Advertisement IC Markets

Mean reversion on intraday charts: what the costs do to it

Discuss 1-minute to 15-minute price action setups, fading intraday momentum, key support/resistance zones, and proven short-term trading methodologies.
Post Reply
FXS London Scalper
Posts: 978
Joined: Sun Oct 18, 2026 12:25 pm

Mean reversion on intraday charts: what the costs do to it

Post by FXS London Scalper »

Mean reversion strategies on short timeframes often show attractive results before costs and disappointing results after. Because profits per trade are small, spread and commission take a large share.

A worked example: a strategy averaging 4 pips gross per trade with 300 trades a year would earn 1,200 pips before costs. With a 1 pip spread and 0.6 pip commission equivalent, costs total 480 pips, cutting the net result by 40 percent. Slippage on entries can reduce it further.

A method to stress test is to double assumed costs and see whether the strategy still holds. If it breaks, the edge may be too thin for real trading.

Timing the trades for the most liquid hours, when spreads are tightest, can help. Some traders restrict mean reversion to the London-New York overlap.

It is also worth checking whether the strategy depends on fills at exact levels, which may not happen in practice.

How do you account for costs in intraday tests?
FXS Official Team | London Session Desk
Recommended broker for automated trading & scalping IC Markets
Amanhasnoname
Posts: 2060
Joined: Fri Oct 30, 2026 9:45 am

Re: Mean reversion on intraday charts: what the costs do to it

Post by Amanhasnoname »

My first serious attempt at intraday mean reversion was a Bollinger fade on EURGBP five-minute charts, and the backtest looked lovely until I put realistic costs in. Gross, it made about three pips a trade. My broker at the time charged a spread that averaged well over a pip on that cross outside London hours, plus commission, so the net was close to zero and the drawdowns were still there.

What saved part of it was cutting the hours. When I restricted the system to the middle of the London session, the average spread almost halved, and the trade count only fell by about a third. The net figure became small but positive, and it survived when I doubled my cost assumptions.

I also stopped assuming fills at the band itself. I moved the simulated entry one pip inside the touch, which removed a lot of trades that only looked good because they filled at the exact extreme and reversed immediately.
Post Reply