All That Glitters: The Research Says You Don't Pick Stocks — Your Attention Does

Behavioral finance has a public relations problem: everyone knows the lab experiments and almost nobody knows the account records. Ask a trader to name a behavioral finding and you’ll get prospect theory, loss aversion, the framing effects — the elegant, quotable stuff from the psychology department. Fine work, and we’ve written about it. But the studies that should actually frighten a retail trader were not run on undergraduates choosing between hypothetical gambles. They were run on real brokerage accounts, at population scale, with tax records and driving records bolted on. And they are considerably more damning.

They point at a conclusion most traders have never seriously entertained: that the fundamental decision in your trading — which instruments you even consider — is very likely not being made by you.

Key takeaways

• Barber & Odean (2008) found individual investors are net buyers of attention-grabbing stocks: those in the news, with abnormal volume, or extreme one-day moves. Institutions aren’t.
• The cause is a search problem. You buy from thousands of candidates, so attention filters them. You sell only from what you own — so selling isn’t attention-driven.
• Grinblatt & Keloharju (2009) found speeding tickets predict trading frequency, controlling for wealth, income and age. How much you trade is partly a personality trait.
• The uncomfortable conclusion: your watchlist is not a strategy. It’s an attention funnel — and in 2026 it’s being fed by an industrial-grade attention machine.

The search problem: why buying and selling are not symmetrical

The foundational paper here is Brad Barber and Terrance Odean’s “All That Glitters: The Effect of Attention and News on the Buying Behavior of Individual and Institutional Investors,” published in the Review of Financial Studies in 2008. Its finding is simple enough to state in a sentence and awkward enough to sit with for a week: individual investors are net buyers of attention-grabbing stocks — stocks in the news, stocks with abnormally high trading volume, and stocks with extreme one-day returns.

The mechanism Barber and Odean propose is the part worth memorizing, because it is not about greed or stupidity. It is about arithmetic. When you go to buy, you are choosing from thousands of possible stocks. That is a genuinely impossible search, so the mind does what minds do with impossible searches: it quietly reduces the candidate set to the things that have already presented themselves. You consider only the stocks that first caught your attention. But when you go to sell, you are choosing from the handful you already own — a trivial search, requiring no filter. So attention shapes your buying and barely touches your selling.

Sit with the implication. It means the single most consequential act in discretionary trading — determining the universe you’re choosing from — is being outsourced, by default, to whatever is loud. And the stocks that are loud are, by construction, the ones that have already moved on volume and news.

Attention doesn’t just influence what you buy. It decides what you were ever able to buy — and it hands you that list only after the move has started.

Note also who doesn’t do this. Barber and Odean found the pattern in individual investors, not institutions. The professionals aren’t immune to bias, but they don’t have this particular bug, because they run screens against pre-defined criteria. Their choice set is constructed on purpose. Yours, if you’re not careful, is constructed by a feed.

The speeding ticket study: how much you trade is partly who you are

If “All That Glitters” explains what retail traders buy, the second study explains how often — and it does so with one of the great research designs in all of finance.

Mark Grinblatt and Matti Keloharju, publishing in the Journal of Finance in 2009, took advantage of the fact that Finland keeps unusually comprehensive records. They combined equity trading data with investors’ tax filings, their driving records, and the mandatory psychological profiles from military conscription. Then they asked a question nobody had been able to ask before: does personality predict trading behavior?

It does. Controlling for wealth, income, age, number of stocks owned, marital status and occupation, they found that overconfident investors and investors most prone to sensation seeking trade significantly more frequently. And their headline proxy for sensation seeking was the number of speeding tickets an investor had accumulated. (Sports car ownership worked too, though less powerfully.) The man who cannot keep to the limit on the motorway is, statistically, the man who cannot keep his hands off the buy button.

This is a profound and rather deflating result. It reframes trading frequency — which most traders experience as a series of reasoned, situation-specific decisions — as substantially a disposition. You are not, in many cases, trading a lot because the market is offering a lot. You are trading a lot because of who you are, and then generating market-shaped reasons afterwards. The reasons feel like causes. The research says they’re mostly commentary.

