Rules Create Unequal Rewards: Elite Tennis Players Allocate Resources Efficiently

TL;DR

The paper investigates how elite tennis players allocate effort under a threshold-based scoring system. By modeling score-dependent point-winning probabilities across $18$ score states and deriving each player's Pareto frontier, the authors quantify efficiency as proximity to the frontier and show that top players convert points to game wins more efficiently. The findings demonstrate that higher-tier players follow optimal allocation patterns—amplifying effort at neutral or leading scores while reducing it when trailing—especially in return games. This efficiency-based perspective highlights a core aspect of expertise: effective adaptation to rule-imposed value structures with potential implications for other domains featuring unequal marginal rewards.

Abstract

In many competitive settings, from education to politics, rules do not reward effort evenly, and thresholds (e.g., grade cutoffs or electoral majorities) make some moments disproportionately important. Success thus depends on efficiently allocating limited resources. However, empirical demonstration has been difficult because effort allocation is rarely observable and feedback is often delayed, limiting our understanding of expertise. Professional tennis provides an ideal natural experiment. Because each game resets after a player wins four points and points in a lost game are wasted, the value of a point varies sharply across scores. Efficient allocation should therefore win games without wasting points, conserving resources for future games. Such allocation manifests in score-dependent point-winning probabilities, from which we derive each player's Pareto frontier-the theoretical limit of the trade-off between game-winning probability and the expected points per game. Here, we show that top players operate closer to this frontier, converting points to game wins more efficiently. Optimal strategies reduce the probability of winning points when the player is far behind (e.g.,0-2, 0-3). This behavior is psychologically difficult-letting go of the current game-but represents a rational energy conservation strategy. Top players exhibit this pattern especially in return games, where winning points is harder than in service games, requiring them to drastically vary their efforts, consistent with game-theoretic predictions. These findings suggest that elite performance reflects efficient adaptation to rule-created value structures; knowing when to give up may be as fundamental to expertise as knowing when to compete.

Figures (3)

• Figure 1: Same ability, different outcomes.(A) Two players with the same average point-winning probability (39%) but different results. This example from women's return games illustrates how the efficient player achieves a higher game-winning probability with fewer expected points per game compared to the inefficient player or a constant-strategy baseline (triangle; fixed 39% at all scores). (B) Score-dependent allocation. Scores reflect the players' position within each game (e.g., 2--0 means the player has 2 points, opponent has 0). The efficient player shows higher probabilities at leading or neutral scores (e.g., 45% at 2--0, 50% at 2--1) and lower probabilities at trailing scores (e.g., 29% at 0--3), whereas the inefficient player shows the opposite pattern. Tennis games require winning at least four points with a two-point margin; at 3--3 (deuce), play enters the advantage (AD) cycle. Tennis scoring (0, 15, 30, 40, AD) presented using simplified notation (0, 1, 2, 3, AD) for clarity.
• Figure 2: Elite players convert points into game wins more efficiently.(A) Quantifying efficiency. Using an example player, the efficiency score quantifies proximity to the theoretical Pareto frontier (yellow curve) in the normalized outcome space. The score is derived from the shortest Euclidean distance between the observed performance and the frontier scaled to a 0--1 range, where 1 represents the theoretical optimum (orange diamond). (B) High-tier players show higher efficiency scores. Efficiency scores are compared for service and return games across three match-winning tiers: low ($<50\%$; men $n=37$, women $n=34$), mid (50--70%; men $n=31$, women $n=47$), and high ($>70\%$; men $n=16$, women $n=13$). The white circles indicate medians, while the error bars show 95% CIs. The horizontal bars denote Bonferroni-corrected Mann--Whitney comparisons ($*$$p<0.05$, $**$$p<0.01$, $***$$p<0.001$, n.s.: not significant) with effect sizes (Cliff’s $\delta$).
• Figure 3: Optimal allocation patterns and strategy fit.(A) Lower baseline ability requires higher strategic contrast. A strong negative correlation ($r = -0.90$) links lower average point-winning probabilities to higher optimal contrast, indicating that weaker positions (e.g., return games) demand more extreme strategies to maximize outcomes. Points represent individual players (men $n=84$, women $n=94$); diamonds denote category means; the black line shows linear regression with 95% CIs. Upper state diagrams illustrate optimal patterns; arrows indicate probability deviation from the category mean. (B) High-tier players follow the optimal pattern. Strategy fit---quantifying the proximity of actual allocation to theoretical optima in 18-dimensional space--- is compared across the low ($<50\%$), middle ($50$–$70\%$), and high ($>70\%$) match-winning tiers. Dots represent players; white circles show tier means with 95% CIs. Horizontal bars indicate pairwise comparisons with Cohen’s $d$ (* $p<0.05$, ** $p<0.01$, *** $p<0.001$, n.s.: not significant).