Recency bias: why the last thing that happened feels like the future

Ask someone how risky the stock market is and their answer will depend enormously on when you ask. In the depths of a crash, people describe stocks as a casino no sane person would touch. Eighteen months into a bull run, the same people describe them as easy money and wonder why anyone holds bonds. The market didn't change its nature in between. Their reference window did. Recency bias is the brain's tendency to overweight recent events and assume the near past will continue into the future — and in investing, it's one of the most expensive habits there is, because it whispers 'buy' exactly when prices are highest and 'sell' exactly when they're lowest.

Why the brain extrapolates

Your mind is a pattern-completion engine built for a world where recent information usually was the best guide: if it rained the last three days, betting on rain today was smart. Markets punish this wiring, because prices already reflect the recent trend, and trends in asset prices mean-revert more often than they persist. So the extrapolation that served your ancestors — 'the recent past predicts the near future' — becomes a machine for arriving late to every move: piling into whatever just went up and fleeing whatever just went down, which is the precise opposite of buying low and selling high.

The fingerprints of recency bias

• Chasing last year's winner: pouring money into the fund, sector, or asset class with the best recent returns, right as it becomes most expensive.
• Panic during drawdowns: a 20% drop feels like the start of a permanent decline rather than a normal, recurring event that markets have always recovered from.
• Recency in your risk tolerance: feeling aggressive after gains and timid after losses, so your 'risk tolerance' tracks the market instead of your actual life.
• Recency about inflation or rates: assuming whatever the current environment is will last forever, and reshaping a 30-year plan around this year's headline.
• Performance-chasing advisors and picks: judging a strategy by its last 12 months rather than its full-cycle behavior.

The record recency ignores

The single most useful fact against recency bias is that market history is a long series of scary declines followed by recoveries that felt impossible at the bottom. The broad US stock market has fallen 20% or more roughly once every handful of years and has, over every long period in its history, gone on to new highs afterward — though past performance never guarantees future results, and this is education, not a prediction. The point isn't that markets always bounce on a schedule; it's that a person who extrapolates the last six months into the next thirty years is using the least reliable possible sample.

Building a recency-proof system

1. Automate contributions on a fixed schedule. Dollar-cost averaging buys more shares when prices are low and fewer when high — the mechanical opposite of recency-driven timing.
2. Write an investment policy statement while calm, stating your allocation and the (rare) conditions under which it changes. Reread it during extremes instead of deciding fresh.
3. Rebalance on a calendar, not a feeling. Rebalancing forces you to trim what's recently soared and add to what's recently sagged — selling high and buying low by rule.
4. Zoom the chart out. Before reacting to a move, look at a 20-year chart instead of a 20-day one; the panic-inducing drop usually vanishes into the noise.
5. Judge strategies over full cycles, not trailing 12 months. A year of returns is a mood, not evidence.

The bottom line

Recency bias takes the most recent stretch of market weather and paints it across your entire horizon, which is why it reliably buys high and sells low. You can't stop the extrapolation reflex, but you can outvote it with structure: automate the buying, write the plan while calm, rebalance by the calendar, and zoom the chart out until this month becomes a rounding error. The investors who do well through cycles aren't the ones who read the last six months best. They're the ones who arranged, in advance, not to be asked.