Boyd's World-> Breadcrumbs Back to Omaha-> A Flaw in the ISR's?

About the author, Boyd Nation

A Flaw in the ISR's?

I'm stuck this week with a problem I always have this time of year. It's what most people consider the most important part of the season, and I don't really have anything appropriate to say. I mean, if you're enough of a college baseball fan to be here (given that I don't do a whole lot of promotion to non-fans or the only sort-of-interested), then you probably have just as valid an opinion of the CWS field as I do. I don't do quotes or player features because, well, have you ever really read those things? In general, all I can do is wait with the rest of you and see how it goes. Therefore, I'm going to do what I do best, which is write about rating systems. It turns out that there's a topic there that might affect your expectations for what happens this week, though, and I want to dig a bit.

It's been pointed out a few times over the years in varying degrees of depth that it's quite possible for a win to hurt a team's ISR. The reason for this, without drowning in the technical details, is that, for each game that a team plays, a score gets averaged into their ISR that is their opponent's ISR (the actual rating, not the ordinal ranking) plus (or minus for a loss) 25 points (plus a home-field factor, but let's ignore that for this week). Therefore, if you beat a team that's more than 25 points worse than you, your ISR actually goes down.

This is certainly true. The question, which we won't answer today, is whether it should be true. After all, part of what we're measuring in the ISR's is the level of competition that the team's level of success has been achieved against, and there's no question that scheduling a, to use a highly technical term, crappy opponent lowers the overall quality of the schedule. On the other hand, all a team can do is beat the opponents that are on the field with them, and we're not on any sort of moral crusade here; we're just trying to figure out how good everyone is. My general take on it, which I'm going to stick with until someone (including me) finds a better way to handle it, is to leave the ISR's alone but notice that some sort of uncertainty factor would be a useful thing to have.

As a concrete example of why this would be helpful, let's look at this year's most notable case. Every year there seems to be one team who is both a legitimate national contender and has a large number of teams who drag down their ISR. It used to be Miami, but they've fixed (or at least tried to fix) their schedule. Last year, it was Florida State. This year, it turns out to be South Carolina. Of the 65 games that the Gamecocks have played this year, 15 have been against teams who were more than 25 ISR points worse than they were. First of all, they should be ashamed of that. More relevantly, though, it makes it much harder to know how good they are. If you just throw those 15 games out, they're 35-15 instead of 50-15, which isn't as impressive. If you imagine those 15 games replaced by games against teams the level of the rest of their schedule, you get a schedule that would actually be top 20 in the nation. Now, given their success rate against the rest of their schedule, we could assume that they would have gone 10-5 or 11-4 in those games, but that might not be a safe assumption -- the extra workload for the front of the rotation might have caused problems, or they could have still gone 15-0 by replacing the really weak with the only mildly weak (replacing Delaware State with William and Mary, for example, would be unlikely to stress them but would give us a better idea).

A different version of this is teams who are really dominant in weak conferences. I've talked about this before, but we really have no idea how good Southern was in 2003 or Oral Roberts was this year.

The short version of this, then, is that teams who play a large number of really weak opponents could be Miami from 1999 and they could be Florida State from 2003, and we'll have to watch to see which one South Carolina is. A couple of lists:

The Ten Teams Who Played the Most Relatively Weak Opponents for 2004:

22 Oral Roberts
15 South Carolina
 9 Dallas Baptist
 8 East Carolina
 7 Vanderbilt
 7 Notre Dame
 7 North Carolina
 7 Georgia
 6 College of Charleston
 6 Central Florida

Number of Relatively Weak Opponents for 2004 CWS Teams

15 South Carolina
 7 Georgia
 5 Texas
 2 Arkansas
 2 Louisiana State
 2 Miami, Florida
 0 Arizona
 0 Cal State Fullerton

The ISR-based probabilities for the CWS:

Texas, 32%
Miami, 13%
South Carolina, 12%
Cal State Fullerton, 11%
Louisiana State, 11%
Arizona, 8%
Arkansas, 7%
Georgia, 6%

Rather than keep returning to the subject of pitch counts and pitcher usage in general too often for my main theme, I'm just going to run a standard feature down here where I point out potential problems; feel free to stop reading above this if the subject doesn't interest you. This will just be a quick listing of questionable starts that have caught my eye -- the general threshold for listing is 120 actual pitches or 130 estimated, although short rest will also get a pitcher listed if I catch it. Don't blame me; I'm just the messenger.

Starting with a couple more double appearances from the regionals that I missed:

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-> A Flaw in the ISR's?

About the author, Boyd Nation