Boyd's World-> Breadcrumbs Back to Omaha-> Tweaking the ISR's

About the author, Boyd Nation

You know, I'm a guy. I like to tinker. I like to adjust things, realign them, bounce them around, change them until they're perfect. That's always been a source of consternation with me when it comes my rating system, the ISR's. The formula was quite simple in its initial form, and I always figured I'd change it around and perfect it as I went. The problem with that notion is that there's no great way to tell what "perfect" is. I mean, I know your team should be ranked 20 spots higher and all, but what about everybody else? You can do some studies based on performance after some point or something, but it's hard to get enough data from that to know if what you're seeing is an indictment of the rating system or just random chance. Lacking that stiff proof, there's no real reason not to stick with the simple method.

There's a little bit about the ISR method in the Ratings FAQ, which I'll reproduce here. The basic idea is an iterative one. Begin with all teams set to an even rating -- 100 in this case. Then, for each game played, give each team the value of their opponent's rating plus or minus a factor for winning or losing the game -- 25 in this case. Total all of a team's results, divide by the number of games played, and that's the end of a cycle. Then use those numbers as the start of the next cycle until you get the same results for each team for two consecutive cycles. That's always been the whole of it.

I've decided, though, to make one change to that, based on something that I've been able to prove to my satisfaction. The home team has a small advantage. We all know that, but a lot of things that we know are wrong, so I wanted to be sure. I'm comfortable, based on this column and some other studies I've done, that the home field advantage is around 55%, so I'm introducing a factor to the ISR's that takes that into advantage. Essentially, I'll deduct 2.5 points for each home game and add 2.5 for each road game. This is a small change, and I think it makes the ratings a bit more accurate.

In practical terms, there's not a huge change due to this. Last year's top 10, for example, remain the same, although a couple of pairs of teams swap places. However, there are around a dozen teams each year that move by 10 or more spots, so I think it's worth doing. As a general rule, the teams that play on the road most are from cold-weather schools, so that's the group that will benefit the most, with a corresponding drop for warmer teams. Here are the teams who moved the most last year:

 22 Maine
 15 Marist
 15 Illinois-Chicago
 13 Wisconsin-Milwaukee
 13 Michigan State
 12 Kent State
 12 Dayton
 12 Central Michigan
 12 Bowling Green State
 12 Ball State
 11 Western Michigan
 11 LeMoyne
 10 Stony Brook
 10 Purdue

-10 Davidson
-10 Cal State Sacramento
-11 Troy State
-11 North Carolina-Charlotte
-11 Memphis
-11 Louisiana Tech
-11 Kentucky
-11 Charleston Southern
-11 Arkansas State
-11 Alabama-Birmingham
-13 Duke
-16 Hawaii-Hilo
-18 Hawaii

None of these teams were all that close to being tournament-worthy, so the effect on my analysis of the postseason choices would be unaffected. The 22-spot move by Maine is the largest in the five years I have data for.

The ratings will start for this year on March 4. That'll be a little early for them to be meaningful, but I like to go ahead and start then to let you watch how things unfold. Arizona State's 9-0 start obviously makes them an early favorite, but we'll see how the next month plays out.

If you're interested in reprinting this or any other Boyd's World material for your publication or Web site, please read the reprint policy and contact me

Boyd's World-> Breadcrumbs Back to Omaha-> Tweaking the ISR's

About the author, Boyd Nation