Ed. note: I need to make a clarification. In these calculations, only Division I competition is counted. I can already hear the murmurs about how can Baylor be #1 in SOS, but those matches from UMHB and UT-Dallas are not included.
I have completed the RPI and SOS calculations for both men’s and women’s roups. To make the data a bit easier to read, I deleted all of the teams that had fewer than 15 recorded results. Having so few results lends itself to outliers and I am not in the mood or have the cognitive powers to get into that right now.
The lists are pretty straight forward. The files can be found under the rankings tabs above. The first column of each report is sorted by RPI, the second by SOS.
I am using the following definition of RPI (ratings percentage index) = team’s winning percentage (25%), its opponents’ winning percentage (50%) and the winning percentage of those opponents’ opponents (25%).
That was the NCAA definition when it was created in 1981. They have since jacked with it to add a factor for home v. away that completely jacks it up too much.
The SOS (strength of schedule) = opponents’ winning percentage (33.3%) + winning percentage of those opponents’ opponents (66.7%).
The direct links to the pages are here:
MEN - http://www.texascollegetennis.com/?page_id=921
WOMEN - http://www.texascollegetennis.com/?page_id=924
One thing I would like to get into later and have discussed some, is the HUGE advantage teams in the SEC have. When I was testing the numbers, I was checking the winning percentage of those opponents’ opponents. These totals are completely dominated by the SEC. I will post them tomorrow with more detail. It is extremely out of whack.
Again it is late (or early, depending on how you see it), so I need to go to bed.
If you are into number and this sort of thing, I suggest Jerry Palm’s websites, http://collegebcs.com/ and http://collegerpi.com/. He is the best and I have followed his work for years.
Or is you just want to profess your geekiness, http://shop.neatorama.com/product-info.php?i-heart-curve-math-tshirt-pid102.html. I love these.
I looked at your team, RPI and SOS rankings tonight before the ITA’s come out tomorrow and again I agree with you 100%. I have always thought that Fresno State, Florida and Arkansas did not belong in the top 16 where they have been most of the year and after reading your charts tonight, I feel even more so. I have thought North Carolina, Florida State and Michigan beloned higher up in the rankings before NCAA regional’s are announced tomorrow but none of them will be there unfortanetley
Fresno State is a very good team but they play a very soft schedule and Arkansas and Florida to me seem way over rated this year and belong closer to the 20-22 range rather then where they are. I am not a mathematician like you are, but I do follow the top 25 very closely and in the past on what is going on here. The ITA has to upgrade their formulas they’re using to determine the correct rankings.
They should look at your rankings and Tennis Recruiting as the two are very similar on determining rankings.
Great job!
tennisguy2894