Cross Country Season on Flotrack 2013Nov 11, 2013 by James Butler
Reintroducing the Butler Projections
Reintroducing the Butler Projections
James Butler is the current Assistant Coach (Distance) at the University of Missouri - Kansas City. Previously, Butler had worked for three years the University of New Mexico. He has a bachelors degree in Computer Science from Wake Forest and a Masters degree in Sports Administration from New Mexico.
We love The Butler Projections because everything is objectively based. It's all about who beat whom. Plus, we know a few NCAA coaches who use the projections to estimate where their team needs to finish in order to fulfill or surpass expectations.
We'll have the projections for every regional meet later this week. For now, here's a quick reintroduction to The Butler Projections.
This program was developed when I was a collegiate runner in the fall of 2000. It was made to objectively rank individuals through their performances and then to derive team rankings from the individual rankings. It is a derivative of the ELO rankings used to rank professional chess players.
The system compares the expected outcome of a race based on the rankings and the actual outcome of the race based on finishing time. It then adjusts the rankings to more closely match the result.
All the runners start with a ranking of 1000. As the computer starts comparing performances those that do better than expected gain points while those that underperform lose points. Once the computer has run through the entire season, the resulting rankings are used and the season is run through several thousand more times.
Eventually, the rankings converge on specific values, that is to say they no longer increase or decrease as the results are run through. These final values are what is used to rank the individuals with team scores derived from the individual rankings.
The values also have direct significance. Prior to the 2011 NCAA meet last year, Lawi Lalang had a ranking of 1057 and Chris Derrick had a ranking of 1046. Thus by dividing Lalang's ranking by Derrick's, we get a factor of 1.01052. So in a race, Derrick was projected to run 1.01052 times Lalang's time.
Lalang's actual time was 28:44 or 1724 seconds. Given that, Derrick would have been expected to run 1742 (1724*1.01052) seconds or 29:02. Derrick actually ran 28:57 or 1737 seconds and outperformed his ranking versus Lalang.
There is no outside influence on the rankings, the system is only given the results of the meets and produces the rankings given those results.
Below are the projections for the last three NCAA Championships (except for the women's race in 2011) and the actual results.
We love The Butler Projections because everything is objectively based. It's all about who beat whom. Plus, we know a few NCAA coaches who use the projections to estimate where their team needs to finish in order to fulfill or surpass expectations.
We'll have the projections for every regional meet later this week. For now, here's a quick reintroduction to The Butler Projections.
This program was developed when I was a collegiate runner in the fall of 2000. It was made to objectively rank individuals through their performances and then to derive team rankings from the individual rankings. It is a derivative of the ELO rankings used to rank professional chess players.
The system compares the expected outcome of a race based on the rankings and the actual outcome of the race based on finishing time. It then adjusts the rankings to more closely match the result.
All the runners start with a ranking of 1000. As the computer starts comparing performances those that do better than expected gain points while those that underperform lose points. Once the computer has run through the entire season, the resulting rankings are used and the season is run through several thousand more times.
Eventually, the rankings converge on specific values, that is to say they no longer increase or decrease as the results are run through. These final values are what is used to rank the individuals with team scores derived from the individual rankings.
The values also have direct significance. Prior to the 2011 NCAA meet last year, Lawi Lalang had a ranking of 1057 and Chris Derrick had a ranking of 1046. Thus by dividing Lalang's ranking by Derrick's, we get a factor of 1.01052. So in a race, Derrick was projected to run 1.01052 times Lalang's time.
Lalang's actual time was 28:44 or 1724 seconds. Given that, Derrick would have been expected to run 1742 (1724*1.01052) seconds or 29:02. Derrick actually ran 28:57 or 1737 seconds and outperformed his ranking versus Lalang.
There is no outside influence on the rankings, the system is only given the results of the meets and produces the rankings given those results.
Below are the projections for the last three NCAA Championships (except for the women's race in 2011) and the actual results.
|
The Projections |
The Actual Results |
|
2012 Butler Projections |
2012 NCAA DI XC Results |
|
2011 Butler Projections (men only) |
2011 NCAA DI XC Results |
| 2010 Butler Projections | 2010 NCAA DI XC Results |