Boyd's World-> Breadcrumbs Back to Omaha-> Better, For Now

About the author, Boyd Nation

Better, For Now

Well, the teams have sort of fixed it, but the NCAA has just broken it all over again.

Once again, I'm taking feedback from a reader and running with it, although in this case the reader is Paul Kislanko, who probably qualifies more as a colleague than as a reader. Most of Paul's work is carried over at Southeastern Baseball, and it's worth a subscription by itself, even if you're not a fan of Southeastern baseball in general.

Paul's been doing some excellent work over the last few years on measuring the connectivity of the game, using things like the percentage of teams that are connected either directly or through at least one common opponent. He and I reached the same conclusion this week through different means. He, as a better geek than I am, noticed that the correlation between the RPI and the ISR is the highest it's ever been so far this year. I vaguely noticed (I'll be honest, I don't pay much attention to even my own rankings this early in the year) that the top of each list was tracking much more closely this year than usual -- whoever has happened to be #1 has been #1 on both lists for most of the last month, for example. This ties into Paul's measurements that the game has been increasingly better connected

Here, then, are the numbers that we're talking about, with a little historical perspective.

Year   3/14    End

1997  0.9098  0.9026
1998  0.9066  0.9066
1999  0.8997  0.9012
2000  0.9056  0.9108
2001  0.9097  0.9117
2002  0.9114  0.9135
2003  0.9137  0.9149
2004  0.9129  0.9153
2005  0.9162  0.9182
2006  0.9195  0.9220
2007  0.9230  ----

These are the correlations between the ordinal rankings (#1, #2, and and so on) for the RPI's and ISR's for each season at mid-March and at the end of the year. I used the ordinal rankings because, in this particular case, the ordering is all we really care about.

Remember, the RPI is not inherently broken, it just hasn't worked historically for baseball. It did work for basketball until they broke it in a piece of social engineering a couple of years ago (the current basketball formula differs significantly from the baseball one), so there is a level of interconnectivity where the output is acceptable. Looking at those correlation numbers above, I don't think we've quite reached that point, but we are (or the teams were) making a decent bit of progress. The big remaining problem is that, in the cases where the RPI is wrong, it tends to be quite wrong -- for example, today UC Riverside is sitting at #26 in the ISR's and #50 in the RPI's. Just to forestall the inevitable "Boyd's pointing at the ISR's as gospel" arguments again, they're at #34 in Ken Massey's ratings.

The progress that's been made, though, is that, through a combination of increased spending on travel and some creativity by the teams, increased awareness of scheduling issues by the teams, and generally dropping air fares, more teams are playing more interregional games than ever before, and it's all been to the good of the game, both in increasing the accuracy of our rating systems (one side effect is that, although the ISR's do better with sparse data than the RPI's do, they cope better with more data as well).

And now, it all changes when the schedule rules change next year. Now, the usual disclaimer applies when talking about the NCAA: With apologies to Gertrude Stein, there's really no "there" there (apologies should actually probably go to Oakland residents -- on a recent visit to the area, I was reminded that there's nothing really wrong with Oakland or the East Bay; it's just that San Francisco is cool enough to spoil anybody). The NCAA staff don't really make many decisions. That's done by committees made up of the varying interests within and between the schools. In this case, the Northern teams felt that their best interests were better served by a common, later start date, and they had the numbers to enforce that will.

Now, I won't argue that this was an unalloyed bad decision. Aesthetically, a unified start date will work much better, and there are some fairness issues involved in the weather problems. But, because the teams were also unwilling to extend the season further into the summer (frankly, by the time you're finishing at the end of June, you're a little bit pregnant and might as well get on with having the baby), the number of weekends available for non-conference (and therefore interregional) series will drop by about half in most cases. That means that we're going to, if we're lucky, see a lot more cases of, say ACC teams playing someone the quality of Coastal Carolina or Winthrop rather than getting to match up with UCLA. The individual games will probably be about as good, but the West will return to its isolation, leaving us stuck with more of the eternal regional wars. If we're not lucky, of course, we'll get more schedules the quality of Florida State's in recent years, and not even Seminole fans want to see that.

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.

(*) Pitch count is estimated. As always, I welcome actual pitch count corrections.

Boyd's World-> Breadcrumbs Back to Omaha-> Better, For Now

About the author, Boyd Nation