Late-Game Reversals in Professional Basketball, Football, and Hockey

Gill, Paramjit

doi:10.1080/00031305.2000.10474518

Cited by 14 publications

(5 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In fact, many studies concerning sports have been developed. For instance, those related to frequency distributions in soccer [13], baseball [28,29,30], golf [29], hockey [29,30], football [29,30], and basketball [29,31]. Here, we concentrate our attention on In our analysis about basketball baskets, we verified that the q-exponential function gives a better adjustment than the Weibull distribution since the parameter r, obtained from the q-Weibull one, is close to one (r = 0.91).…”

Section: Empirical Analysismentioning

confidence: 66%

q-exponential, Weibull, and q-Weibull distributions: an empirical analysis

Picoli

Mendes

Malacarne

2003

Physica A: Statistical Mechanics and its Applications

View full text Add to dashboard Cite

In a comparative study, the q-exponential and Weibull distributions are employed to investigate frequency distributions of basketball baskets, cyclone victims, brand-name drugs by retail sales, and highway length. In order to analyze the intermediate cases, a distribution, the q-Weibull one, which interpolates the q-exponential and Weibull ones, is introduced. It is verified that the basketball baskets distribution is well described by a q-exponential, whereas the cyclone victims and brand-name drugs by retail sales ones are better adjusted by a Weibull distribution. On the other hand, for highway length the q-exponential and Weibull distributions do not give satisfactory adjustment, being necessary to employ the q-Weibull distribution. Furthermore, the introduction of this interpolating distribution gives an illumination from the point of view of the stretched exponential against inverse power law (q-exponential with q > 1) controversy. 89.90.+n

show abstract

Section: Empirical Analysismentioning

confidence: 66%

q-exponential, Weibull, and q-Weibull distributions: an empirical analysis

Picoli

Mendes

Malacarne

2003

Physica A: Statistical Mechanics and its Applications

View full text Add to dashboard Cite

show abstract

“…For hockey, this literature would include Mullet (1977), Hurley (1995), Danehy and Lock (1995), Anderson-Cook and Thornton (1998), Berry (2000), and Gill (2000), and an interesting series of papers on when to pull the goalie (see Morrison 1976;Morrison and Wheat 1986;Erkut 1987;Nydick and Weiss 1989;Washburn 1991;Zaman 2002). Research on times for soccer goals includes Chu (2003), Dixon and Coles (1997), and Lee (1997).…”

Section: The Sample Datamentioning

confidence: 97%

Is the Overtime Period in an NHL Game Long Enough? An Example for Teaching Estimation and Hypothesis Testing in the Presence of Censored Data

Brimberg

Hurley

2008

The American Statistician

View full text Add to dashboard Cite

This note outlines an approach for introducing students to estimation and hypothesis testing with censored data using a National Hockey League example. We consider the effects of extending the overtime period in a regular season game beyond its current length of five minutes. The rationale for this change is that more games would be decided on the basis of four-on-four play rather than on a shootout. In order to make this assessment, we must estimate the parameter of the exponential distribution. We used data from the 281 NHL games that went to overtime during the 2005-2006 season. Of these, more than half went to a shootout and hence these potential observations of the time to a goal are censored and must be taken into account in the estimation. For more advanced students, we offer the mechanics of a likelihood ratio test to confirm that four-on-four play in overtime is a much different game than five-on-five in regulation play.

show abstract

“…As we note in Section 4, both the Sagarin ratings' predictions L Sag and the Las Vegas line L Vegas have normally-distributed errors X Sag and X Vegas with means that are not significantly different from zero. Others have also noted and/or used this normal distribution in football (e.g., Gill 2000;Berry 2003;Fanson 2020). Since X is empirically shown to be normally distributed with mean zero, we make the mild assumption that its components E (error in teamstrength estimates) and R (in-game randomness) are also both normally distributed and all independent of each other, with variances σ 2 E • and σ 2 R , respectively, and mean zero for R. (Because ratings are relative to each other, the mean μ E of E is not important.)…”

Section: Modelsmentioning

confidence: 99%

Randomness, uncertainty, and the optimal college football championship tournament size

Muller

Hood

Sokol

2023

Journal of Sports Analytics

View full text Add to dashboard Cite

Every year, there is a popular debate over how many teams should take part in the NCAA’s FBS-level college football championship tournament, and especially whether it should be expanded from 4 teams to 8 or even 12. The inherent tradeoff is that the larger the tournament, the higher the probability that the true best team is included (“validity”), but the lower the probability that the true best team will avoid being upset and win the tournament (“effectiveness”). Using simulation based on empiricially-derived estimates of the ability to measure true team quality and the amount of randomness inherent in each game, we show that the effect of expanding the tournament to 8 teams could be very small, an effectiveness decrease of only 2-3% while increasing validity by 1-4%, while a 7-team tournament provides slightly better tradeoffs. A 12-team tournament would decrease effectiveness by 5-6%.

show abstract

Late-Game Reversals in Professional Basketball, Football, and Hockey

Cited by 14 publications

References 8 publications

q-exponential, Weibull, and q-Weibull distributions: an empirical analysis

q-exponential, Weibull, and q-Weibull distributions: an empirical analysis

Is the Overtime Period in an NHL Game Long Enough? An Example for Teaching Estimation and Hypothesis Testing in the Presence of Censored Data

Randomness, uncertainty, and the optimal college football championship tournament size

Contact Info

Product

Resources

About