CHL Rankings
As some of you may be aware, I have devised my own “power ranking” algorithms to help me analyze several professional sports for the purposes of sports betting. I only bet when I have an edge and this algorithm helps me get that edge. The algorithm uses key stats from each sport to help me identify arbitrage opportunities to take advantage of. I’ve been moderately successful with this on my trips to Vegas, but not enough to make me want to quit my day job. Surprisingly enough, the sports betting market is VERY efficient in Las Vegas so there isn’t much arbitrage to be had. I’m still tweaking the formulas all the time. Maybe when I get it just right, I’ll publish a book.
Anyway, I have applied this in the past to create Puddinman’s Aggregate Ranking Algorithm™ for the CHL. Last year, I got it to be a pretty damn accurate predictor of the outcome of games. Too bad there isn’t a market for betting on CHL games! (Yes, I have used it for NHL betting and it has been pretty successful.) Just to give you an idea of how good the system is, let’s look at last year. At mid-season of the 2008-2009 CHL season, my Algorithm gave me this as the top 10 Power Ranking:
- Texas
- Colorado
- Laredo
- Mississippi
- Odessa
- RGV
- OKC
- Bossier-Shrievport
- Rocky Mountain
- New Mexico
I’d say, given how the rest of the season and playoffs played out, that was pretty damn accurate. The only thing it didn’t forsee was the rise of CC to overtake AZ and NM, but that can be ascribed to the coaching change in CC among other things which my formulas don’t really account for (I can’t see the future, dammit, or read the players minds!). So, inspired by my friend, BucksFanAddict at chlforums.com, I decided to apply the Algorithm to this years budding season. Here’s what we have so far:
2009-2010 CHL Power Rankings as of 11/3/09
- Odessa
- Amarillo
- Colorado
- Allen
- Laredo
- Corpus Christi
- Texas
- Arizona
- Rapid City
- Tulsa
Now, understand that I would never use these rankings as the sole determinant of my betting decisions at this point in the season; the sample size is way too small with the average games played of only 6 games. There are also several other factors I consider before making a bet that can’t really be plugged into a formula, such as rivalries, roster moves, injuries, last-5-game stats etc. For example, with David Nimmo injured, I would drop Amarillo a few notches. But I think this is a pretty dang fair analysis of the performance of the teams up to this point in the season.
So what say you? Do you like it or do you think I’m full of shit?
