The reflex to close winners at the first sign of a pullback

[Shadow Trader]

Price moves a pip in favor of the position. Then it ticks back a quarter pip. The hand twitches toward the close button.

For a scalper, a floating profit feels fragile. It might vanish in the next tick. The urge to lock it in is powerful, and acting on it feels safe.

Repeated over hundreds of trades, though, the habit shapes the results: average wins get small, losses stay the size they were, and the payoff ratio deteriorates. The strategy ends up with a built-in negative skew.

What makes it hard is that each individual early exit seems reasonable. Sometimes price did reverse, and the exit saved the profit. The cases where it didn't are less memorable because nothing happened: the trade would have made more, but that's a hypothetical.

Reviewing the data helps. For each early close, record where price went in the next minute. A tally of how often it went further in favor than the exit point gives a direct measure of the cost of the reflex.

Some switch to exits placed as orders at the start of the trade, taking the decision out of the moment.

How much did early exits cost when the follow-up prices were checked?

[Amanhasnoname]

I did exactly the check you describe and it was humbling. Over about two hundred trades I tracked every manual early close and noted where price was a minute later. In roughly two out of three cases price went further in my favour, often reaching my original target, and the total I left on the table was bigger than my net profit for the period.

The trades where the early exit saved me stuck in my memory, which is why I kept doing it. The ones where it cost me just disappeared, because nothing bad happened on the screen, there was only a smaller win than there could have been.

What helped was placing the target and stop as orders the moment I enter and then minimising the trade window. If I really think the trade is going wrong I'm allowed to close it, but I have to type a short reason in my journal first. Most of the time, typing the reason makes me see there isn't one.