Prefer to listen? To hear this entire conversation, and more episodes in the weeks ahead, follow Ones and Tooze wherever you get your podcasts.
This year, the winners of the men’s and women’s singles at the U.S. Open tennis tournament in New York will each win $5 million—a 39 percent increase from the prize in 2024, when Jannik Sinner and Aryna Sabalenka each won $3.6 million. But the total amount of money won by players at this year’s U.S. Open, as at most high-level tennis events, represents only a fraction of the money that the sport generates.
How do the origins of tennis as an aristocratic pastime shape the sport today? Why do tennis players earn only a small share of the money? And how can economic theory explain the development of individual tennis matches?
Those are just a few of the questions that came up in my recent conversation with FP economics columnist Adam Tooze on the podcast we co-host, Ones and Tooze. What follows is an excerpt, edited for length and clarity. For the full conversation, look for Ones and Tooze wherever you get your podcasts. And check out Adam’s Substack newsletter.
Cameron Abadi: Tennis is largely identified as an upper-class sport in Europe. Does that have something to do with the history of the sport, specifically the way it traces its origins to aristocratic recreation?
Adam Tooze: I think that’s very fair. The sport’s deepest origins are in the medieval jeu de paume. I think the first thing called tennis was a game now known as real tennis, which is like a hybrid between indoor tennis and squash. And in the odd place like Cambridge or Oxford, there are still courts for real tennis, which is the game that somebody like Henry VIII played in the 16th century.
The game of tennis in its modern form was absolutely a product of late 19th-century trans-Atlantic, haute bourgeois, quasi-aristocratic leisure culture. It was formalized in the current rules in 1873 by the fabulously named Walter Clopton Wingfield. And it was a pursuit for ladies and gentlemen in the summer in their large private gardens, country clubs, and clubs in cities like London and Paris. It required a lawn, time, equipment, and the de rigueur white outfits that Wimbledon still insists on.
Tennis was, from the very beginning, both a game for men and women—but for gentlemen and gentlewomen. So it was part of a kind of bourgeois aristocratic feminism. And it was included in the modern Olympics from the very start, beginning in 1896. That’s significant because the [original] ethos of the Olympics is that of the amateur/gentleman athlete: so not professional athletes but athletes that, for the sake of performance and their love of the sport and the perfection of human physical activity, engage in their pursuits.
This defined the early history of tennis, with a profound division between the original core players—gentlemanly amateurs who had access to the big lawn tennis clubs in the United States, in Paris, in London—and professionals, who in many cases were more adept players, working-class, and played on an exhibition circuit and could earn money doing so.
The Grand Slam circuit of Roland Garros, Wimbledon, the U.S. national championship, and that of Australia was originally exclusively for amateur players without prize money. That logic dominated the game through the late 1960s. What’s called the modern phase of the game started in 1968, the Open Era. They’re “open” because they admitted professional and amateur players.
So the form of the game that we have now is the tortured product of these struggles over a class history that previously separated the Grand Slam tournaments from everyone else precisely on the basis that they didn’t give any prize money. They were like the Olympics. And from ’68 onward, you get the emergence of the modern game with that boundary being erased and professionals being paid prize money who are playing at the peak of the sport.
This is different from the history of soccer, for instance, which was professional from the start. It’s much closer to the history of a game like rugby, which for a long time has had a distinction between rugby union—the professional working-class game, which is played in the north of England and Australia—and rugby league, the game that comes out of private schools and is played in national competitions between the great rugby nations of the world.
CA: Professional tennis players seem to only get a small share of the money that the sport generates. How does tennis compare in that respect with other major sports?
AT: I think it’s structurally much more like golf, where you have professional golfers but there’s a vast bourgeois industry of golf clubs and golf playing that dwarfs the amount of money that the professionals earn. Whereas in soccer, there is an amateur game played by adults, but it doesn’t involve a lot of money, and soccer leagues are vastly capital-intensive and generate stupendous incomes.
But as you say, the really striking thing about tennis as a competitive sport at the professional level is how small the fraction of the purse is that the players get. It’s about 17 percent of the revenue being generated. I’ve seen figures for the Bundesliga [the German soccer league] that players’ salaries are about 50 percent of the revenue being generated; for basketball, I think it’s about 40 percent. This is a remarkable discrepancy.
I think it has to do with the history of tennis—the relatively late stage at which the high-prestige Grand Slam tournaments started paying prize money at all. At Wimbledon in ’68, Rod Laver, the first men’s champion of the Open Era, was paid 2,000 pounds—which even in ’68 money was not a lot. Unlike the American sports leagues, they don’t have union representation. And the balance of power in the tennis labor market is totally unlike that in soccer—where, thanks to court rulings in the European Union, all European soccer players are free agents all the time. That has fundamentally shifted the bargaining power in soccer in Europe.
CA: How can economic theory, or game theory more specifically, explain the development of individual matches of tennis?
AT: One of the things that makes tennis such a genius game is that it’s essentially like a repeated jewel, right? It’s repeated jewel, which then also because of the scoring rules has these key points. It’s very dramatic because you have a moment of winning a point, a tiebreak, a set, a match. And again and again, the narrative reaches a climax. So the game scoring system generates a drama that compounds the inherent natural drama of two players facing off.
Why this is so classically a game theory moment is that game theory attempts to understand the way in which you optimize a decision when the outcome of your decision depends on the reaction of another person—another player who’s also trying to optimize their decision.
The classic model of this is the prisoner’s dilemma: two prisoners separated from each other in cells, having to decide whether they confess or don’t confess. The classic conclusion is that it’s quite likely that both will end up confessing. That may not be optimal because the ideal thing will be for neither of them to confess—but the risk is too high that the other side will confess and you don’t and you therefore take the whole rap. This generates propositions about likely strategies, and the most classic one is min-max, when you try to minimize the maximum payoff that your antagonist can possibly realize from a situation.
This is classically instantiated in the case of a tennis opening, which in the modern game has become more dominant, especially in the men’s game, as the key moment. A tennis player can serve wide and drive the opponent out to the edge and then try to force them to run and not be able to reach the server’s return. You can serve into the body and discombobulate them. Or you can serve directly down the middle, which is often the fastest and gives them very little time to react. How successful each of those options will be depends on whether your antagonist anticipates it.
And you’ll hear in the commentary that when you go to the second serve—which the player has to serve more slowly to make sure that the ball goes in—the receiver will become more aggressive and move toward a position where they can strike the ball back harder. This is a classic game theoretic position. My optimal strategy to go wide, to go to the body, or go to the T is determined by what I think the other person is going to do in positioning themselves for receiving the ball and returning it. Economists have done these incredible behavioral studies where they’ve scanned hundreds of thousands of serves and concluded that in general, the optimal strategy for tennis players approximated very closely to the randomization strategies that min-max under game theoretic conditions would predict.
Basically, what you tend to do is pick the service that’s strongest for you and minimizes the receiver’s chance of returning it—while allowing for the fact that the more often you do that, the more predictable you become, so to a degree you have to randomize. You will see players optimizing over these two elements: the most powerful and most successful serve, as opposed to the most predictable serve. What the studies also showed is that the most successful players approximate most closely to the model’s prediction of the optimal strategy. Because the serve is so crucial in the men’s game now, it’s particularly in men’s games that convergence to the min-max optimal strategy is most predictive of success.
The upshot of all of this is: Yes, game theory appears to rule, and the players by way of a kind of intuitive behavioral approximation to it—or in some cases, through endless video study—actually get pretty close to performing on the basis of what the model will predict.