The four mistakes that account for most retail losses

When the desk started writing every trade down in 2009, the purpose was not analysis. It was simply that memory had become unreliable — winning trades felt larger in recollection than they were, and losing ones had a way of becoming other people's fault. The journal was meant to settle arguments with myself.

What it did instead was reveal a pattern. Sorted by loss size rather than by date, the entries stopped looking like a series of unrelated bad days and started looking like four situations repeating, in different instruments, across different years.

One — sizing up after a winning streak

The largest single losses in the journal all shared a feature: they came directly after a run of wins. Not because the market changed, but because confidence did. Three good trades in a row produce a feeling that the next one is more likely to work, and that feeling expresses itself as a larger position.

The market does not know about the streak. A larger position simply means a larger loss when the fourth trade behaves like any other trade. The fix is mechanical rather than psychological — position size is calculated from volatility, and the calculation does not have a field for how the last three went.

Two — averaging into an invalidated position

Adding to a losing position feels like conviction and is usually the opposite. The question that separates the two is simple: does the reason you entered still hold? If the setup has been invalidated and you are adding anyway, the trade has stopped being an idea and become an attempt to be proved right.

In the journal, every one of these entries had the same tell — the reasoning field for the second entry was vaguer than the first. When the reason gets harder to write down, that is the signal.

Three — cutting winners out of relief

This one costs less per occurrence and more in total. A position moves into profit, the discomfort of being wrong lifts, and closing it converts an uncertain feeling into a certain small gain. It feels like discipline. It is the same impulse as averaging down, running in the opposite direction.

The consequence is arithmetic rather than emotional. If losses run to their planned size and wins are cut short of theirs, the expected value of the whole system inverts, no matter how often you are right.

Four — trading a market that had not set up

The quietest of the four and the hardest to notice, because each individual instance is small. The market is open, nothing qualifies, and a marginal setup gets taken anyway — because sitting still does not feel like working.

Aggregated over a year, these trades were close to a coin flip minus costs. That is not a strategy losing money slowly; that is paying brokerage for the privilege of feeling busy.

What to do with this

None of these four are insights about markets. They are observations about behaviour, which is why reading about them changes very little on its own. What changed things for this desk was the record — not the knowledge that these mistakes exist, but the monthly evidence of which one was currently costing the most.

1. Log the reasoning for every entry, in a sentence, before placing the order. Vague sentences predict bad trades.
2. Review weekly, sorted by loss size rather than by date. The pattern lives at the top of that list.
3. Pick the single most expensive pattern and work only on that one for a quarter. Attacking all four at once fixes none of them.

The four have not gone away. They occur less often, and the ones that do get caught earlier, which is the only realistic definition of progress available here.