The Stop-Loss You Set vs. The Stop-Loss You Moved

Two trades close for the same loss. One had a stop placed before entry, at the level that invalidated the setup, and it got hit. The other had a stop placed before entry too — but somewhere in the middle of the trade it got dragged further away, once, then again, and by the time it finally closed the loss was three times the size originally planned. In most journals, both trades show up as one row each: instrument, direction, P&L. Identical. They are not the same decision, and treating them as the same number is exactly how the second habit survives for years without being caught.

Why the moved stop is invisible by default

A journal that only logs the stop-loss price at the moment a trade closes cannot tell these two trades apart. It sees the final stop, not the first one. The moment where the decision actually happened — the point mid-trade where the plan was quietly rewritten — leaves no trace unless something was recording before that edit as well as after it. That is not a minor gap. It is the single most common way a defined-risk strategy turns into an undefined-risk one without the trader ever deciding, on purpose, to change strategy.

The reasoning in the moment is rarely reckless-sounding. “It just needs a bit more room.” “The level held on the last two touches.” “I don't want to get stopped out right before it turns.” Each one sounds like a market read. What it actually is, almost every time, is the original risk decision being replaced by however much loss currently feels tolerable to sit with — which is a number that gets larger every time it is asked, because the trade is now more expensive to be wrong about than it was five minutes ago.

What moving a stop is actually evidence of

A stop set before entry is a risk decision made with a clear head, before there is money on the line to defend. A stop moved after entry is a decision made under the exact opposite conditions — with a loss already open, with the sunk cost of being wrong already felt, and with every incentive pointing toward avoiding the discomfort of closing at a loss right now. Those are not two versions of the same judgment. The second one is structurally worse at the moment it is made, regardless of how the trade eventually turns out.

This is also why outcome is the wrong test for it. A moved stop that ends up back in profit does not retroactively make the decision sound — it makes it lucky, and luck is exactly what teaches the habit to repeat. The trade that would have proven the point, the one where moving the stop turned a planned 1R loss into an account-damaging one, is the trade that does not get talked about afterward. Only a record that keeps both numbers — the original stop and the one it became — makes that asymmetry visible instead of remembered selectively.

What to actually track

• The stop-loss at entry, recorded before the trade is placed, not filled in afterward from memory.
• Every subsequent change to that stop, in the direction of more room — moving a stop closer to lock in profit is a different action and not the one this is about.
• The gap between planned risk and realized risk on trades where the stop moved, in R or in account percentage, not just in currency.
• How often it happens — a single instance is a bad day; a pattern across a month is a hole in the strategy that no amount of good setups will fix.

None of this requires guessing at intent. It only requires knowing what the stop actually was at two points in time instead of one.

Why this is hard to catch by hand

A stop moved manually in a MetaTrader 5 terminal does not announce itself — the platform shows the current order, not its history, and a manual spreadsheet only gets updated with whatever the trader remembers to write down, which is rarely the moment of the edit itself. getALPHA syncs closed trades directly from MT5, and the process review in getALPHA's AI coach checks realized risk against what a trade was sized for at entry, so a widened stop shows up as a pattern to look at rather than a number that quietly blends into the average.