2010 Race Predictions

April 20, 2010

Based on excel regression and simulation. I am predicting Phi Delta Theta to be the race winners, with the Cutters coming in second. The rest of the field as follows:

1 Phi Delta Theta
2 Cutters
3 Phi Kappa Psi
4 Phi Gamma Delta
5 Sigma Chi
6 Black Key Bulls
7 Beta Theta Pi
8 Gray Goat Cycling
9 Delta Tau Delta
10 Delta Upsilon
11 Acacia
12 Delta Chi
13 Evans Scholars
14 Sigma Phi Epsilon
15 Sigma Nu
16 Cru Cycling
17 Hoosier Climber?
18 Theta Chi
19 Wright Cycling
20 Delta Sigma Pi (M)
21 Kappa Sigma
22 Pi Kappa Phi
23 Zeta Beta Tau
24 Collins Bunny Gamma
25 Air Force Cycling
26 Cycling For Freedom
27 Phi Kappa Sigma
28 Achtung!
29 Phi Sigma Kappa
30 Dodds House
31 Sigma Alpha Mu
32 Emanon Cycling
33 Lambda Chi Alpha

This year will certainly be a close race, with 5 teams being within the standard error of the model and thus having a shot at the win. Probabilities of a given team winning are as follows:

Team Chance of Winning
Phi Delta Theta 24.5%
Cutters 18.3%
Phi Kappa Psi 16.8%
Phi Gamma Delta 11.5%
Sigma Chi 7.4%
Black Key Bulls 6.0%
Beta Theta Pi 4.2%
Gray Goat Cycling 3.1%
Delta Tau Delta 2.8%
Delta Upsilon 2.4%

Check back next week. I will be comparing actual results to the predications

How it works:

A regression of historical data reveals that 67% of the race outcome can be explained by ITT times, team pursuit times, and qualification times. The ITT time of the team’s fastest rider and the team pursuit time were the most significant variables to explain race outcome. The ITT time of the fourth rider was insignificant and not included in the regression model. The model can predict a team’s race time to within 177 seconds. The expected race times are assumed to be normally distributed. This assumptions allows for simulation of 1000 iterations using the standard error of 177. This revealed each team’s chances of winning as listed above.

The model assumes that a team’s four fastest riders ride in the race. This assumption if, of course, imperfect but it provides the only simple way to regress ITT times to race outcome. All times are normalized to their respective yearly averages. This accounts for anomalies in weather, track conditions, and excessive yellow flags.

Times below/above event average
Team Rider 1 Rider 2 Rider 3 Rider 4 Team Pursuit Quals Race predicted
Phi Delta Theta -14.54 -12.17 -8.92 -5.84 -47.67 -5.36 -425.28
Cutters -16.58 -11.39 -6.70 -4.62 -25.64 -7.82 -388.10
Phi Kappa Psi -13.09 -12.29 -11.83 -7.85 -34.13 -6.77 -374.91
Phi Gamma Delta -11.79 -10.50 -8.81 -7.02 -29.49 -2.88 -312.51
Sigma Chi -10.56 -10.35 -9.93 -6.28 -24.98 -3.91 -281.77
Black Key Bulls -10.81 -9.52 -7.70 -5.54 -23.19 -4.09 -266.57
Beta Theta Pi -11.15 -7.20 -5.23 -1.98 -21.28 -6.13 -238.75
Gray Goat Cycling -10.03 -6.68 -5.73 0.89 -22.00 -6.47 -218.27
Delta Tau Delta -10.37 -7.77 -4.40 -4.22 -6.84 -6.36 -189.33
Delta Upsilon -9.77 -9.45 -6.77 8.02 2.41 -1.19 -175.45
Acacia -8.42 -0.96 1.18 7.59 -16.17 3.43 -96.49
Delta Chi -6.83 -4.90 0.78 1.83 -11.14 -2.65 -95.41
Evans Scholars -7.76 -4.13 2.13 2.36 5.40 -3.54 -56.08
Sigma Phi Epsilon -5.51 -4.90 -1.71 5.37 -0.98 1.58 -51.87
Sigma Nu -4.44 -4.30 0.87 -8.73 2.21 -40.16
Cru Cycling -4.68 -3.67 -1.73 -0.30 3.94 -1.29 -14.11
Hoosier Climber? -3.04 -1.75 -0.95 2.15 -2.19 -3.61 17.17
Theta Chi -2.84 -2.17 -0.78 3.02 9.46 4.02 51.19
Wright Cycling 4.61 -8.86 -4.67 2.70 -10.03 0.86 52.25
Delta Sigma Pi (M) -6.45 3.14 10.28 10.83 7.47 1.68 67.06
Kappa Sigma -0.59 0.79 0.84 1.32 0.13 -0.79 95.83
Pi Kappa Phi -1.69 -1.53 -0.64 5.84 17.27 -2.63 95.93
Zeta Beta Tau 1.34 96.30
Collins Bunny Gamma -1.47 -0.13 4.22 5.78 6.25 4.87 101.33
Air Force Cycling -6.04 3.36 8.92 16.11 20.85 4.32 108.58
Cycling For Freedom -0.53 0.08 6.55 13.19 6.07 110.61
Phi Kappa Sigma 3.46 5.84 157.68
Achtung! -2.68 6.67 11.37 27.06 19.66 4.67 201.88
Phi Sigma Kappa 1.22 2.29 3.94 7.70 19.74 4.22 205.08
Dodds House 1.30 3.40 3.46 11.94 17.57 5.93 209.98
Sigma Alpha Mu 1.00 6.74 7.11 8.34 22.13 5.37 259.30
Emanon Cycling 3.38 6.38 6.58 8.84 19.84 3.55 288.57
Lambda Chi Alpha 10.31 11.67 68.99 5.62 568.30

Regression Statistics:

Coefficient Standard t-Value p-Value
Regression Table Error
Constant 95.8205 40.2274 2.3820 0.0189
Rider 1 17.2550 6.3459 2.7191 0.0076
Rider 2 9.2340 8.2817 1.1150 0.2673
Rider 3 3.1905 7.2550 0.4398 0.6610
Team Pursuit 2.6771 0.9173 2.9184 0.0043
Quals 0.3559 4.3517 0.0818 0.9350

Special thanks to Wayne Winston for help: http://www.waynewinston.com

Advertisements

2 Responses to “2010 Race Predictions”


  1. […] Here’s a link to the full post which is a must read for any Little 500 enthusiast. […]

  2. Justin Says:

    ROLI,

    Nice job! You would have really raised your value as mechanic if you could have done this a few years ago. I’ll be down for the race, hit me up.

    Proud to be…


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: