The comparative accuracy of judgmental and model forecasts of American football games

Song, Chi-Ung; Boulier, Bryan L.; Stekler, H. O.

doi:10.1016/j.ijforecast.2007.05.003

Cited by 40 publications

(28 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…, 31, as derived from the adjusted and transformed bookmakers odds, we want to obtain measures for consensus and (dis)agreement among the bookmakers. An intuitive and straightfoward strategy would be to compute this information on a logit scale, employing team-wise means for the consensus and team-wise standard deviations for the disagreement across bookmakers (as suggested by, e.g., Song et al, 2009Song et al, , 2007Zarnowitz and Lambros, 1987). For our application this simple strategy might be appropriate because we could expect that the teams are sufficiently different and the bookmakers have rather similar models and information about the teams.…”

Section: Model Classmentioning

confidence: 99%

“…They were successfully used to predict the outcome of single games (e.g., Spann and Skiera, 2009;Song et al, 2007;Forrest et al, 2005;Dixon and Pope, 2004;Boulier and Stekler, 2003). Based on these ideas, Leitner et al (2008) and Leitner et al (2009) To aggregate information of different forecasters a measure of "consensus" is needed.…”

Section: Introductionmentioning

confidence: 99%

“…To measure "uncertainty" or "disagreement" the standard deviations of the predictive probability distributions could be used (e.g, Clements, 2008;Zarnowitz and Lambros, 1987;Lahiri and Teigland, 1987). These strategies are applied to sports competitions by Song et al (2009Song et al ( , 2007. Alternative strategies for the aggregation of forecasts are discussed by Kolb and Stekler (1996); Schnader and Stekler (1991).…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Bookmaker consensus and agreement for the UEFA Champions League 2008/2009

Leitner

Zeileis

Hornik

2010

IMA Journal of Management Mathematics

View full text Add to dashboard Cite

Bookmakers odds are an easily available source of "prospective" information that is thus often employed for forecasting the outcome of sports events. To investigate the statistical properties of bookmakers odds from a variety of bookmakers for a number of different potential outcomes of a sports event, a class of mixed-effects models is explored, providing information about both consensus and (dis)agreement across bookmakers. In an empirical study for the UEFA Champions League, the most prestigious football club competition in Europe, model selection yields a simple and intuitive model with team-specific means for capturing consensus and team-specific standard deviations reflecting agreement across bookmakers. The resulting consensus forecast performs well in practice, exhibiting high correlation with the actual tournament outcome. Furthermore, the teams' agreement can be shown to be strongly correlated with the predicted consensus and can thus be incorporated in a more parsimonious model for agreement while preserving the same consensus fit.

show abstract

Section: Model Classmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Bookmaker consensus and agreement for the UEFA Champions League 2008/2009

Leitner

Zeileis

Hornik

2010

IMA Journal of Management Mathematics

View full text Add to dashboard Cite

show abstract

“…Point spreads are in fact so accurate that they have been shown to have greater predictive power than sports experts and economists' models (Song et al, 2007), and it is somewhat rare to find any way to systematically exploit them for a profit (Woodland & Woodland 1994;Woodland & Woodland 2001. ).…”

Section: Background: the Nba Betting Market And Betting Datamentioning

confidence: 99%

Racial Bias in the NBA: Implications in Betting Markets

Larsen¹,

Price²,

Wolfers³

2008

Journal of Quantitative Analysis in Sports

View full text Add to dashboard Cite

show abstract

“…In this paper we use that coefficient to evaluate the levels of agreement among the forecasts of 74 experts and 31 statistical systems for outcomes of National Football League regular season games played during the 2000 and 2001 seasons. This data set is the same used earlier to analyze the predictive accuracy of these experts and systems (Song, et al, 2007). The experts and systems made two types of binary forecasts.…”

Section: Introductionmentioning

confidence: 99%