In the fall of 2013 we dipped our toes in the waters of Ohio High School soccer, candidly, with mixed reviews. Ohio presented some new challenges with many, many more schools, and more ways to slice and dice groups of schools. We could handle 3 divisions and 6 districts, but with conferences of isolated groups of teams as well, it was quite a challenge. Last year, we topped out at about an 82% prediction rate. Sounds ok, but when you consider that groups of teams rarely played outside their isolated conferences, it was hard to ever be proven wrong! I jokingly thought it seemed like a problem we could live with, but it wasn’t passing the smell test. There were teams obviously ranked wrongly.
So, we are back at it this year, and better late than never, just in time for tournament play. See the results in the charts below. We are able now to finish seasons reliably in the 87% range, which is around where folks really started to believe in the system – at least in Kentucky. For this year, you can see the system just eclipsing 85%, and yes, we have had to collect data going back to 2005 to build up enough history in the database to sort out all of the isolation related issues. Still common sense often rules in these matters, so please, comment on your direct observations, and we can continue get better!
As always, please keep on the look out for silly errors. With the amount of data collected, I can’t track schools that have merged, changed names, started new teams or stopped playing. I can use the help!