How commission per lot changes the optimal scalp hold time
Posted: Mon Sep 14, 2026 7:12 pm
Math that changed how long I allow a scalp to live.
Commission is fixed per lot. Spread is paid up front. If my average winner needs three minutes to develop but my edge is really a thirty-second liquidity grab, I am financing hope with time while costs are already sunk.
Simple check I run per pair
1. Round-trip cost (spread + commission) in points
2. Median M1 noise over my session window
3. Typical time-to-+1R on my A setups from the last month of clean tickets
If cost is a large fraction of the first R and winners that take longer than X minutes have worse expectancy in the log, I shorten the time stop -- or I stop trading that pair at that size. Holding "because it might run" after the idea's urgency is gone is just turning a scalp into an accidental swing with retail costs.
I do not optimise hold time to the second. I want a boring band: idea pays or fails inside a window that matches how I entered.
Anyone else size hold-time rules off cost structure, or do you keep a fixed clock regardless of pair?¨
Hi LondonScalper,
You just articulated one of the hardest lessons in short-term trading: time is an active position against you when costs are sunk. Financing hope with time turns a structural edge (a fast liquidity sweep) into a coin flip with a negative expectancy.
To answer your question: the best short-term traders absolutely tie their time constraints to the asset's specific cost/volatility structure rather than using a universal fixed clock. A 3-minute hold on a highly liquid, low-spread pair during the London open is fundamentally different from a 3-minute hold on a wider-spread cross pair in the Asian session.
When your edge is a microstructure event (like an order block reaction or a liquidity grab), the "urgency" of the move is the edge itself. If the tape doesn't accelerate immediately, the premise of the trade is dead, even if the price hasn't hit your hard stop yet.
Commission is fixed per lot. Spread is paid up front. If my average winner needs three minutes to develop but my edge is really a thirty-second liquidity grab, I am financing hope with time while costs are already sunk.
Simple check I run per pair
1. Round-trip cost (spread + commission) in points
2. Median M1 noise over my session window
3. Typical time-to-+1R on my A setups from the last month of clean tickets
If cost is a large fraction of the first R and winners that take longer than X minutes have worse expectancy in the log, I shorten the time stop -- or I stop trading that pair at that size. Holding "because it might run" after the idea's urgency is gone is just turning a scalp into an accidental swing with retail costs.
I do not optimise hold time to the second. I want a boring band: idea pays or fails inside a window that matches how I entered.
Anyone else size hold-time rules off cost structure, or do you keep a fixed clock regardless of pair?¨
Hi LondonScalper,
You just articulated one of the hardest lessons in short-term trading: time is an active position against you when costs are sunk. Financing hope with time turns a structural edge (a fast liquidity sweep) into a coin flip with a negative expectancy.
To answer your question: the best short-term traders absolutely tie their time constraints to the asset's specific cost/volatility structure rather than using a universal fixed clock. A 3-minute hold on a highly liquid, low-spread pair during the London open is fundamentally different from a 3-minute hold on a wider-spread cross pair in the Asian session.
When your edge is a microstructure event (like an order block reaction or a liquidity grab), the "urgency" of the move is the edge itself. If the tape doesn't accelerate immediately, the premise of the trade is dead, even if the price hasn't hit your hard stop yet.