Volume 9, Issue 2

This is the first paper to investigate factors that affect the size of the TV audience for the NCAA Men’s Final Four. The model is based on Nielsen data for 54 markets for 10 years. One of the most interesting results is that college basketball teams have a measurable effect in their local markets, but even the biggest name programs do not have a national effect. However, the little known teams that succeed in the tournament, known as Cinderellas, have a national effect likely due to the media attention and success on the court. Broadcasters and advertisers are interested in maximizing the TV audience, and the model allows prediction to compare games between big market teams, big name teams, David vs. Goliath games, and a Championship game between two Cinderella teams.

Golf is a global sport with professional golfers playing on many organized tours throughout the world. The largest and most important tours for male professionals include the PGA Tour, European Tour, Japanese Tour and Asian Tour. The Official World Golf Ranking, or OWGR, is a system for ranking male professional golfers on a single scale. We say a ranking system is unbiased if otherwise identical golfers who happen to play on different tours have the same (or very similar) ranks. In this paper, we investigate whether the OWGR system is biased for or against any of the tours, and if so, by how much. To investigate any potential bias, we compare the OWGR system with two unbiased methods for estimating golfer skill and performance. The first is a score-based skill estimation (SBSE) method, which uses scoring data to estimate golfer skill, taking into account the relative difficulty of the course in each tournament round. The second is the Sagarin method, which uses win-lose-tie and scoring differential results for golfers playing in the same tournaments, to rank golfers. Neither the score-based skill method nor the Sagarin method use tour information in calculating player ranks, and therefore neither method is biased for or against any tour. Using data from 2002 to 2010 and comparing the results ranks from the OWGR and score-based methods, we find that PGA Tour golfers are penalized by an average of 26–37 OWGR ranking positions compared to non-PGA Tour golfers. Qualitatively similar results are found when comparing OWGR and Sagarin ranks. In all cases, the bias is large and statistically significant. We find a persistent bias through time and also find that the bias tends to be the largest for golfers with SBSE ranks between 40 and 120.

The team batting second in a day-night cricket match faces different playing conditions to the team batting first. This paper quantifies the effect on the runscoring ability of the team batting second by using the difference in differences estimator. This approach indicates that batting during the evening reduces the expected number of runs scored per over by 0.2. The effect explains around 12% of the total margin of loss for teams in the treatment group who lose the match. It is also found that batting during the evening increases the expected number of wickets lost at any given point in the innings.

There is much interest in studying the hot hand behavior of individuals and teams in sports. We collect success/failure data for each of the hitters in a particular baseball season, focusing on the spacings or number of at-bats between consecutive successes. Consistent and streaky models are defined on the basis of the underlying geometric probabilities and a Bayes factor statistic is developed to compare the two models. By looking at the ensemble of test statistics over all players, we see if the patterns of streakiness differ from what one would predict from a consistent hitting model. An interesting observation from this study is the pattern of streakiness depends on how one defines a success in a plate appearance.

In this paper we apply survival techniques to soccer data, treating a goal scoring as the event of interest. It specifically concerns the relationship between the time of the first goal in the game and the time of the second goal. In order to do so, the relevant survival analysis concepts are readjusted to fit the problem and a Cox model is developed for the hazard function. Time dependent covariates and a frailty term are also considered. We also use a reliable propensity score to summarize the pre-game covariates. The conclusions derived from the results are that a first goal occurrence could either expedite or impede the next goal scoring, depending on the time it was scored. Moreover, once a goal is scored, another goal scoring becomes more and more likely as the game progresses. Furthermore, the first goal effect is the same whether the goal was scored or conceded.

The Colley Matrix Rating System (CMRS) has been a component of the Bowl Championship Series (BCS) rankings. As such it is important that the assumptions that lead to and differentiate it from other rating systems be well understood. This work investigates an approach to justifying the CMRS based on an approach taken by Massey. This approach leads to both the CMRS and a version of a Massey Matrix Rating System. When approached in the manner proposed here these rating systems can be presented as special cases of one more general rating system.

In this paper, we empirically evaluate the predictive power of eight sports ranking methods. For each ranking method, we implement two versions, one using only win-loss data and one utilizing score-differential data. The methods are compared on 4 datasets: 32 National Basketball Association (NBA) seasons, 112 Major League Baseball (MLB) seasons, 22 NCAA Division 1-A Basketball (NCAAB) seasons, and 56 NCAA Division 1-A Football (NCAAF) seasons. For each season of each dataset, we apply 20-fold cross validation to determine the predictive accuracy of the ranking methods. The non-parametric Friedman hypothesis test is used to assess whether the predictive errors for the considered rankings over the seasons are statistically dissimilar. The post-hoc Nemenyi test is then employed to determine which ranking methods have significant differences in predictive power. For all datasets, the null hypothesis – that all ranking methods are equivalent – is rejected at the 99% confidence level. For NCAAF and NCAAB datasets, the Nemenyi test concludes that the implementations utilizing score-differential data are usually more predictive than those using only win-loss data. For the NCAAF dataset, the least squares and random walker methods have significantly better predictive accuracy at the 95% confidence level than the other methods considered.

The desire to promote healthier and more environmentally conscious methods of commuting has generated increased interest in professional and recreational bicycling in recent years. One of the most important factors cyclists consider when riding is the amount of exertion they will perceive on a given path. In this paper, we build and test a model of the perceived exertion of different categories of cyclists on a daily path within a long bicycle tour. We first propose an additive formula for calculating the perceived exertion of cyclists on component parts of a tour and then present the results of a survey designed to verify the accuracy of the model. Distance, elevation gain, average percent grade, maximum percent grade, and cyclists’ level of expertise are shown to be significant predictors of perceived exertion (p<0.005). Repeated measures analysis indicated that 109 of the 120 perceived exertion levels produced by our model fit the reported perceived exertion levels of the 242 avid cyclists who participated in the validation survey.