| Boyd's World-> Breadcrumbs Back to Omaha-> The 2001 Tournament Field | About the author, Boyd Nation |
|
Hey! Click here to do a good thing! The Boyd's World KidOne Campaign |
Publication Date: May 22, 2001
Not Unlike a Prostate Check
Well, it's that time again. That annual occurence so painful to college baseball fans, when we must look at the carnage wrought upon our game by the field chosen by those entrusted to choose which teams will get to play for the most coveted title in the game -- the national championship.
I must admit that my initial reaction to seeing this year's field was fairly positive. I suspect that's because they've lowered my expectations to the point where I'm pleased any year that don't actually select Maryland-Eastern Shore as an at large #1 seed. After having some time to actually look into it at some depth, though, I see that only a little progress has been made, and that some of my joy was caused by a couple of the teams that I thought would get really undeserved at large bids winning their conference tournaments and making the question moot.
In the sense of fairness, though, I must point out that there were a couple of encouraging notes this year. Although they did pair them with absurdly good #2 seeds, the committee did correctly place Stanford, Cal State Fullerton, and Southern California in the top four national seeds. They also took eight SEC teams for the first time, which granted the inclusion of a couple of teams I expected to make my most mistreated list. Both of these actions have met with some criticism from representatives of the two groups involved, both of whom seem to be missing the actual source of their problems.
I'll include a list of reasons that the committee tends to make errors below, but the biggest reason seems to be that the RPI misrates many teams from mid-major conferences, which means that the committee takes those teams rather than deserving Western teams. This is not a new problem this year. There are other problems involving overselection of Northern teams, but that involves a relatively small number of bids and is a political problem, not a process problem.
I'll have a table included in my discussion of each regional, so I need
to describe the columns.
W-L -- Won-loss record against Division I opponents
pRPI -- Pseudo-RPI ranking
ISR -- ISR ranking
S -- Correct seed according to the ISR's
Reason -- Probable reason for mis-seeding, according to the list
below
Probs -- The probability of the team winning the regional,
super-regional, and championship, respectively, according to the method
described below.
Why They Get It Wrong
These are the codes for the reasons column, which shows most of the reasons that I'm guessing are used by the committee when they get a team in the wrong place, or in or out when they shouldn't be. In one or two cases, this is a matter of policy rather than misunderstanding, which doesn't make it any more fair.
C -- Conference record. The committee uses conference record (and frequently, conference standing) as a shortcut to try and thin out the contenders. This ignores the fact that it's not unusual for an obviously better team to finish behind a lesser team due to problems of sample size in the number of conference games played. It also ignores the fact that each game is a zero-sum event, so in conferences with a large number of excellent teams, somebody has to lose each game no matter how good they are. This latter fact cost the Pac 10 and Big West dearly this year.
N -- Reputation (Name). There are schools that get bonus points just for who they are. I honestly believe that this is not intentional; it's just a matter of awareness. The committee is made up of people with full-time jobs; even if they're trying to pay attention as much as they can during the season, it's got to be more eye-catching to see results from a big-name program than from an upstart. This also holds true at the conference level. This factor can also work against you if you have some unexpected losses, as Wichita State seems to have learned this year.
R -- RPI. Did I mention the RPI's stink? If anything, this year's committee seems to have paid more attention to them than usual.
S -- Snow factor. The basic effect of the stated policy of trying to promote the game in the Northeast and Midwest seems to be that teams get a boost if there is snow on their home field for some extended time during the season. While growing the game nationally is a laudable goal, rewarding mediocre performance with tournament bids is not a way to achieve it, simply because it's unlikely to work; it provides no feedback to the schools to get them to do the things needed to improve.
T -- Conference tournament. The committee always seems to pay too much attention to the conference tournaments, apparently not realizing that giving that much weight to two to five games is silly, no matter when they take place.
W -- Winning percentage. There are a lot fewer examples of this than there used to be (the RPI is better than nothing, apparently), but there are still times when the committee pays too much attention to raw won-loss record with no consideration of strength of schedule. This year they took every team with at least forty wins against Division I teams, which led to the biggest stinker among the at large picks.
? -- Huh? There's one pick for which I simply cannot come up with an explanation.
