|
Activity Number:
|
272
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Tuesday, August 4, 2009 : 8:30 AM to 10:20 AM
|
|
Sponsor:
|
Section on Statistics in Sports
|
| Abstract - #304898 |
|
Title:
|
Models for Motorcycle Grand Prix Racing Times
|
|
Author(s):
|
Leanne Streja*+ and Robert E. Weiss and Catherine A. Sugar
|
|
Companies:
|
University of California, Los Angeles and University of California, Los Angeles and University of California, Los Angeles
|
|
Address:
|
, Los Angeles, CA, 90095,
|
|
Keywords:
|
motorcycle racing ; loess regression ; analysis of variance
|
|
Abstract:
|
Motorcycle Grand Prix (MotoGP) is the most elite form of motorcycle racing with approximately 18 events per year at tracks world-wide. MotoGP teams are interested in understanding the factors predictive of fast race times in order to improve their performance/chances of winning. However, to our knowledge no one has ever directly modeled lap times and used it for strategies to win. In this analysis I generate a statistical model of lap times from MotoGP 2002--2007 race seasons with the goal of predicting results under optimal conditions. I plot and evaluate lap times throughout a race for all riders, tracks, and years of racing. I systematically remove the slow laps from the analysis to generate a model representing optimal race conditions. Several analysis of variance models are considered along with several versions of the data sets and compared to an actual MotoGP race.
|
