Thinking Strategically: The Competitive Edge in Business, Politics, and Everyday Life explains how to win at nearly any competition, from poker and product marketing to politics and warfare, using game theory. Published in 1991 by economics professors Avinash K. Dixit of Princeton University and Barry J. Nalebuff of Yale University, the work became an international bestseller. The eBook version of the 1993 re-issue forms the basis for this study guide.

Summary

In life, people often vie to win resources; this applies to everything from poker games to job searches, dating, marketplace competition, elective office, and warfare. In the science of game theory, each of these competitions is considered a “game” for which there are strategies that improve the chances of winning.

Game theory breaks competitions into various types, each with a different strategy for victory. A basketball player with “hot hands” forces the defense to focus on him, freeing up his teammates to make more baskets. A sailboat leading a race copies whatever the second-place boat does and stays in the lead; similarly, large corporations let upstart firms take innovative risks, then copy their successes. Before making commitments, people demand more benefits because they know that, once they’ve signed on, their bargaining power drops. These are examples of the many types of games and how to win them.

Players make their moves either simultaneously or in sequence. Sequential games, like sports or politics, can be diagrammed with a “game tree.” For each possible move, game trees portray the possible follow-up moves as branching lines that contain further branches for the next round of moves, and so on until the game ends. Simultaneous games—for example, voters on Election Day, or businesses that release competing products on the same day—can be charted with tables that list one player’s options as rows and another player’s options as columns; the resulting squares display the results of each combination of options.

In many games, players find they have a “dominant strategy,” an approach that works at least as well or better than any other option. No matter what the other players do, the player’s dominant strategy is the best one to use. Similarly, “dominated” strategies are worse than all other approaches and should be removed from consideration. Some games end up in an “equilibrium,” where each side has one consistent strategy no matter what the other side does. 

A common game in society is the prisoner’s dilemma, where two or more parties agree to do something costly to each but beneficial to all, except the players also are tempted to defect and get a free ride on the others’ sacrifices. If all players think this way, cooperation collapses and the result is worse for everyone than if they hadn’t agreed in the first place. Price fixing is a common example of this type of cooperation: It’s hard to prevent members from cheating.

In games that last indefinitely—business competition, or international diplomacy, as examples—cooperation can be enforced because players will face future punishments if they defect. When a game has a final round, players will defect at the end because they won’t suffer punishment; this realization causes all cooperation to collapse. In games where players always punish negative behavior, any lapse may cause the game to become a feud, a loop of endless punishments.

Players sometimes can control events with “strategic moves,” promises either to do something unilaterally or to punish or reward the others under certain conditions. A candidate might promise to lower taxes no matter what; a country might promise to retaliate if another nation attacks the country’s allies. Strategic moves only work if the promise is believable; opponents are more likely to believe a player who makes a commitment of resources to fulfill the promise.

Several approaches can improve a player’s credibility: These include a good reputation, written contracts, cutting off communication and burning bridges to remove the temptation to renegotiate or retreat, moving in small steps to prevent defaulting, using self-reinforcing teams, and negotiating through an arbitrator.

Players need to randomize their attacks. A tennis player who always serves to an opponent’s backhand will do much worse than if the serves are randomized to both backhand and forehand. It’s possible to figure out the best mix of serves based on the opponent’s strengths and weaknesses; even if the opponent figures out that mix, moment-to-moment randomization of serves makes it impossible to predict the next serve.

One form of strong commitment is brinkmanship, the art of dragging an opponent toward a disastrous outcome if the opponent doesn’t make concessions. During the Cuban Missile Crisis of 1962, US president Kennedy demanded that the Soviet Union remove nuclear missiles it had placed in Cuba, just off the US border; he ordered a blockade of Russian ships nearing Cuba with more missiles. Soviet premier Khrushchev backed down on condition that the US secretly remove its own nuclear missiles from bases in Turkey, close to the Soviet Union. Kennedy’s brinkmanship used the threat of total disaster to scare his opponent. Such tactics always are risky; there’s no way to design such a game without the possibility of sliding accidentally toward complete destruction.

People’s individual choices sometimes combine, in group settings, into results that are worse than no choice. This is due partly to a lack of pricing: Pollution has no price, for example, because it’s not for sale. It’s also due to incentives that work for individuals but not for groups: If everyone is speeding, it’s safer to go along with them but less safe than if everyone was obeying the speed limit. Sometimes people adopt a product standard that’s worse than others—the QWERTY keyboard, or gasoline engines, or light-water nuclear reactors—because of flukes during the product’s development processes. In these situations, outside intervention sometimes can reset the incentives so the group arrives at a better solution.

Voting is a game in which participants try to promote their candidate by placing votes where they do the most good, not always by voting directly for their preferred candidate. No voting system is perfect, and ballots with multiple candidates can generate especially strange results, so that winners sometimes aren’t anyone’s first choice.

Negotiation is an important game, with results dependent on the rules of the interaction and the amount of each side’s needs. For evenly matched opponents, generally the resources in contention get split 50/50, but the less-patient side tends to give in to the more-patient one.

Companies design incentives so that employees focus on doing good work. A software firm, for example, might offer computer programmers a bonus if their work produces applications that sell well. Partnerships, meanwhile, must be designed so that one side doesn’t get a free ride or quit suddenly. Firms should put up a large sum toward the project and suffer financial penalties if they back out; each partner should profit in proportion to its costs, though this can cause them to pad their expenses.

Every chapter in Thinking Strategically contains a Case Study—a real-world example of the chapter’s principles—with a discussion on the best solutions. Chapter 13 contains 23 Case Studies that cover these and other game theory situations.

The book is easy to read for the non-professional; its style is enthusiastic and sometimes humorous. The work includes many charts, diagrams, and tables that clarify the content, along with 36 pages of notes.