Five trades is not a track record, because it is merely a handful of coin flips with a story attached. Your brain treats the most recent results as the most informative, which is exactly backwards when the sample is tiny.
Recency bias is the tendency to weight recent events more heavily than older ones when you form judgments. In trading, it shows up as confidence after a short win streak, panic after a short losing streak, and strategy changes based on outcomes that could easily be noise.
The fix is not willpower; instead, it is thinking in sample sizes and expected value.
What recency bias does to your decisions
Your memory is not a fair archive, since it highlights what just happened and fades what happened last month. After three winners, the next trade feels safer than it is, while after three losers, a perfectly good setup feels broken.
This is a prediction problem, because you are trying to estimate the quality of a strategy from a handful of outcomes, and a handful of outcomes is a terrible estimator. A strategy with a real edge can lose five in a row, while a strategy with no edge can win five in a row. Both happen more often than most people expect.
There is a second layer, because recency bias does not just change what you believe; it also changes what you do. You size up after wins, skip valid setups after losses, or abandon a plan mid-stream, and this behavior change then produces new results, which feed the same loop.
Sample size: the number that should slow you down
Sample size is the count of independent observations behind a conclusion. If you want to know whether a strategy works, the question is not whether it worked last week. The question is how many trades you have, and whether those trades were similar enough to compare.
Small samples produce wide ranges of plausible outcomes, so if you flip a fair coin ten times, five heads is normal, but so is seven or three. Nothing about seven heads means the coin is biased, and the same logic applies to a trading method with a modest edge.
This cuts both ways, and that is the point. A short hot streak is weak evidence of skill, while a short cold streak is weak evidence of failure. Treating either as proof is a category error, not a judgment call.
Sample size also has a quality dimension. Twenty trades taken in the same market regime, on the same setup, may tell you less than ten trades spread across different conditions. Correlated outcomes inflate the apparent sample.
If all your trades were long tech during one rally, you have closer to one observation than twenty.
Expected value: the number that should guide you
Expected value is the average result you would get per trade if you repeated the strategy many times. It combines win rate and payoff size. A method that wins 40 percent of the time but makes twice as much on winners as it loses on losers can have a positive expected value.
A method that wins 70 percent of the time with tiny gains and large losses can have a negative one.
You cannot observe expected value directly, so you estimate it from results, but the estimate is noisy when the sample is small. This is why a run of five trades cannot confirm or refute an edge; it can only nudge your estimate, and usually by less than it feels like.
Expected value also explains why streaks are normal, since even a positive-expectancy strategy produces losing runs. The length of those runs depends on win rate and payoff structure, not on your discipline or your mood. Expecting smooth equity curves is a separate bias wearing a friendly mask.
For traders: weeks to several months
If your horizon is weeks to a few months, your sample is almost certainly too small for confident conclusions. That is a constraint, not a personal failing.
The practical response is to define in advance how many trades a test needs before you judge it, and to log every trade rather than the memorable ones.
Keep a simple record: setup, entry reason, exit reason, result in risk units. Review it on a schedule, not after every loss. A scheduled review may give recent trades less weight than a review conducted the moment a position closes.
Position sizing matters more than opinion here. If a short losing streak can force you to stop trading, your size is too large for the sample uncertainty you face. Risk per trade should survive a normal losing run without changing your process.
Be careful with strategy switching, because changing rules after three losses feels like adaptation, yet often it is recency bias with a spreadsheet. Consider setting a minimum number of trades before you revise a plan, and write down why you chose that number.
For investors: longer time horizons
Longer horizons reduce the problem but do not remove it. An investor reviewing quarterly returns is still looking at a small sample. A fund that trails for a period can be executing exactly as intended, and one that leads for a period can be benefiting from a style tailwind.
For investors, the relevant sample is often market cycles, not months, which is a high bar, and it means most short-term performance comparisons carry little information. A useful habit is to separate process from outcome: did the portfolio behave as designed, or did it happen to go up?
Rebalancing decisions are a common place where recency bias sneaks in. Selling what just fell and buying what just rose is a recency-driven move dressed as discipline. A written policy with fixed rules reduces the room for that drift.
Also watch how you read performance reports, since a single strong year can dominate your impression of a decade. Look at rolling periods and at the range of outcomes, not just the latest number.
Where this reasoning can go wrong
The counter-argument is fair: sometimes recent information is genuinely informative. A regime change, a structural shift in a business, or a broken process can make old data less relevant. Recency bias is a bias because it is usually wrong, not because it is always wrong.
The distinction is whether new information changes the underlying mechanism: a few trades do not change how a market works, although a change in a company’s competitive position might. Therefore, ask which one you are looking at before you rewrite your plan.
There is also a risk of overcorrecting, because some people use sample size as an excuse to ignore all feedback. If a process is clearly broken, waiting for a large sample is its own mistake. The goal is proportion: give recent results the weight they deserve, not more and not zero.
Conclusion: judge the process, not the streak
Your last five trades are a tiny, noisy sample, and they can feel like a verdict because your brain is built to weight what just happened. That does not make them evidence of much.
Track sample size, estimate expected value, and decide in advance how much data a change requires. For traders, that means a written review schedule and sizing that survives losing runs, while for investors, it means judging process against policy over cycles rather than quarters.
Recent results are information; they are just smaller information than they feel like.
- Confirmation Bias: How to Actively Seek Evidence Against Your Trade
- Keeping a Trading Journal That Improves Results
- Discipline and Patience: The Trader’s Most Underrated Skills
Frequently Asked Questions
What is recency bias in trading?
Recency bias is the tendency to give more weight to recent events than to older ones when making judgments. In trading, it shows up as overconfidence after a short win streak or doubt after a short losing streak. The recent outcomes feel more informative than they actually are.
How many trades do I need before I can judge a strategy?
There is no universal number, because it depends on win rate, payoff size, and how similar the trades are. What matters is setting a minimum in advance and logging every trade, not just the memorable ones. A small sample of correlated trades tells you less than the count suggests.
Can a good strategy lose several trades in a row?
Yes. Losing runs are a normal feature of any strategy with less than a perfect win rate, and their length depends on the win rate and payoff structure. A short losing streak does not by itself mean the edge has disappeared.
What is expected value in trading?
Expected value is the average result per trade if you repeated the strategy many times. It combines win rate with the size of wins and losses, so a lower win rate can still produce positive expected value. You can only estimate it from results, and small samples give noisy estimates.
How do I reduce recency bias in my decisions?
Log every trade, review on a fixed schedule rather than after each result, and set a minimum sample before changing rules. Writing down your process and position sizing in advance limits how much a recent streak can steer you. Separate whether the process worked from whether the outcome was good.
AI Notice: This article was created wholly or predominantly with the assistance of artificial intelligence and was published without human editorial review.
This article is for general information only and does not constitute investment advice. Always do your own research before making trading decisions.