ForexVue
Level 8 · Lesson 6 of 14 · 6 min read

Confirmation Bias in Trading

You decided EUR/USD is going up. Now every indicator magically confirms it. That's the trap.

Laurent Researched and written by

How It Works

You've decided EUR/USD should go up based on your initial analysis. Now you look at the chart: "The 50 EMA looks like support. RSI is rising. There's a bullish candle forming." Meanwhile, you don't notice: price is below the 200 EMA (bearish), the weekly chart shows a downtrend, and there's major resistance 30 pips above.

You didn't lie to yourself. You just selectively focused on evidence that confirmed what you already wanted to do. This is confirmation bias, and every human does it. It's how our brains manage information overload: we filter for relevance, and "relevant" becomes "confirms my existing belief."

A Real Scenario

A trader's story:

Back in 2024, with USD/JPY pressing the 152.00 area, a trader read a bullish analysis from a well-known forex analyst. They opened the chart and started looking for buy opportunities. They found: bullish divergence on the H4 RSI, a support zone at 149.50, and a bullish engulfing candle forming. They went long.

What they didn't look for: the daily chart was in a clear downtrend below the 200 EMA. The weekly candle was a bearish engulfing. The BoJ had just signaled potential policy normalization (JPY bullish). Three strong bearish signals were right there on the chart, but the trader wasn't looking for bearish signals because they'd already decided to buy.

Here is the full bearish evidence list that was sitting on the same screens and was filtered out by the bias:

  • H4 rejection wick at 152.00, a level that capped 3 prior attempts in the last 14 trading days.
  • RSI bearish divergence on the daily: price made a higher high but RSI made a lower high, a classic momentum exhaustion signal.
  • DXY breaking down below its 50-day moving average that same morning, removing the dollar-strength tailwind that had supported the pair for 6 weeks.
  • BoJ verbal intervention earlier that week: the Finance Minister explicitly named 152.00 as a level of concern, raising actual-intervention odds materially.
  • Japanese real wages turned positive YoY for the first time since early 2022, a fundamental shift that BoJ hawks were citing as justification for further normalization.

Five concrete bearish data points, all visible, all ignored. The trade closed at -2.1R the next day when BoJ stealth intervention pushed the pair 180 pips lower in a 90-minute window. The analyst's bullish call was not the problem. The trader's failure to look for evidence that contradicted it was.

The Anti-Confirmation-Bias Protocol

4-step protocol you must apply on every trade:
  1. Before placing the order, write 3 specific reasons the trade could FAIL. Not generic ("market is risky"). Specific: "DXY just broke below 50-day MA," "RSI shows bearish divergence on daily," "NFP tomorrow could reverse this." If you cannot find 3 specific reasons, you have not done the analysis.
  2. Set a calendar reminder 24 hours after entry to revisit those reasons honestly. A reminder, on your phone, with a sound. Not "I'll check later." Real notification.
  3. If 2 of the 3 fail-reasons are now actually playing out, exit at market. Not "wait and see." Not "give it a bit more room." Market exit, immediately. Your reasons-to-fail list was the kill-switch. Use it.
  4. After the trade closes, score yourself on whether you applied steps 1 to 3. Yes/No journal entry. Track this score weekly. If it drops below 90%, you are sliding back into confirmation bias and need to reset.
The research says: Psychologists have found that once people form a hypothesis, they search for information that confirms it far more readily than they seek out disconfirming evidence (see Nickerson's 1998 review of confirmation bias research). In trading, this means the bullish signal jumps out at you when you want to buy, while the bearish signals fade into the background.

The Devil's Advocate Exercise

Before every trade, force yourself to answer: "What are 3 reasons this trade could fail?"

  • "Resistance is only 40 pips above, limiting upside."
  • "NFP is tomorrow, which could reverse any move."
  • "The weekly trend is still bearish."

If you can find 3 solid reasons against the trade and still take it, fine. At least you made an informed decision. If the reasons against are stronger than the reasons for, skip it.

Try it yourself: On your next 10 trades, write down 3 bullish arguments and 3 bearish arguments BEFORE entering. Review them a week later. You'll find that the trades where the bearish arguments were strong (but you ignored them) were your worst performers.
✅ Check your understanding
The "devil's advocate exercise" requires you to:
✅ Check your understanding
In the anti-confirmation-bias protocol, if 2 of your 3 pre-written fail-reasons start playing out, you should:

Key Takeaways

  • • Confirmation bias = selectively noticing evidence that supports your existing belief.
  • • In trading: you ignore bearish signals when you want to buy, and ignore bullish signals when you want to sell.
  • • The fix: before every trade, actively list 3 reasons why it could fail.
  • • If you can't find reasons against the trade, you probably aren't looking hard enough.
Glossary: Technical Analysis
Overtrading: Doing Too Much Recency Bias and Anchoring