CoChat

Comparing two methods for testing the efficiency of sports betting markets

Open access
Hegarty, Tadgh, Whelan, Karl · 2024 · Sports Economics Review

The DOI matches its Crossref record; 41 of the 41 references it cites were found in a registry or an index; none of the 5 sources its claims cite is checked for support yet.

Overview

CoChat Summary

TL;DR

The paper compares normalized-probability and inverse-odds regressions for testing strong-form efficiency in sports betting markets, showing theoretically and empirically that the inverse-odds method is biased against detecting favorite-longshot bias. Using soccer and tennis data plus simulations, it finds normalized probabilities correctly reject strong-form efficiency while inverse odds often fail to do so.

Key findings

  • In soccer and tennis data, average payouts decline as decimal odds rise (favorite-longshot bias); all deciles average below one, with the lowest 1% of odds averaging 0.997 for soccer and 0.985 for tennis, so this does not provide evidence against weak-form efficiency.
  • Normalized-probability regressions reject strong-form efficiency: soccer γ = 0.075* (SE 0.007), F = 106.5*; tennis γ = 0.066* (SE 0.006), F = 186.61*. Inverse-odds regressions do not reject for soccer (γ = 0.009, SE 0.007; F = 1.85) and produce a weaker rejection for tennis (γ = 0.0166, SE 0.006; F = 8.39).
  • Predicted payouts from the normalized-probability model: soccer 343 of 252,690 bets (0.14%) have predicted payouts above one, and only 14 (0.01%) have 95% confidence intervals entirely above one; tennis 94 of 111,976 bets (0.08%) have predicted payouts above one, with none having 95% confidence intervals entirely above one. The inverse-odds model produced no predicted payouts above one.
  • Simulations (10,000 per case): under favorite-longshot bias, normalized probabilities rejected the null in all simulations, while inverse odds rejected 13% of the time for soccer and 47% for tennis. Under efficiency, inverse odds nearly always rejected the true null, while normalized probabilities had median t-statistics near zero.

Methods

  • Theoretical derivation of regression-based tests under strong-form efficiency: normalized probabilities equal bookmaker probabilities under the null, while inverse odds are a multiple greater than one of true probabilities, causing downward bias in the estimated coefficient.
  • Data: 84,230 European professional soccer matches across 22 leagues from 2011/12 to 2021/22 (252,690 bets) and 55,988 ATP/WTA tennis matches from 2011 to 2022 (111,976 bets), using average closing odds; 16 matches were excluded because reported odds had an overround below 1.
  • Estimation: weighted least squares with variances approximated by P(1−P), clustered standard errors at the match level; regressions of outcomes on normalized probabilities or inverse odds; simulations calibrated to the soccer and tennis data, with 10,000 simulated datasets per type.

Limitations

  • Stated: linear probability models can be objected to because they can predict probabilities outside [0,1], though the null implies linearity and the paper reports similar significance results with Logit/Probit alternatives; average rather than maximum odds are used; 16 matches were excluded as likely mis-reported.
  • Stated: the paper focuses on strong-form efficiency for online fixed-odds betting and does not test weak-form profitability; it notes that an even stronger cross-event efficiency does not hold and is not tested.
  • Evident from the text: evidence is limited to soccer and tennis fixed-odds markets and simulations calibrated to those datasets, so generalizability to other sports, bet types, or market structures is not established.

Model-written summary. Check it against the paper before relying on it.

Integrity signals

Registry

Crossref · 10.1016/j.serev.2024.100042

Cites

41 of 41 references verified0 citation errors · 0 not checkable

Statistics

Nothing to recompute.

Verification

Claims the paper makes

0 Valid · 0 Not valid · 4 Not checked · 35 Own results and interpretations

15 more claims

Create a free account to see all 39 claims and every source verdict.

See all claims — free account
  • It may state two separate claims as one. 18
  • Probably not a substantial claim. 2

Create a free account to see these claims.

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

Each contradiction between two passages listed here was confirmed by an independent check.

2 of 2

Citations

Cites

Show all 41 references

Research works printed but not linked: 2

  1. Cain, 2001, Vol. 69, 197
  2. Davies, 2022, Revealed: How bookies clamp down on successful gamblers and exploit the rest, The Guardian

Cited by

See all 5

Metadata

Authors
Hegarty, T., & Whelan, K.
Affiliations
University College Dublin (IE)
Corresponding author
Whelan, Karl
Venue
Sports Economics Review
Publisher
Elsevier BV
Published
2024-12
Volume
8
Pages
100042
ISSN
2773-1618 (ISSN-L)
OpenAlex
W4402679764
Cited by
5
References
38
Citations by year
2026: 4 · 2025: 1
FWCI
1.15
Type
journal article
Language
en
Open access
Locations
  • Sports Economics Review (publishedVersion)
  • Munich Personal RePEc Archive (Ludwig Maximilian University of Munich) (acceptedVersion)
  • RePEc: Research Papers in Economics (submittedVersion)
  • Econstor (Econstor) (submittedVersion)
Topics
  • Social Sciences › Economics, Econometrics and Finance › Economics and Econometrics › Sports Analytics and Performance
  • Social Sciences › Economics, Econometrics and Finance › Finance › Financial Markets and Investment Strategies
  • Social Sciences › Decision Sciences › General Decision Sciences › Decision-Making and Behavioral Economics
Keywords
sports betting markets, market efficiency, strong efficiency, betting odds, favorite-longshot bias, tennis, soccer

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

A language model drew the claims from the paper on 2026-11-19, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).

A second model labelled each claim on 2026-11-19 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review

How to cite

Cite CoChat for the checks on this page, and the paper for its own findings.

This page
CoChat. (2026). Citation check: Comparing two methods for testing the efficiency of sports betting markets. https://app.cochat.ai/papers/10.1016/j.serev.2024.100042
The paper
Hegarty, T., & Whelan, K. (2024). Comparing two methods for testing the efficiency of sports betting markets. Sports Economics Review. https://doi.org/10.1016/j.serev.2024.100042

Data sources

Bibliographic data: Semantic Scholar, OpenAlex, Crossref.

Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).

Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.