Markets: A Price Is a Bet, Not a Measurement

A company reports the highest profits in its history. The share price falls that morning.

This looks irrational until you understand what a price is. A price is not a measurement of how a company is doing. It is a collective bet on how it will do, made by people who already read the same news you did.

If everyone expected record profits, then record profits are already in the price. What moves the price is the gap between what was expected and what arrived. Record profits that fall short of what was expected are, in the only sense the market cares about, bad news.

Once you hold that idea, most of the confusing behaviour of markets becomes considerably less confusing.

Prices as aggregated expectations

A market price is produced by a very large number of people, each acting on their own information and judgement, buying and selling until they reach a number where the last buyer and the last seller are content.

That number contains an enormous amount of information — more than any individual participant possesses. Nobody set it. It is an emergent property of many people acting independently, which is why it can be surprisingly good at incorporating news, and why no committee could replicate it.

Two consequences follow immediately.

Public information is mostly already reflected. By the time you read something in the news, so did everyone else, including people whose job is reacting to it within milliseconds. Acting on public information is not usually an edge, because you are not the first to have it.

You are betting against the aggregate, not against a company. Buying something because a company seems well run only makes sense if you believe the market has underestimated how well run it is. "This is a good company" is not a thesis; "this is better than the price implies" is.

This is the useful core of the efficient markets idea, and it is worth stating in its honest form rather than the caricature. The strong claim — that prices are always exactly right — is clearly false; bubbles and crashes happen. The defensible claim is narrower and more useful: prices are hard to beat consistently, because a great many well-resourced people are trying to beat them too. That is a Red Queen situation, and it is why most active managers underperform simple index funds over long periods, after fees.

Where the models break

Markets are not smoothly rational machines, and it is worth being clear about where the tidy picture fails.

Distributions are not normal. Standard financial models often assume price movements follow a bell curve. They do not. Extreme moves happen far more often than that assumption predicts — the tails are fat, closer to a power law. Risk models built on the wrong distribution understate danger in exactly the situations where getting it right matters most.

Everyone using the same model correlates behaviour. If many institutions use similar risk systems, they receive similar signals and act at the same time. Diversification that exists on paper disappears in a crisis, because the models told everyone to sell simultaneously. The model itself became a source of the risk it was measuring.

Liquidity is not constant. The ability to sell at a quoted price is an assumption that holds until it is most needed. Markets that seem deep in calm conditions can become thin very quickly.

Narratives move prices before numbers do. Because prices are bets on the future, they respond to changes in the story about a company — a new competitor, a shift in the growth outlook — often well before those changes appear in reported results.

None of this makes markets useless or fake. It means they are a mechanism with known failure modes, and that models are maps rather than the territory.

What this implies for an individual

The practical conclusions are unglamorous and reasonably well established.

Your purchase price is not information. The market does not know what you paid, and it has no bearing on what an asset is worth now. Holding a loser because selling would "lock in" the loss is loss aversion, and it is one of the most expensive habits available.

Time in the market contributes more than timing. Missing a small number of the best days substantially reduces long-run returns, and those days are clustered unpredictably, often near the worst ones. Since you cannot reliably identify them in advance, staying invested through cycles beats attempting to step aside.

Costs compound against you. Fees are one of the few certain quantities in investing. A small annual difference in cost becomes very large over decades, for the same reason compound interest works — it just runs in the wrong direction.

Diversify because you cannot forecast. Not because it maximises returns; it does not. Because it survives being wrong, which is a margin of safety applied to a domain where being wrong is normal.

Distrust confident stories, including your own. Markets generate compelling narratives constantly, and after the fact everything looks like it was obvious. Survivorship bias ensures you mostly hear from people whose confident bets happened to work.

The record-profits example is worth keeping in mind as a corrective. A price that falls on good news is not the market being irrational. It is the market telling you that what you just learned was already known — and that the only thing that ever moved it was the part nobody saw coming.