Is scalping futures profitable?

It depends almost entirely on whether the edge per trade survives the cost per trade. Commission, fees and slippage do not shrink with the stop, so the tighter the stop, the larger the share of each trade's risk they take. The only way to know for a given strategy is to test it with costs charged per fill, at the stop it will actually use.

Which futures market is cheapest to scalp?

The one whose tick is small relative to its typical movement, because that keeps each round trip small relative to the risk. In the screening engine's terms, a round trip at the menu's 0.8-ATR stop on 30-minute bars costs 0.012 R on NQ, 0.053 R on ES and 0.084 R on natural gas, and every one of those figures multiplies as the stop shrinks.

How much does slippage matter for scalping?

More than for any other style, because slippage is charged in ticks while a scalper's stop is only a few ticks wide. One tick of slippage on a four-tick stop is a quarter of the trade's risk.

Does order-flow delta predict the next bar?

Not per bar. On the research record, per-bar delta and the bar's own return carry nearly the same information from one-minute to thirty-minute bars, so the next bar conditioned on delta is the next bar conditioned on the last bar's return. Cumulative delta measured against price, as a divergence over a window, is where the information survives.

Can I backtest a scalping strategy on range or Renko bars?

Not reliably. A bar that closes after a fixed price distance cannot contain an adverse move bigger than itself, so stops placed outside the brick fill at prices the market may never have offered. On the research record, range bars added tens of thousands of dollars a year to a breakout backtest, all of it artefact.

Strategy style

Scalping futures: what survives the costs

A scalping strategy takes many small trades a day, holding each for seconds or minutes and aiming for a few ticks. Its edge per trade is small by design, which makes it the style most exposed to the one thing a backtest most often gets wrong: what each trade costs.

Why costs decide a scalping strategy

Every round trip pays commission, exchange fees and slippage, and none of them shrink when the stop does. Measured as a share of the risk on each trade, cost rises in proportion as the stop tightens: halve the stop and the same round trip takes twice the bite. A strategy that risks two ticks and gives up one tick of slippage has spent half its risk before the market moves.

The screening engine charges costs in exactly these terms, in R, per market. At the menu's 0.8-ATR stop on 30-minute bars a round trip costs 0.012 R on NQ and 0.053 R on ES. A scalper's stop is a small fraction of that stop, and the table below shows what the same round trip costs as the stop shrinks.

Round-trip cost in R on each market as the stop shrinks from the menu's 0.8-ATR stop
Stop, ATRStop	0.80.8 ATRthe menu's stop	0.4Half0.4 ATR	0.2A quarter0.2 ATR	0.08A tenth0.08 ATR
ES E-mini S&P 500	0.053 R	0.106 R	0.212 R	0.530 R
NQ E-mini Nasdaq-100	0.012 R	0.024 R	0.048 R	0.120 R
YM E-mini Dow	0.023 R	0.046 R	0.092 R	0.230 R
RTY E-mini Russell 2000	0.029 R	0.058 R	0.116 R	0.290 R
CL WTI Crude Oil	0.053 R	0.106 R	0.212 R	0.530 R
GC Gold	0.041 R	0.082 R	0.164 R	0.410 R
SI Silver	0.081 R	0.162 R	0.324 R	0.810 R
NG Henry Hub Natural Gas	0.084 R	0.168 R	0.336 R	0.840 R

The first column is what a round trip costs at the menu's 0.8-ATR stop on 30-minute bars. The others hold that dollar cost fixed while the stop shrinks, which is what happens to a strategy that tightens its stop and keeps its broker. A screen written in the Strategy Language pays the same round trip for every contract it enters, at whatever stop you write.

What the research record says about short horizons

Order flow is the scalper's favourite signal, and the bar-level version of it is mostly price restated. Per-bar delta and the bar's own return are close to the same variable on every timeframe from one minute to thirty; once the bar's return is controlled for, the information in per-bar delta collapses toward zero. What survives is cumulative flow measured against price, a divergence over a window, not the figure printed under each bar.

Timing is expensive at this scale. On one deployed stream, seven to nine percent of gross profit sat in the first minute after the signal, so a strategy that looks robust to its entry timing may simply never have been tested at the resolution where it breaks.

The charts scalpers favour can invent results. A range or Renko bar closes after a fixed price distance, so it cannot show an adverse move larger than itself, and a backtest on those bars fills stops at prices the market may never have offered.

Testing a scalping idea

The screener tests scalping ideas on bars down to one minute. Every rule is decided at a bar's close and filled at the next minute's open. A stop and a target in points or ticks rest as orders checked minute by minute, and when one minute reaches both, the stop is taken to have come first. A round trip is charged for every contract entered, so the first question the screen answers is whether the edge per trade clears the cost at the stop you intend to use. Holds shorter than a minute and tick-by-tick fills are outside it.

From the research record

Scalping questions

Is scalping futures profitable?: It depends almost entirely on whether the edge per trade survives the cost per trade. Commission, fees and slippage do not shrink with the stop, so the tighter the stop, the larger the share of each trade's risk they take. The only way to know for a given strategy is to test it with costs charged per fill, at the stop it will actually use.
Which futures market is cheapest to scalp?: The one whose tick is small relative to its typical movement, because that keeps each round trip small relative to the risk. In the screening engine's terms, a round trip at the menu's 0.8-ATR stop on 30-minute bars costs 0.012 R on NQ, 0.053 R on ES and 0.084 R on natural gas, and every one of those figures multiplies as the stop shrinks.
How much does slippage matter for scalping?: More than for any other style, because slippage is charged in ticks while a scalper's stop is only a few ticks wide. One tick of slippage on a four-tick stop is a quarter of the trade's risk.
Does order-flow delta predict the next bar?: Not per bar. On the research record, per-bar delta and the bar's own return carry nearly the same information from one-minute to thirty-minute bars, so the next bar conditioned on delta is the next bar conditioned on the last bar's return. Cumulative delta measured against price, as a divergence over a window, is where the information survives.
Can I backtest a scalping strategy on range or Renko bars?: Not reliably. A bar that closes after a fixed price distance cannot contain an adverse move bigger than itself, so stops placed outside the brick fill at prices the market may never have offered. On the research record, range bars added tens of thousands of dollars a year to a breakout backtest, all of it artefact.

Screen a scalping idea

Describe it in a sentence. The desk turns it into an exact specification and screens it through seven gates written down before the run, from real per-fill costs to a drift null and an era split. The verdict usually comes back in 15 to 25 seconds.

Screen an ideaThe first is free.