Herding

Herding is the tendency to follow others’ decisions because the crowd offers a form of safety, even when independent analysis suggests a better course. The book traces this pattern across professional football and investment management: imitation can make an established practice appear self-evidently correct, while reputational pressure makes departure from convention costly. In both settings, people may choose what looks defensible to peers over what their own evidence suggests, narrowing the range of decisions and information available to the group.

Convention and Imitation

In professional football, the book uses fourth-down decisions to show how a shared convention can reproduce itself. Teams often punt in situations where analysis indicates that they would be better off trying to advance the ball. The conservative choice has the standing of a fundamental truth in the sport; without compelling evidence to challenge it, coaches can more readily explain why the existing practice makes sense than imagine an alternative. If no one else goes for it, that absence itself seems to prove that going for it is unwise. Imitation thus supplies its own apparent justification.

The pattern is reinforced by football’s clubby, insular decision-making culture. Although the sport has had genuine innovators, its approach to statistical analysis is described as unusually resistant to change. A relatively undiverse pool of decision-makers is less likely to develop radical alternatives and less likely to embrace them when proposed. The resulting errors are correlated: coaches tend to make mistakes in the same direction, rather than bringing different judgments to bear. This is not simply a matter of one coach overlooking an option; shared assumptions can keep an entire professional community from testing it.

The book contrasts that convention with the Oakland A’s, where Billy Beane and Paul DePodesta rejected social proof and cultivated different approaches to evaluating and developing players. Their ability to build a successful team on little money is presented as a consequence of refusing to treat accepted strategic wisdom as decisive. In football, Bill Belichick is identified as the coach who appeared to take David Romer’s fourth-down analysis seriously, perhaps applying it in games; his willingness to reject conventional wisdom accompanied the New England Patriots’ success in winning two Super Bowls in three years. These examples frame independence not as novelty for its own sake, but as a willingness to test whether the crowd’s settled practice is actually sound.

Risk and Reputation

The incentive to conform is not only intellectual. Going for it on fourth-and-two may make strategic sense while remaining psychologically difficult: under the strategy discussed, a team would fail to score roughly half the time it was inside its opponent’s ten-yard line. The long-run logic can favor the riskier choice, but a failure is immediate and visible. Punting may make little strategic sense, yet it limits the scale of a possible public setback. A cautious decision can therefore feel safer even when it is less advantageous over time.

That difference matters especially in a community where peers follow the same practice and failure cannot be hidden. A coach who adopts a new strategy risks failing conspicuously; one who follows the crowd can fail in a manner that seems ordinary and professionally defensible. The book calls this pattern herding: like water buffalo gathering in the face of a lion, decision-makers find security in numbers. The slogan “No one ever got fired for buying IBM” captures the same reputational logic beyond sport. The point is not that people are forced to imitate, but that the professional costs of visible deviation can make conformity seem prudent.

Money Managers and the Appearance of Reason

Mutual-fund managers illustrate how herding can persist even among people whose apparent job is to outperform their competitors. David S. Scharfstein and Jeremy C. Stein’s study considered managers who follow similar strategies and gather into the same stocks. Such imitation seems self-defeating: if managers are paid to beat the market, copying one another should make a difficult task harder. The book explains the behavior by distinguishing the work managers must do from the evidence investors can observe. Managers must invest wisely, but they must also persuade clients that they are doing so.

Investors cannot easily tell whether a manager’s choices are genuinely wise. Short-term performance is an imperfect measure of skill: results in a quarter can depend on factors unrelated to stock-picking or asset allocation. Investors therefore use comparison with peers as additional evidence. A manager whose choices resemble other managers’ choices appears at least not irrational; one who departs from prevailing practice can look foolish, even when the contrarian decision is better grounded. The benchmark of professional reasonableness can thus become conformity itself.

This pressure is strengthened by investors’ limited patience. If they could wait indefinitely, the numerical difference between good and bad strategies might eventually become clear. But managers can perform badly for significant stretches even when they are highly capable, and clients may not wait long enough to distinguish skill from short-term outcomes. Protecting a job can therefore mean following the strategy that seems rational to investors rather than the strategy that is rational in substance. Managers mimic competitors because the appearance of sound judgment is easier to defend than an unconventional decision whose benefits have not yet appeared.

What Imitation Costs

When managers imitate one another, they sacrifice the very information advantage that might have distinguished their decisions. They are no longer trading on independent information, but relying on what others have done. The effects spread beyond an individual portfolio: imitation narrows the range of investments and reduces the market’s overall intelligence, because conforming managers bring less new information into it. Herding can therefore make each participant’s choices look safer while leaving the group less informed.

The concept carries a tension. Following others is not always foolish: information is dispersed among many people, and attention to their actions can improve a decision that would otherwise rely on only one person’s private knowledge. But when observation turns into uncritical imitation, the crowd’s shared behavior can suppress the distinct information each member might have contributed. The book’s account locates herding in that gap between learning from others and surrendering independent judgment. Its force comes from the fact that conformity can be individually defensible—protecting a coach or manager from conspicuous failure—while collectively weakening innovation, diversity of judgment, and the quality of decisions.