Blogs : NCAA Division III Blog

Banked Conversion Comparisons: Part II

      Last week, this blog pointed out that the DIII qualifying standards for indoor track nationals impose greater penalties for marks run on banked tracks than the DI or DII qualifying standards. This incited great controversy spawning an 8 post debate as to why this is. Is this equitable? Should DIII athletes receive greater penalties in proportion to their slower times, or, should they in fact be penalized less since they are running at lower velocities and thus, are aided less by banked tracks? I set out to find an answer. 

      Because I am inherently lazy, I limited the data I used to the DI and DIII qualifying standards for the 400m, 800m, mile, 3k (for DI) and 5k. To compare the DI and DIII penalties, I did several linear regressions. To convert the data into a form that I could graph, I converted the automatic qualifying marks for an unbanked 200m track into the velocity needed to runs those marks. I also expressed the time penalty imposed on banked tracks in seconds per meter imposed, multiplied by 10,000 for convenience’s sake. I used velocity as the independent variable, the time penalty as the dependent variable.   

      I graphed the data for DI men, DI women, DIII men and DIII women. I obtained a linear regression equation and an R-squared value for each of the data sets, as you can see below: 
 
 

      If the men’s DI equation is used to predict the DIII men’s penalties for the mile and 5k, the penalties come out to .75 seconds and  1.5 seconds, respectively. For DI, the penalties are .8 and 1.9 seconds. Using the DI curve, the DII women’s mile and 5k penalties come out to .5 seconds for the mile and 1 second for the 5k, compared to .6 seconds and 1.6 seconds for DI. Note that when the DI curves are used to obtain time penalties for DIII, the mile penalties are very close between divisions, but the 5k penalties for DIII are significantly smaller than the penalties for DI. This is the result of one of the biggest problems with my technique, which is that the regression equations cannot be used to extrapolate accurately. Since the DIII 5k qualifying marks are slower than any of the velocities in the DI data sets, the calculated DIII 5k penalties are much lower than they should be.