Recency Bias Trading: Why Last Month’s Results Distort Your Judgment.
By Sofia Harchich | Trading Psychologist & Behavioural Finance Writer | thewealthmirror.com
Four winning trades in a row don’t make the fifth one safer. Your brain disagrees, and it’s very convincing.
Something shifts after a good week. Position sizes creep up. Setups that would have been passed over a month ago start looking obvious. Conviction rises — not because the analysis got better, but because the last few trades worked, and the mind has quietly rewritten “the last few trades worked” into “this is how the market behaves now.”
That quiet rewrite has a name. Recency bias trading is the tendency to weigh the most recent information far more heavily than everything that came before it — treating a short, recent stretch of results as a reliable preview of what’s coming, when in most markets it’s closer to noise. It runs in both directions: a hot streak breeds overconfidence, and a rough patch breeds a conviction that the losing will never end. Neither feeling is a measurement. Both are memory playing tricks with weight, and both feel, from the inside, exactly like accumulated experience rather than a handful of recent data points wearing experience’s clothing.
Why the Last Few Trades Feel Like the Whole Story.
This bias is worth distinguishing from ordinary pattern recognition, which is a genuinely valuable skill in trading and shouldn’t be discarded out of a fear of being biased. The difference lies in the sample being judged: noticing that a specific setup has performed well across many months, in varied conditions, is legitimate pattern recognition. Noticing that the last three trades worked and concluding the market has fundamentally shifted is recency bias wearing pattern recognition’s clothes. The skill isn’t in ignoring patterns — it’s in being honest about how large a sample is actually being generalized from.
Working memory holds recent events far more vividly than distant ones — a well-documented pattern that shows up across contexts from courtroom testimony to financial decisions. Psychologists call the broader phenomenon the serial position effect: people tend to remember the first and last items in any sequence better than the ones in the middle, with the tail end of a sequence enjoying a particular advantage because it’s still sitting in short-term memory, unrehearsed and easy to retrieve. Whatever happened most recently sits closest to the surface, easiest to recall, and the mind mistakes ease of retrieval for importance.
The market doesn’t help. It provides a constant stream of new information, and each new candle nudges the recent past further to the front of attention while pushing everything older further into the background. A trader watching XAU/USD after three consecutive breakout wins isn’t wrong to notice the pattern — the error is in the leap from “this happened three times recently” to “this is what gold does now,” which is a much larger and much less supported claim than the noticing itself.
This tends to be strongest at specific moments: the end of a trading week, the close of a month, the day after a particularly sharp move. These are exactly the points where a trader is most likely to sit back and draw a conclusion about “how things are going” — and exactly the points where the most recent handful of trades has the loudest, least contested voice in the room.
The last few trades are the loudest data in the room. That doesn’t make them the most important.
How Recency Bias Quietly Resizes Every Position
The most costly version of this bias rarely shows up as a bad entry. It shows up as a bad size. After a run of wins, risk creeps upward — not through a deliberate decision to take on more risk, but through a felt sense that the current approach simply works, so slightly larger positions feel proportionate rather than reckless. The reasoning, if it gets articulated at all, sounds something like “this setup has been working, so it makes sense to lean into it a bit more” — which sounds prudent and is, in fact, exactly the mechanism the bias runs on.
After a run of losses, the opposite happens: risk shrinks defensively, sometimes below what the original strategy actually called for, driven by a conviction that the losing streak reflects something broken rather than something ordinary. A perfectly sound strategy can produce a losing week purely through the normal variance any probabilistic approach carries — and recency bias, in this direction, convinces a trader that a run of bad luck is actually a verdict on the strategy itself, prompting a retreat that has nothing to do with whether the underlying edge has changed at all.
Both movements happen below the level of a stated decision. Nobody sits down and decides “I will now increase my risk because recency bias has distorted my judgment.” The sizing simply changes, gradually enough not to be noticed in the moment, and the story arrives afterward to explain why it made sense at the time.
The size of the next trade often reveals more about the last five trades than about the next five.
Recognising Recency Bias Trading in Real Time.
- Check the sample size before trusting the pattern. Three or four trades is an anecdote, not a track record — a useful gut-check before adjusting size or conviction based on a short streak.
- Compare the current setup to the monthly chart, not the weekly one. A wider window dilutes the outsized weight the last few candles are carrying, showing a recent run as one small stretch inside a much longer story.
- Ask what would have to be true for this to be coincidence. If the honest answer is “not much,” the pattern is probably real; if the honest answer is “quite a lot,” recency is likely doing the talking.
- Keep position sizing rules written down and separate from mood. A fixed sizing rule, decided in advance, doesn’t care whether last week was good or bad — it applies the same regardless of the story currently being told about it.
- Review a full month of trades before drawing any conclusion about “how the market is behaving.” A month contains enough noise to average out what any single week cannot, and it’s a much fairer sample to judge a strategy against.
The correction isn’t ignoring the last few trades. It’s refusing to let them outvote everything that came before them.
There’s a useful test for telling a real pattern from a recency illusion: does it survive being restated in the driest possible terms? “Gold has broken this level and held three times this week” is a real pattern, stated plainly. “Gold always does this now” is the same observation inflated by recency into something it hasn’t earned yet. The plain version invites a check against a larger sample. The inflated version tends to resist checking, because checking is exactly what would deflate it.
The Deeper Pull: Why Presence Matters More Than Memory
Eckhart Tolle’s writing on presence describes a mind that habitually pulls attention into either the remembered past or the imagined future, rarely resting in what’s actually happening now. Recency bias is a precise example of that pull in financial form — not attention drifting into distant memory, but attention gripped by the very recent past and mistaking it for the present moment itself. The last three trades aren’t happening now. They already happened. Treating them as a live signal rather than a closed chapter is exactly the kind of confusion Tolle’s work points toward: the observer mistaking a memory for the moment, and reacting to the memory with the same urgency the present moment would deserve.
Noticing that pull — simply seeing “this is memory dressed up as prediction” — does more to correct the bias than any amount of willpower applied to the next trade itself. The observer that can name what’s happening is already a step removed from being run by it. This is a subtle but real distinction: trying to force conviction back down to a “correct” level rarely works, because the conviction was never really about the current setup in the first place. It was about three trades that already closed. Seeing that clearly tends to dissolve the false certainty far more reliably than arguing with it does.
A trade closed in the past has no vote in the trade opening now — unless memory dresses it up as one.
Where to Start
- Pull the last 30 days of trades and calculate the win rate honestly, rather than relying on how the week “felt.”
- Write a fixed position-sizing rule this week, before the next win or loss has a chance to influence it.
- Before the next trade, name out loud whether conviction is coming from the setup or from the last few results.
- See how recency bias tends to show up in your own trading specifically — the free quiz takes minutes.
About the Author
Sofia Harchich is a Trading Psychologist and Behavioral Finance Writer with a Master’s in Psychology. She works at the intersection of Jungian shadow work, neuroscience, and market behaviour — helping traders understand the psychology driving their decisions, not just the strategy.
Read more at thewealthmirror.com/about