Put the two together and you have the modern retail trader

Here is the composite portrait the account-level literature paints, and it is not flattering. A trader whose candidate list is selected by whatever is loudest. Who therefore buys, structurally, after the move and into the crowd. Whose frequency of trading is driven substantially by a temperament trait rather than by opportunity. And who, per Barber and Odean’s earlier work, gives up something like 6.5 percentage points a year to that turnover — and who, per the Taiwan day-trading data, has roughly a 5 percent chance of being consistently profitable.

Now notice what has happened to that trader’s environment since these papers were written. In 2008 the attention machine was CNBC and a newspaper. In 2026 it is an infinite, algorithmically optimized, personalized firehose: push alerts, X, TikTok, Discord, AI screeners that never run out of ideas, and a brokerage app engineered — as Daniel Schlaepfer argues — to monetize activity rather than improve it. The bias hasn’t changed. The apparatus for exploiting it has been industrialized.

The honest caveat: attention is not always the enemy

It would be too neat to conclude that anything attention-grabbing is untradeable. It isn’t. Volume and news often mark genuine catalysts, and some of the best traders alive make their living precisely there — Daljit Dhaliwal trades macro events, which are about as attention-grabbing as markets get.

But look at the difference, because it’s the whole lesson. Dhaliwal defined his edge first, from his own records, and then went looking for the specific events that matched it. The attention didn’t select the trade; his written criteria did, and the event merely had to qualify. That is attention as a filter you designed. The failure mode is attention as the thing that chose for you — where the alert arrives, the chart is already moving, and the thesis is assembled in the ninety seconds between noticing and clicking.

Same stock. Same news. Completely different trade, and the difference doesn’t live in the market. It lives in whether the criteria existed before the ticker did.

From belief to behavior: is your attention or your process choosing?

The good news about account-level research is that it studies exactly the thing you also have: an account. Every finding above converts directly into something you can check in your own record.

The fix that follows from the research is not “pay more attention.” It is the reverse: shrink the choice set on purpose. Write down the setups you trade and the instruments you trade them in. Make a stock qualify against criteria that existed before you saw it. Trade from a screen, not from a feed. This is unglamorous, and it is the single structural change most likely to move a retail account, because it attacks the decision that all the other decisions inherit from.

Resources and further reading

• Attention: Barber, B. M. & Odean, T. (2008), “All That Glitters: The Effect of Attention and News on the Buying Behavior of Individual and Institutional Investors,” Review of Financial Studies 21(2): 785–818.
• Personality and turnover: Grinblatt, M. & Keloharju, M. (2009), “Sensation Seeking, Overconfidence, and Trading Activity,” Journal of Finance 64(2): 549–578 — Finnish trading records combined with tax filings, driving records and psychological profiles.
• The cost of turnover: Barber, B. M. & Odean, T. (2000), “Trading Is Hazardous to Your Wealth: The Common Stock Investment Performance of Individual Investors,” Journal of Finance 55(2): 773–806.
• The outcome: Barber, Lee, Liu & Odean’s Taiwan day-trading research — roughly 5% of day traders consistently profitable, with most losing money net of fees.
• The disposition effect: Odean, T. (1998), “Are Investors Reluctant to Realize Their Losses?”, Journal of Finance — the companion account-level finding on selling winners early and holding losers.

Frequently asked questions

Essay in Gecko’s trading psychology series. Findings are drawn from Barber & Odean (2008, 2000), Grinblatt & Keloharju (2009), Odean (1998), and the Taiwan day-trading research of Barber, Lee, Liu & Odean. Figures are approximate and specific to the samples, markets and periods studied, and may not generalize to your situation. Gecko is an educational and informational tool. Nothing here is financial, investment, or trading advice, and nothing here is a recommendation for or against any security or strategy. Trading carries substantial risk of loss.