nep-spo New Economics Papers
on Sports and Economics
Issue of 2026–07–27
five papers chosen by
Humberto Barreto, DePauw University


  1. Agreeing to Disagree: The Economics of Betting Exchanges By Whelan, Karl
  2. Regulatory Arbitrage in the Market for Player Rights: Evidence from European Soccer By Ciliberto, Federico; Scotti, Davide; Vismara, Silvio
  3. LLM Agents as Static Level-k Players in Behavioural Games By Po Han Teo
  4. Decomposing First-Mover Advantage: A Decision-Making Perspective By Zegners, Dainis; Bhaskarabhatla, Ajay; Kretschmer, Tobias
  5. Comparison games and ranking of players By Daniela Bubboloni; Stefano Moretti

  1. By: Whelan, Karl
    Abstract: Betting exchanges match people to take opposite sides of a bet. We present a model of a betting exchange in which participants disagree about outcome probabilities but are, on average, correct. Traders maximize subjective expected profits and equilibrium emerges from a simple matching process. The model predicts those who post quotes (Makers) will earn higher returns than those who accept them (Takers) and that loss rates for Takers will rise as the probability of their accepted bet winning falls. Using a large sample of bets on soccer from Betfair Exchange, we implement a transaction-level empirical strategy that identifies Maker and Taker sides of each trade. We show that pre-match and early in-play behavior aligns closely with the model’s predictions. However, as matches progress, behavior shifts: longshot bets generate large, systematic losses even for liquidity providers, and profits emerge for those who accept offers on favorites
    JEL: G14 G41
    Date: 2025–09
    URL: https://d.repec.org/n?u=RePEc:cpr:ceprdp:20633
  2. By: Ciliberto, Federico; Scotti, Davide; Vismara, Silvio
    Abstract: Firms facing accounting-based regulation can restructure transactions to meet formal constraints without reducing risk. We study this arbitrage in European soccer, where Financial Fair Play (FFP) rewards inflated capital gains from bilateral player exchanges (“cross-transfers†). Using transaction-level data from 20 European leagues (2015–2023), we identify 412 cross-transfers generating €1.37–€2.94 million in excess capital gains per leg, roughly €0.6–1.2 billion in aggregate. The premium is absent before FFP, and gains are positively correlated within pairs, consistent with coordinated inflation rather than liquidity trading. Arbitrage peaks at 8.5% of reported capital gains, declining after regulators adopted transaction-level scrutiny.
    JEL: G28 L51 Z20
    Date: 2025–08
    URL: https://d.repec.org/n?u=RePEc:cpr:ceprdp:20534
  3. By: Po Han Teo
    Abstract: Large Language Models (LLMs) are increasingly used as stand-ins in behavioural games. These stand-ins rely on the assumption that the LLM's distribution of choices meaningfully matches how humans play the same game. This study tests that assumption through two games. The first is a p-beauty contest, and the second one is a public goods game. The study first investigates five local-model settings within the same model family. These settings are varied together in a 360-cell factorial, which balances temperature, scale (0.5-32B), quantisation, instruct vs base, and framing. Each cell's distribution is then compared against whole choice distributions in published human data. Each deployment setting, except for quantisation, governs a different aspect of fidelity. Mechanically, while the dispersion of human players can be somewhat recovered through deployment settings, the strategic process behind it cannot. Through the lens of the level-k cognitive theory, we find that LLMs act as static, category-retrieved level-k players, where k is set by the model scale. The models also do not run within-game belief-updating or backward induction throughout multiple-round horizon settings. While human contributions decayed in the public goods game, LLMs stayed flat or rose at every scale. When the horizon test was administered, LLMs were more cooperative under an indefinite horizon compared to a finite one. However, LLMs ignore their relative round position, so no last-round defection was displayed. This implies that LLMs retrieved levels relative to the horizon category rather than working out iteratively from the specific game setting.
    Date: 2026–06
    URL: https://d.repec.org/n?u=RePEc:arx:papers:2606.27845
  4. By: Zegners, Dainis; Bhaskarabhatla, Ajay; Kretschmer, Tobias
    Abstract: We develop a theory of first-mover advantage (FMA) grounded in decision-making ability and error propensity. Our simulation model demonstrates how small initial advantages compound through systematic differences in error rates between first- and second-movers. We show that FMA amplifies when both competitors have higher decision-making ability and when abilities are closely matched, whereas FMA increases for low complexity of the decision-making environment but decreases for higher complexity. To test these predictions, we analyze 3.66 million chess games, exploiting quasi-random assignment of move order and objective ability measures. Results strongly support our theory: FMA increases from 3.5% for low-ability players to 11% for high-ability players, peaks when competitors are evenly matched, and strengthens in complex decision environments. Move-level analysis reveals second-movers make larger errors when operating from disadvantageous positions under complexity, explaining FMA persistence through compounding mistakes rather than resource advantages.
    Date: 2025–07
    URL: https://d.repec.org/n?u=RePEc:cpr:ceprdp:20448
  5. By: Daniela Bubboloni; Stefano Moretti
    Abstract: This work addresses the problem of assessing player importance in coalitional settings where the available information concerns the relative strength between pairs of coalitions, rather than the absolute worth of each coalition. We introduce a novel framework that is flexible enough to represent all coalitional pseudo-games and, through the use of coalitional networks, naturally accommodates scenarios with limited or heterogeneous coalition comparisons. Importantly, this framework still enables the computation of semivalues of pseudo-games, such as the Banzhaf and Shapley values, that can be expressed as weighted sums of differences in specific coalition comparisons, thus offering interpretations beyond traditional approaches. Furthermore, for ranking players rather than computing exact numerical attributions, we introduce the concept of a player's score, which simplifies the process of determining rankings based on semivalues, and shifts the perspective from average marginal contribution to average coalitional worth. This turns out to be particularly enlightening for the Banzhaf value.
    Date: 2026–06
    URL: https://d.repec.org/n?u=RePEc:arx:papers:2606.31955

This nep-spo issue is ©2026 by Humberto Barreto. It is provided as is without any express or implied warranty. It may be freely redistributed in whole or in part for any purpose. If distributed in part, please include this notice.
General information on the NEP project can be found at https://nep.repec.org. For comments please write to the director of NEP, Marco Novarese at <director@nep.repec.org>. Put “NEP” in the subject, otherwise your mail may be rejected.
NEP’s infrastructure is sponsored by the Griffith Business School of Griffith University in Australia.