp that aims at three ticks, it is most of the trade. That is the one claim this article defends: for scalping, the cost line decides the result, and the rule itself comes second.

The order of work follows from that. First convert your commission and slippage into ES ticks and write down the cost floor. Second, run the backtest with those costs charged on every trade from the first candidate, so the ranking already reflects what you could trade. Third, read the average gross trade against the floor before you look at anything else. AlphaPrime (Valynox) applies the commission and slippage you set inside every candidate it runs in backtesting, with next-bar execution, so a scalp that lives on two ticks falls out of the ranking in the first pass rather than surviving to the final review.

What does scalping mean on ES emini futures?

Scalping on ES is holding a position for a few ticks over a few minutes on one-minute or smaller bars, which makes the cost per trade the largest single line in the result compared with day or swing trading. Scalping is a trading style that takes many small trades per day, each aimed at a move of roughly two to six ticks, and closes every position within minutes. On ES, the E-mini S&P 500 future, one tick is 0.25 index points and is worth $12.50 per contract, so a four-tick scalp is a $50 move before costs. The free futures contract calculator converts ticks, points and dollars for ES and the other contracts you test.

Compare this with the other two common holding periods. A day trade on 5 to 15-minute bars aims at perhaps 10 to 40 ticks and takes one to three trades a day. A swing trade on daily bars aims at 100 ticks or more and may take fewer than one trade a day. A fixed cost of, say, two ticks per round trip is a small share of a 100-tick target, a noticeable share of a 20-tick target, and half or more of a 4-tick target. The same commission and the same slippage, in dollars, weigh three completely different amounts.

Two more facts about scalping change how you test it. The trade count is high, often ten or more trades per day, so a few years of one-minute data produce thousands of trades. And the data requirement is strict: one-minute bars with the correct session template and clean rolls between contract months, as covered in the guide to contract specs, rollover and tick value in futures backtesting. A bad tick that would not matter on daily bars can create or destroy a four-tick scalp.

How many ticks per trade do you need after commission and slippage?

Convert commission and one tick of slippage per side into ES ticks to get the cost floor, then require the average gross trade to be several times that floor before the rule is worth more testing. The cost floor is the total cost of one round trip (entry plus exit) expressed in the same unit as the backtest result. On ES, one tick of slippage on the entry and one on the exit is already two ticks, or $25.00 per round trip, before any commission.

The calculation is one line:

net per trade = gross ticks × tick value − commission and fees − slippage

Here gross ticks is the average move the rule captured before costs, tick value is $12.50 on ES, commission and fees is the broker commission plus exchange, clearing and regulatory fees for the round trip, and slippage is the number of ticks lost per side times two sides times the tick value.

A useful rule of thumb from the example: the gross average trade should be several times the cost floor, not just above it. With a $29.00 floor on ES, a gross of about 7 ticks ($87.50) is three times the floor and still clears zero when slippage doubles. A gross of 3 ticks does not. The free expectancy calculator gives the same per-trade figure from win rate and average win and loss, if your report shows those instead of gross ticks.

How much capital does scalping ES need?

Capital for ES scalping is margin plus a drawdown buffer taken from the backtest plus a cushion for costs, and the Micro E-mini lets the same rule be tested with a tenth of the dollar exposure. Intraday margin is the deposit your broker requires per contract while a position is open during the session; brokers set it, they change it, and it is usually lower than the exchange's overnight margin, so check your broker's current figure rather than a number from an article. The guide to futures margin by broker explains how intraday and overnight margin differ and how they feed into backtest size.

1. Margin. The broker's intraday margin for one ES contract, taken from the broker's page on the day you check.
2. Drawdown buffer. The maximum drawdown in the backtest, after costs, multiplied by a safety factor of about 1.5 to 2, because the live drawdown is usually deeper than the one backtest you saw.
3. Cost cushion. The extra cost of one month of trades if slippage runs one tick worse than you modeled. For a rule that takes 20 trades a day, that is 20 × 21 days × $25.00 on ES, which is a real line in the account.

