Bayesian Analysis

A Comparison of Truncated and Time-Weighted Plackett–Luce Models for Probabilistic Forecasting of Formula One Results

Daniel A. Henderson and Liam J. Kirrane

Advance publication

This article is in its final form and can be cited using the date of online publication and the DOI.

Full-text: Open access

Abstract

We compare several variants of the Plackett–Luce model, a commonly-used model for permutations, in terms of their ability to accurately forecast Formula One motor racing results. A Bayesian approach to forecasting is adopted and a Gibbs sampler for sampling from the posterior distributions of the model parameters is described. Prediction of the results from the 2010 to 2013 Formula One seasons highlights clear strengths and weaknesses of the various models. We demonstrate by example that down weighting past results can improve forecasts, and that some of the models we consider are competitive with the forecasts implied by bookmakers odds.

Article information

Source
Bayesian Anal. (2017), 24 pages.

Dates
First available in Project Euclid: 28 February 2017

Permanent link to this document
https://projecteuclid.org/euclid.ba/1488250819

Digital Object Identifier
doi:10.1214/17-BA1048

Keywords
Bayesian inference Gibbs sampling latent variable models permutations ranks sport

Rights
Creative Commons Attribution 4.0 International License.

Citation

Henderson, Daniel A.; Kirrane, Liam J. A Comparison of Truncated and Time-Weighted Plackett–Luce Models for Probabilistic Forecasting of Formula One Results. Bayesian Anal., advance publication, 28 February 2017. doi:10.1214/17-BA1048. https://projecteuclid.org/euclid.ba/1488250819


Export citation

References

  • Bell, A., Smith, J., Sabel, C. E., and Jones, K. (2016). “Formula for success: multilevel modelling of Formula One driver and constructor performance, 1950–2014.”Journal of Quantitative Analysis in Sports, 12: 99–112.
  • Benter, W. (1994). “Computer-based horse race handicapping and wagering systems: a report.” In Hausch, D. B., Lo, V. S. Y., and Ziemba, W. T. (eds.),Efficiency of Racetrack Betting Markets, 183–198. London: Academic Press.
  • Bernardo, J. M. and Smith, A. F. M. (1994).Bayesian Theory. Chichester: Wiley.
  • Caron, F. and Doucet, A. (2012). “Efficient Bayesian inference for generalized Bradley-Terry models.”Journal of Computational and Graphical Statistics, 21: 174–196.
  • Chib, S. (1995). “Marginal likelihood from the Gibbs output.”Journal of the American Statistical Association, 90: 1313–1321.
  • Dawid, A. P. and Musio, M. (2014). “Theory and applications of proper scoring rules.”Metron, 72: 169–183.
  • Diaconis, P. (1988).Group Representations in Probability and Statistics, volume 11 ofIMS Lecture Notes. Institute of Mathematical Statistics.
  • Dixon, M. J. and Coles, S. G. (1997). “Modelling association football scores and inefficiencies in the football betting market.”Applied Statistics, 46: 265–280.
  • Eichenberger, R. and Stadelmann, D. (2009). “Who is the best Formula One driver? An economic approach to evaluating driver talent.”Economic Analysis & Policy, 39: 389–406.
  • Friel, N. and Pettitt, A. N. (2008). “Marginal likelihood estimation via power posteriors.”Journal of the Royal Statistical Society, Series B, 70: 589–607.
  • Glickman, M. E. and Hennessy, J. (2015). “A stochastic rank ordered logit model for rating multi-competitor games and sports.”Journal of Quantitative Analysis in Sports, 11: 131–144.
  • Gneiting, T. and Raftery, A. E. (2007). “Strictly proper scoring rules, prediction, and estimation.”Journal of the American Statistical Association, 102: 359–378.
  • Graves, T., Reese, C. S., and Fitzgerald, M. (2003). “Hierarchical models for permutations: analysis of auto racing results.”Journal of the American Statistical Association, 98: 282–291.
  • Guiver, J. and Snelson, E. (2009). “Bayesian inference for Plackett–Luce ranking models.” Proceedings of the 26th Annual International Conference on Machine Learning.
  • Harville, D. A. (1973). “Assigning probabilities to the outcomes of multi-entry competitions.”Journal of the American Statistical Association, 68: 312–316.
  • Henderson, D. A. and Kirrane, L. J. (2017). “Supplementary Material for “A comparison of truncated and time-weighted Plackett–Luce models for probabilistic forecasting of Formula One results”.” \em Bayesian Analysis.
  • Hoeting, J. A., Madigan, D., Raftery, A. E., and Volinsky, C. T. (1999). “Bayesian model averaging: a tutorial.”Statistical Science, 14: 382–417.
  • Hunter, D. R. (2004). “MM algorithms for generalized Bradley–Terry models.”Annals of Statistics, 32: 384–406.
  • Ibrahim, J. G. and Chen, M.-H. (2000). “Power prior distributions for regression models.”Statistical Science, 15: 46–60.
  • Kass, R. E. and Raftery, A. E. (1995). “Bayes Factors.”Journal of the American Statistical Association, 90: 773–795.
  • Luce, R. D. (1959).Individual Choice Behavior. New York: Wiley.
  • Marden, J. I. (1995).Analysing and Modeling Rank Data. London: Chapman and Hall.
  • McCarthy, D. and Jensen, S. T. (2016). “Power-weighted densities for time series data.”Annals of Applied Statistics, 10: 305–334.
  • Pettitt, A. N. (1982). “Inference for the linear model using a likelihood based on ranks.”Journal of the Royal Statistical Society, Series B, 44: 234–243.
  • Phillips, A. J. K. (2014). “Uncovering Formula One driver performances from 1950 to 2013by adjusting for team and competition effects.”Journal of Quantitative Analysis in Sports, 10: 261–278.
  • Plackett, R. L. (1975). “The analysis of permutations.”Applied Statistics, 24: 193–202.
  • Silverberg, A. R. (1980). “Statistical models for $q$-permutations.” Ph.D. thesis, Department of Statistics, Princeton University.
  • Smith, R. (2013).Formula 1: All the Races. The World Championship Story Race-by-Race 1950–2012. Yeovil, Somerset: Haynes Publishing, second edition.
  • Yellott, J. I. (1977). “The relationship between Luce’s choice axiom, Thurstone’s theory of comparative judgement, and the double exponential distribution.”Journal of Mathematical Psychology, 15: 109–144.

Supplemental materials

  • Supplementary material: Supplementary Material for “A comparison of truncated and time-weighted Plackett–Luce models for probabilistic forecasting of Formula One results”. The Supplementary material contains further details on the Gibbs sampling algorithm of Section 3, details of predictive simulations, an analysis of sensitivity of predictions to prior assumptions and an investigation into optimal choices of the time weighting parameter.