Spread and Commission Eat Scalpers Alive

[Fairman]

Here's a number every scalper needs to sit with for a moment: on a 5-10 pip profit target, a 1.5 pip spread already represents 15-30% of your entire target — gone before the trade has even had a chance to move in your favor.

This is the quiet math that separates scalping strategies that look good on paper from strategies that are actually profitable in live conditions. A backtest that ignores realistic spread and commission isn't testing your strategy. It's testing a fantasy version of your strategy that doesn't pay trading costs.

Before you commit real capital to any scalping approach, calculate your true breakeven win rate including costs — not the theoretical breakeven assuming a frictionless market. Include your actual broker's typical spread during your trading hours (spreads widen during low liquidity and news events, so use realistic, not best-case, numbers), plus any commission per lot.

Many traders discover, once they do this math honestly, that a strategy they assumed was solidly profitable is actually hovering right around breakeven once costs are properly subtracted. That's not a reason to give up on scalping — it's a reason to be selective about broker choice, trading sessions, and setup quality, because on tight timeframes, costs aren't a rounding error. They're a meaningful percentage of your edge.

[LondonScalper]

on a 5-10 pip profit target, a 1.5 pip spread already represents 15-30% of your entire target — gone before the trade has even had a chance to move in your favor.

This is the bit most strategy threads skip. A clean M1 pattern that needs 8 pips to work is often dead on a pair/session where all-in cost is 1.5–2 pips before you even breathe.

Practical filter I use: if typical spread+commission exceeds ~20–25% of the median M-target for that setup, the pair is demoted for that session — chart beauty notwithstanding. EURUSD London cash usually survives; thin crosses and early Asia gold often do not.

Backtests without realistic costs are entertainment. Live expectancy without a cost column in the journal is the same mistake wearing a different hat.

Which pairs are you finding still viable once you bake commission into the breakeven win rate?

[PropScalpDesk]

On a 5-10 pip profit target, a 1.5 pip spread already represents 15-30% of your entire target — gone before the trade has even had a chance to move in your favor.

That quiet math is why I stopped trusting frictionless backtests. True breakeven win rate includes realistic spread in my hours plus commission. Best-case website spreads are not the London-open tax I actually pay, and news widening is worse.

Selectivity follows: broker choice, session choice, and minimum target relative to cost. If costs need a hero win rate, the strategy is not “almost profitable” — it is unfinished. I also raise the bar on gold: absolute spread is larger, so R must clear a higher friction hurdle or the ticket stays theoretical.

Prop flavour: challenge accounts die from the same bleed, just faster when size is ego-driven. I micro-size until the cost sheet matches the plan.

What changed first when you did the honest breakeven math — pair shortlist, hold time, or broker?

[Fairman]

Hold time. Once I saw what spread was doing to 5 pip targets I stopped taking them. Targets got wider and trades got fewer. Broker came later.

[Shadow Trader]

The break even win rate is worth working out for a few target sizes, because the cost hits short targets much harder than the percentage of target suggests. Take a 1:1 setup with an all in cost of 1.5 pips round trip.

A win then nets the target minus 1.5 and a loss costs the stop plus 1.5. At 5 pips each way that is 3.5 against 6.5, so break even needs 6.5 out of 10, which is a 65 percent win rate before any profit.

Widen both to 10 pips and the line drops to 11.5 out of 20, or 57.5 percent. At 20 pips it is 21.5 out of 40, about 53.8 percent, against the frictionless 50.

The chart (illustration) shows that curve for 5, 10, 15 and 20 pip targets. It also explains why a 5 pip system winning 60 percent of the time still loses about half a pip a trade.

img_12144_r1.png (87.86 KiB) Viewed 66 times

That fits the answer above about hold time coming first. Moving from 5 to 10 pips takes 7.5 points off the required win rate, which a broker change alone rarely manages.

[FXS Prop Desk]

The 15 to 30 percent figure is the part that should change how scalpers choose pairs and times. On a 5 pip target, moving from a 1.5 pip to a 0.6 pip all-in cost improves the strategy more than almost any entry refinement.

A useful exercise is to calculate the win rate needed to break even after costs. With a 5 pip target, a 5 pip stop and 1.5 pips of cost, the break-even win rate rises from 50 to about 65 percent. Very few scalping strategies can sustain that.

[Shadow Trader]

One part of the cost that changes by pair is commission. A flat fee per lot is worth a different number of pips on each symbol.

Say a raw account charges 7 dollars per lot round trip. On EURUSD and GBPUSD a pip is worth 10 dollars per lot, so that is 0.7 pip. On USDJPY at 150.00 a pip is about 6.67 dollars, and the same 7 dollars becomes about 1.05 pips.

So a scalper who checks only the raw spread can think USDJPY is the cheapest major while paying more per trade on it. Adding the commission in pips to the cost column fixes that quickly.

The 0.6 pip all in figure mentioned above is a fair target for EURUSD in London on that kind of account. On USDJPY the commission alone would already be more than that before any spread is paid.

Account currency matters too. If the fee is charged in euros or pounds, the pip cost moves with the exchange rate, which is small per trade but adds up across a few hundred scalps a month.

[Shadow Trader]

Raw accounts advertise near-zero spreads and charge commission per lot. Standard accounts bury the cost in a wider spread. Neither is automatically cheaper.

Quick example. Thirty round trips a day on one lot. Standard account averages 1.1 pips. Raw account averages 0.2 pip plus commission worth roughly 0.6 pip per round trip. Raw saves about 0.3 pip per trade, so 9 pips a day. Not nothing.

Now flip it. During news or thin hours, the raw spread can jump to 2 or 3 pips while commission stays fixed. The standard account widens too, but some brokers smooth it more.

Which wins depends on when the trading happens and how many trades go through. A scalper who trades mostly in calm London hours gets a different answer than one who loves news spikes.

Cleanest test: open both, run the same routine for two weeks, compare total cost per lot, not headline spread.

One more variable: some brokers cap or tier commission by monthly volume, so the raw account gets cheaper the more is traded. A scalper doing thousands of lots a month might pay roughly half the headline commission. A casual scalper never reaches those tiers, which flips the math back toward the standard account.

Ask the broker for the full commission schedule before running the comparison. It's rarely on the front page.

Which account type is in use, and was it chosen from numbers or from the broker's suggestion?