Boyd's World-> Breadcrumbs Back to Omaha-> Who's Hotter?

About the author, Boyd Nation

Last week, I looked at the question of whether regional performance was a good predictor for super-regional performance. The results were too fuzzy to be of much use, but I thought I'd follow up just to be complete. Interestingly, this past weekend's results completely reversed the mild trends of past years, as the 3-1 teams went undefeated (with one win over a 4-1 team and two over 3-0 teams), while a 3-0 team beat a 4-1 team. This just cements the notion that there's no predictive value there.

This week, I want to look at an idea that may also not work as a predictor, but is a lot more likely to succeed than last week's. I was reminded this week of a common question about the ISR's, "Why don't they give more weight to recent results?" There are two answers to that, one aesthetic and one practical. First of all, I'm not trying to predict the future (other than at times like this); I'm trying to measure how well teams have played so far, and those can be contradictory goals. There's no official definition of what the word "better" means in this context (whoever wins the title next week, please remember this), and I choose to value the entire season rather than a one-month postseason. Secondly, the last time I looked at this issue (around 1999, I think), I couldn't find any subset of the season that actually did better at predicting the postseason than the whole season did. I have several more years of data now, so I'm going to take another look at it during the offseason; it might be that I could produce another measure that does a better job of predicting the future than the ISR's do, although I probably wouldn't replace the ISR's.

With that in mind, there are anecdotal cases of teams improving over the course of the season. Last year, South Carolina was #9 in the country from the beginning of the season to the beginning of the CWS, but they were #5 from March 15 until the beginning of the CWS. LSU follows a more extreme version of this same pattern most years. Since there may be some value in the postseason to taking a month to get your kinks worked out, let's look at the ISR's for this year based on a few different starting dates:

1/01             2/01             3/01             4/01             5/01

 1. CSUF          1. Rice          1. Stanford      1. Texas         1. Stanford
 2. Rice          2. Stanford      2. Rice          2. Florida St.   2. USC East
 3. Stanford      3. CSUF          3. Texas         3. Stanford      3. SMSU
 4. Texas         4. Arizona St.   4. CSUF          4. S. Alabama    4. Va. Comm.
 5. Arizona St.   5. Texas         5. LBSU          5. LSU           5. Texas
 6. LBSU          6. LBSU          6. LSU           6. USC East      6. LBSU
 7. Baylor        7. Baylor        7. Washington    7. Arizona St.   7. Florida St.
 8. Florida St.   8. TAMU          8. Arizona St.   8. CSUF          8. Georgia
 9. TAMU          9. Florida St.   9. TAMU          9. La-Lafayette  9. Illinois St.
10. LSU          10. LSU          10. Florida St.  10. S. Miss.     10. Baylor
14. Miami        14. Miami        15. Miami        12. Rice         12. LSU
28. USC East     27. USC East     27. USC East     24. Miami        13. CSUF
37. SMSU         38. SMSU         35. SMSU         40. SMSU         16. Miami
                                                                    17. Rice

My intuition tells me that the difference in March 1 and April 1 is the difference in "improving after a shakedown cruise" and "riding a hot streak", but the point of the exercise is not to trust my intuition. Nonetheless, fear the Cardinal. Note that the May results are just silly (Illinois State went 8-3 in May, including a four-game split with Southwest Missouri). And, just to beat a dead horse a little bit more, note that perpetually-declared-to-be-slumping Arizona State was still one of the eight best teams in the country after April 1.

As I said, I'll do a more seriously mathematical study of this issue in the offseason, but I thought I'd throw these on out there for anyone who wants to try to predict a hot hand or recognize a legitimate in-season improvement.

Chances to reach the final series and win, respectively:

Texas                      32/15
Miami, Florida             16/ 5
Southwest Missouri State    6/ 1
Rice                       46/25
Louisiana State            14/ 6
Cal State Fullerton        44/26
South Carolina              5/ 1
Stanford                   37/21

I've suggested a few times that they move to random placement of the field in Omaha, since that would take the committee out of the position of pretending to know more than they do (quick, what's the difference in the #6 and #7 team?) and would remove a big source of pointless criticism. It occurs to me looking at this that upsets tend to produce the same effect as often as random seeding would; I count two serious contenders on each side of the bracket, and none of them play each other in the first round.

Just as a quick side note, ESPN did a really great job with their coverage last weekend. Their CWS coverage has always been a bit shaky -- one or two games each year get shown tape-delayed, and Harold Reynolds is only there to tell us how he doesn't really like college baseball -- but this last weekend there were several praise-worthy points. First of all, there was lots of it. The PPV-TiVo guys could watch every single pitch, as far as I know, but even on live coverage on ESPN and ESPN2, they got at least seven games over four days, which makes quite a feast. Secondly, the announcers, while still definitely TV guys (there's a lot of airtime to kill, and not everything that's said extemporaneously should be), were glad to be there, knowledgeable about the college game, and thoroughly professional. Finally, the production values were top-notch, indicating that they're putting some effort into the project. Kudos all around.

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.

The McCally count is corrected based on an actual count.

The supers have, in years past, been a magnet for overuse; it's good to see that Steven White was the only pitcher to be really left out there this weekend, as far as I can tell.

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-> Who's Hotter?

About the author, Boyd Nation