The Micro E-mini S&P 500 (MES) is one tenth of ES: $5 per index point and $1.25 per tick, per the exchange's Micro E-mini FAQ. That makes it the right contract for a first live test of a scalping rule, because the drawdown buffer and the cost cushion shrink by ten. One thing does not shrink by ten: the commission per contract. Brokers charge less per micro contract than per E-mini, but rarely one tenth, so the commission share of each MES tick is larger than on ES, and a rule that just clears the cost floor on ES can fall below it on MES. Run the MES backtest with the MES commission, not the ES commission divided by ten. The exchange can change contract specifications; the contract page on CME Group's website carries the current values.

Which session is best for an ES scalping test?

The US regular session, especially its first hours, carries the volume a scalp needs for one-tick slippage to be realistic, so test the regular session and the overnight session as separate backtests. ES trades almost 23 hours a day on CME Globex, from 18:00 US Eastern time to 17:00 the next day, with a maintenance halt from 16:15 to 16:30 Eastern. The US stock market's regular hours, 09:30 to 16:00 Eastern, are the part of that day when ES volume is heaviest, and volume is what lets a market order fill one tick from the trigger price instead of two or three. The free futures market hours tool shows these sessions in your own time zone, and the exchange's contract page is the source if the schedule changes.

For a reader in Taiwan the regular session runs from 21:30 to 04:00 the next morning while the US is on daylight saving time, and from 22:30 to 05:00 the rest of the year. The Globex open at 18:00 Eastern is 06:00 or 07:00 Taiwan time. The overnight hours between the US close and the next US open are the quiet part of the day in ES, where the one-tick slippage assumption is weakest.

Three separate backtests tell you more than one combined run:

1. Regular session only, 09:30 to 16:00 Eastern. This is the base case. One tick of slippage per side is a fair starting assumption here.
2. Opening window only, 09:30 to about 11:30 Eastern. Many scalping rules make most of their gross result here. If the full-session result is positive but the opening-window result is flat, the rule is earning its ticks in a thinner part of the day, where your slippage number is probably too kind.
3. Overnight session only, 18:00 to 09:30 Eastern. Use at least two ticks of slippage per side. A rule that survives here with doubled slippage has a real cushion.

Setting the session template correctly is one of the five settings for ES and NQ futures backtesting to fix before the first run, because a one-minute rule tested on bars from the wrong session is testing a different market.

What short-bar entry rules can you test?

Breakout, pullback and mean-reversion rules on short bars are testable when each has one fixed exit, and the cost floor removes the ones that depend on capturing one or two ticks. A testable scalping rule has an entry trigger you can state in one sentence and an exit that does not depend on judgment: a fixed number of ticks, a fixed number of bars, or a fixed time of day. Rules with two or three exits, each tuned, give the search too many ways to fit the noise in one-minute data.

	Breakout	Pullback	Mean reversion
Entry trigger	Close above the last N-bar high (or below the low)	Price touches a short moving average while the longer trend points the same way	Close outside a volatility band, trade back toward the middle
Exit	Fixed ticks or a fixed number of bars	Fixed ticks or a fixed number of bars	Return to the band's middle or a fixed number of bars
Ticks typically aimed for	4 to 8	3 to 6	1 to 3
Order type at entry	Stop order, fills worse than trigger	Limit order, may not fill	Limit order, may not fill
Removed first by the cost floor	Third	Second	First

Mean reversion on one-minute bars is removed first because its natural target, the middle of the band, is often one or two ticks away, below the floor from the earlier section. Breakouts aim higher but enter on stop orders, which fill worse than the trigger, so their slippage assumption should be the strictest. Pullback entries with limit orders have the opposite problem: a backtest that fills every limit order when price touches it is optimistic, so count how many fills the rule assumes at the exact low of a bar.

If this is your first systematic rule, build and test one of these three on the regular session, with one exit, before you add any filter.

How many trades and which checks validate a scalping backtest?

A scalping rule produces thousands of trades, which makes averages reliable but multiplies any cost error, so the validation checks are doubled slippage, out-of-sample trades and remembering that the one candidate kept from many tried is flattered by selection. The uncertainty of an average trade shrinks with the square root of the trade count, so an average built from three thousand scalps barely moves if a few trades change. That is the good news. The bad news is the same multiplication applied to a mistake.

To show how trade count changes what a report can tell you, here is a real run from the AlphaPrime product engine on a very different bar size. It tested NQ daily bars, from 2019-01-02 to 2025-06-30, long and short, one contract, with $20 commission and $10 slippage charged on every trade. The rule is a stand-in built from standard indicators as its only entry and exit, with no stop, target or filter; it is not a scalping rule and it is not the subject of this article. The result is the one candidate the search kept out of 1,024 it tried, chosen for its return against its drawdown.

