Chicago Fanatics Message Board
http://www.chicagofanatics.com/

Predicting the NCAA tournament using statistical analysis...
http://www.chicagofanatics.com/viewtopic.php?f=101&t=56889		 Page 1 of 1
Author:  Psycory [ Wed Mar 16, 2011 2:20 pm ]
Post subject:  Predicting the NCAA tournament using statistical analysis...
A grad school friend and I do it every year, but in a couple of dollars and put it in my pool. It never wins, however, it can give you some general trends on who to like/avoid. We have data going back the last 12 years for all the teams that are in the tournament a some stats (Sagarin rating, sos, rpi, conference Sagarin, conference rpi, offensive and defensive statistics as well) to create a profile of winners and losers for each round based on the past 12 years of data. In the analysis when two teams fit the "winners" profile, we have two different decision rules for the tie breaker (absolute value of winning score, difference of winner profile score - loser profile score).
For the first set of decision rule (absolute value), here are what we saw
1st round upsets: Villanova, Tennessee Oakland, Missouri, FSU, ODU, Utah State
2nd round upsets: Utah State over Wisconsin Georgetown over Purdue
Sweet sixteen upsets: Kentucky over Ohio St, Uconn over SD st, Georgetown over ND, BYU over Florida
Final four teams: Kentucky, Duke, Georgetown, BYU
Final game: Duke over BYU

2nd decision rule (difference score)
1st round upsets: Villanova, Tennessee, ODU, Utah State
2nd round upsets: Arizona over Texas, Georgetown over Purdue, Utah State over Wisconsin
Sweet sixteen upsets: Kentucky over Ohio state, BYU over Florida
Final four teams: UNC, SD St, Kansas, Pitt
Final game: SD St over Kansas.

Now bear in mind that the system does not take injuries or suspensions in mind, hence liking BYU. One thing is very clear though, the stats do NOT like the big ten (only Ohio State makes it to the sweet 16). Also interesting (at least to me) that the 1st rule tends to be more liberal and the 2nd rule tends to be more conservative, yet the first descision rule leads to the more conservative pick for the champion, I don't know that that means, but it is kinda cool.
If you are interested in the significance tests for the profile analysis, the first three rounds were significant at the .05 level, last three profiles were not. If you want to look at the SPSS output, message me.
Author:  Brick [ Thu Mar 17, 2011 7:08 am ]
Post subject:  Re: Predicting the NCAA tournament using statistical analysi
Interesting as always Psycory. Please adjust your algorithm to make Purdue win it all next year. Thanks.
Author:  Douchebag [ Thu Mar 17, 2011 7:13 am ]
Post subject:  Re: Predicting the NCAA tournament using statistical analysi
After doing my statistical analysis, I'm 100% certain that I am not going to read this Sinaclypse post.
Author:  Psycory [ Thu Mar 17, 2011 10:20 am ]
Post subject:  Re: Predicting the NCAA tournament using statistical analysi
Boilermaker Rick wrote:
Interesting as always Psycory. Please adjust your algorithm to make Purdue win it all next year. Thanks.

Right after I get Penn State to win it all...and on that note, if you would have looked at the ratings/statistics of Penn State, it's amazing that they are under the radar about how bad they are, yet made the tournament and teams like VCU and UAB were getting ragged on. I guess it is nice being in a power conference.
Author:  Psycory [ Sat Mar 26, 2011 7:24 am ]
Post subject:  Re: Predicting the NCAA tournament using statistical analysi
Well at least I had the "No Big 10 team beyond the sweet sixteen" prediction right. Everything else? eh not so much.
Page 1 of 1 All times are UTC - 6 hours [ DST ]
Powered by phpBB® Forum Software © phpBB Group
https://www.phpbb.com/