User Pass
Home Sign Up Contact Log In
Forum > South America AA Leagues > South America AA #4 > Tex's Statistical Power Ratings - Week 6
Tex Scram
offline
Link
 
Here's something I used to do in our old league.

These are purely statistical formulas that I've come up with to measure different aspects of past performance. You really have to take them all into account to begin ranking teams subjectively.

For example, my RPI really only measures the strength of the schedule that a team has played so far, and says very little about the strength of the team itself. So say two teams have the same AIA or JLG score, you could look at their RPIs and see who has achieved that score against tougher competition.

Again, these are ratings not rankings. Rankings are naturally subjective, and I just like playing with numbers and making objective comparisons. I'll leave the sports section write-ups to others who can do a much better job than I could.

The initial order of the table is simply the order listed on the conference/league standings page. You can sort each column by clicking on the column headers.

Feel free to ask questions, but I won't be sharing the actual formulas themselves.


http://grimland.uta.edu/glb/SouthAmericaAA4-Alpha-Week6.html

http://grimland.uta.edu/glb/SouthAmericaAA4-Zeta-Week6.html
Last edited May 4, 2009 22:44:00
 
JefT@nk
offline
Link
 
Interesting, thanks for putting that together.
 
Ravenwood
offline
Link
 
That is definitely interesting. Could you perhaps provide an explanation of what the acronyms to the right of Win% stand for though?
 
H2C
offline
Link
 
Originally posted by Ravenwood
That is definitely interesting. Could you perhaps provide an explanation of what the acronyms to the right of Win% stand for though?


Yes this
 
Qwhatchet
offline
Link
 
<removed explanations>

I'm not sure if what I gave was too much info (just what he has said about them in the past) I will let him post the explanations himself.
Last edited May 4, 2009 20:48:21
 
Tex Scram
offline
Link
 
Oh sorry - I guess I should explain my ratings. I'm a bit of a numbers nerd and these are just a few things I put together last season one day when I was bored at work.

AIA is the simplest formula. It pretty much just looks at W/L, and just a little bit of scoring differential. Wins over better teams are worth more.

JLG is similar, but a little more complex. It also adds in penalties for losing to lesser teams and obviously works on a different scale.

I'm not really happy with OMF and DMF. They were attempts to try and predict how many points a team will score (OMF) and allow (DMF) based on what they've done in the past. I recently tried to tweak it to take into account how their next opponent stacks up against their previous opponents, but it still needs more work.

My RPI is not the standard RPI you might see in college football (maybe I should call it something else). It mostly measures the strength of the schedule a team has already played. It comes in handy to be able to compare the other ratings without having to wait until the end of the season when everyone has played everyone else. There is a small component of team strength (based on several factors) so the further we get into the season, the more accurate it will be in comparing teams to each other directly.
Last edited May 4, 2009 22:50:36
 


You are not logged in. Please log in if you want to post a reply.