The Biases That Cost Traders Money

Cognitive biases are usually presented as a list of names to learn. That is the least useful form of the information, because knowing a bias exists does not stop it — decades of research say the effects persist in people who can define them. What helps is recognising the specific behaviour each one produces at a chart, and then building a rule that makes the behaviour impossible.

Loss aversion and cutting winners

Losses hurt more than equivalent gains please — roughly twice as much, in the classic experiments. In trading this produces one behaviour with two halves, and both halves lose money.

• Winners get closed early, because an unrealised gain feels like something that can still be taken away. Closing it converts the discomfort into a certainty.
• Losers get held, because closing one makes the loss real. While it is open it is still theoretically a trade that comes back.

The result is a portfolio of small wins and large losses, which is precisely the shape that turns a positive expectancy into a negative one. Note the arithmetic: cutting winners at 0.7R while letting losers run to 1.3R turns a 2:1 plan into a 0.54:1 plan, and Track 5's break-even table then demands a 65% win rate the method was never designed for.

Where it shows in your records

1. Compare planned R against realised R. A persistent gap on winners and none on losers is this bias, measured.
2. Count trades closed before the target with no rule permitting it.
3. Count trades where the stop was widened. Any number above zero is worth a rule.
4. Check the average holding time of winners against losers. Losers held longer is the signature.

Recency and over-weighting last week

Recent events feel more representative than they are. Three losses in a row feels like evidence the strategy has stopped working; three wins feels like confirmation it is working better than expected.

Track 5's streak arithmetic says otherwise: a 40% win-rate method produces a run of six losses regularly, and over 200 trades it is close to certain. The streak carries no information at all, and recency makes it feel like the most important information available.

• Count in hundreds, not weeks. A week is 2 to 20 trades, which is noise at any win rate.
• Know your expected streak length before you start, so a run of six is a number you predicted rather than a surprise.
• Keep size fixed by rule, which removes the lever recency reaches for first.

Confirmation and timeframe shopping

Once you have a view, you notice the evidence that supports it and skim past the evidence that does not. This is not carelessness; it is how attention works, and it operates before you are aware of having a view.

Track 3's multi-timeframe lesson described the trading form precisely: you want to be long, the daily says no, so you check H4, then H1, then M15, and one of them agrees. The analysis was real and the conclusion preceded it.

The other shapes it takes

• Indicator shopping. Adding an oscillator that agrees, after the first one did not.
• Reading only the news that supports the position, which is easy because there is always enough of both.
• Re-drawing levels so the one you need is where you need it.
• Remembering the wins from a setup and not the losses, which is why counting matters more than recalling.

Designing around a bias instead of resisting it

The practical conclusion of this lesson: do not plan to be more disciplined. Discipline is a finite resource, it is lowest exactly when the bias is strongest, and a plan that depends on it fails at the moment it is needed.

Build rules that remove the decision instead.

Why this works when willpower does not

1. The decision is made in a calm state and executed in an agitated one, rather than made in the agitated one.
2. Breaking a written rule is visible. It becomes a line in the journal, which the weekly review reads.
3. It converts a character question into a process question. "Was I disciplined?" has no useful answer; "did I deviate, and where?" has a number.

Every rule in that table already exists somewhere in this site — stops and targets on the entry ticket in Track 1, the sizing formula and abandon criteria in Track 5 and Track 6, the fixed timeframe pair in Track 3. That overlap is not repetition. It is that most of what looks like trading technique is really a set of devices for removing decisions from moments when you should not be making them.