Read the box as a lesson in trade count and selection, not as a scalping result. With 105 trades, the average trade of $1,927.38 could move a long way if ten trades were different, and the max drawdown of $80,555 is one path out of many. On NQ, one tick is $5.00, so $10 of slippage is two ticks against a daily-bar rule that holds for days; the same $10 against a one-minute scalp would be most of the trade. And because this candidate was picked after the fact from 1,024 for its return against its drawdown, the numbers flatter it. Bailey, Borwein, López de Prado and Zhu (2014) show that the best backtest in a large search is almost always inflated, and Bailey and López de Prado (2014) propose discounting a result for the number of candidates behind it. Neither the rule nor the indicators behind it earned that money, and no reader should expect it.

For a scalping rule on one-minute bars, the checks to write down before the run are these:

1. Doubled slippage. Rerun with two ticks per side. A rule that turns negative was renting its edge from cheap fills.
2. Out-of-sample trades. Out-of-sample data is the part of the history the search never touched. Count the completed trades in it; a scalp should give several hundred, and the per-trade figure after costs should keep its sign. The guide to how to split in-sample and out-of-sample data explains how much is enough.
3. Top trades removed. Delete the ten largest winners. A scalp's result should barely change, because no single trade should matter.
4. Selection count. Note how many candidates were tried before this one was kept. The larger that number, the larger the discount the result deserves.

The next step is one number: take your broker's ES commission and fees, add $25.00 for one tick of slippage per side, and divide by $12.50 to get your cost floor in ticks. Then open your last scalping backtest and read the average gross trade in ticks next to it. If it is not several times the floor, rerun with doubled slippage before you spend another hour on the rule.

Frequently asked questions

How much capital do you need to scalp ES?

Add three parts: your broker's intraday margin for one ES contract, the maximum drawdown of your backtest after costs multiplied by about 1.5 to 2, and the extra cost of one month of trades if slippage runs one tick worse than you modeled. Brokers set and change intraday margin, so take that figure from your broker's page. The Micro E-mini (MES) is one tenth of ES and lets you run the same rule with about one tenth of the drawdown buffer.

How many ticks per trade does an ES scalp need to be worth it?

First compute the cost floor: commission and fees for the round trip plus one tick of slippage per side, which on ES is already $25.00, divided by $12.50 to express it in ticks. Then require the average gross trade to be several times that floor, not just above it. A gross of three ticks against a floor of about 2.3 ticks turns negative as soon as slippage doubles; a gross of seven ticks still clears zero.

Which time of day is best for ES scalping?

The US regular session, 09:30 to 16:00 Eastern time, has the heaviest ES volume, and the first two hours after the open are where most scalping rules earn their gross result. Test the regular session, the opening window and the overnight session as three separate backtests, with stricter slippage for the overnight hours. In Taiwan the regular session starts at 21:30 or 22:30 depending on US daylight saving time.

How much do costs change an ES scalping backtest?

For a scalp aiming at three or four ticks, commission and slippage usually take half or more of the gross result, and a one-tick increase in slippage per side can turn a positive backtest negative. The same costs take a few percent from a swing trade aiming at 100 ticks. That is why costs belong in the first run of a scalping backtest and why the doubled-slippage rerun is the most important validation check.

References

• Bailey, D. H., Borwein, J., López de Prado, M., & Zhu, Q. J. (2014). Pseudo-Mathematics and Financial Charlatanism: The Effects of Backtest Overfitting on Out-of-Sample Performance. Notices of the American Mathematical Society, 61(5), 458–471.
• Bailey, D. H., & López de Prado, M. (2014). The Deflated Sharpe Ratio: Correcting for Selection Bias, Backtest Overfitting and Non-Normality. The Journal of Portfolio Management, 40(5), 94–107.
• CME Group. E-mini Nasdaq-100 Futures Contract Specs. https://www.cmegroup.com/markets/equities/nasdaq/e-mini-nasdaq-100.contractSpecs.html
• CME Group. FAQ: Micro E-mini Equity Index Futures. https://www.cmegroup.com/articles/faqs/frequently-asked-questions-micro-e-mini-equity-index-futures.html