The Probabilities
The probabilities I give here are based on a system I've devised which uses historical results between teams at each of the possible gaps in the ISR's. For example, over the four years that I've done them, teams who were three ISR points ahead of their opponents have won 59.9% of those games. I take those numbers for each game, compute all of the possible outcomes, and add up the probabilities. These are not exact, and they ignore a couple of relevant factors like home field advantage (small but real for most teams) and pitching matchups, but they're a reasonable raw tool for beginning analysis if you don't take them too seriously. I don't want to go into too much detail here, since I've already got folks nodding off, but feel free to email me if you're interested in more detail.
They're rounded to the nearest percentage point, so anybody under half a percent goes to zero. This is sports, of course, so everyone has at least some chance of winning; feel free to read it as .49% if it makes you feel better about your team's chances.
The Field
W-L pRPI ISR S Reason Probs
1 Cal State Fullerton 37-15 1 2 1 - 76/70/22
2 Arizona State 36-18 27 10 1 RC 17/13/ 1
3 Texas Tech 34-18 44 24 2 RT 7/ 4/ 0
4 Temple 26-29 131 164 4 - 0/ 0/ 0
According to these probabilities, Cal State Fullerton has the best chance of anyone of winning the national championship. That's not a bad thing; they're a deserved #2 in the ISR's. The bad thing is that 22% is the lowest I've seen for the highest total since I started doing these things three years ago; most years someone comes in with a 30% or so chance. They deserved a fairer field than this, and Arizona State certainly deserved better placement.
W-L pRPI ISR S Reason Probs
1 Mississippi State 36-22 11 31 2 RT 67/11/ 0
2 Ohio State 42-15 62 73 - RS 19/ 1/ 0
3 Delaware 43-13 39 87 4 RS 13/ 1/ 0
4 Kent 29-28 150 163 4 - 1/ 0/ 0
OK, I'm in full-throttle internal strife mode here. I'm an MSU graduate and fan, so part of me is really, really pleased with this placement. I'm also an analyst who tries to point out unfairness where he sees it in the selection process, and this is a ridiculous field. Ohio State gets a charity snow factor bid, and then, to further complicate things, they and Delaware both get higher seeds than they deserve. Had the NCAA actually sent a legitimate #1 seed to Columbus, this could have been really ugly; as it is, it's just a mess.
W-L pRPI ISR S Reason Probs
1 Nebraska 45-14 18 9 1 - 67/38/ 3
2 Rutgers 40-15 23 65 - RS 7/ 1/ 0
3 Brigham Young 37-19 59 29 2 R 24/ 9/ 0
4 Northern Iowa 30-26 119 108 4 - 2/ 0/ 0
The cynic in me thinks that they were trying to grease the skids for a Nebraska trip to Omaha (OK, geographically, that didn't make any sense, but you know what I mean), and they just didn't do it very well because of undervaluing the Cougars. Still, being able to save Komine for Saturday while BYU has to take Rutgers seriously will probably bring the Huskers through here and into a potentially interesting super.
W-L pRPI ISR S Reason Probs
1 Rice 43-17 20 5 1 - 55/34/ 3
2 Baylor 32-21 29 21 2 - 27/12/ 0
3 Houston 29-28 34 47 - R 10/ 3/ 0
4 Texas-Arlington 34-23 47 49 3 - 8/ 2/ 0
This is not a bad field at all. Teams that have played really, really well at some point in the season, the way Arlington did for six weeks, are probably more dangerous in the postseason than those who haven't (see last year's Louisiana-Lafayette team for an example). Baylor is better than they showed last week. Houston fell just below the cutoff line for choosing a field purely on an ISR basis, but they play the sort of schedule I'd like to see more teams play, so I'm glad they're in. Add in the local flavor, and this is one I'm looking forward to.
W-L pRPI ISR S Reason Probs
1 Stanford 42-14 2 1 1 - 75/64/20
2 Long Beach State 32-20 15 4 1 C 22/16/ 2
3 Texas 34-24 50 36 3 - 3/ 1/ 0
4 Marist 33-19 157 194 4 - 0/ 0/ 0
Ouch! The thing about this is that the committee showed some signs of hope this year. They correctly got CSUF, Stanford, and USC into the top four overall seeds, showing that they recognize what the West has done this year. But then they go and send a team that's a legitimate #1 seed to one of the best #1 hosts. They were at least on the ball enough to give Texas a bid despite an inaccurately subpar RPI, but that's a minor thing compared to reducing the chances of Stanford winning the national title the way they've done.
This regional starkly points out one of the major problems with the current process -- the committee is explicitly not allowed to seed teams within the seeding levels. That means that the #17 team in the country, as best as they can tell, is to be considered the same as the #32 team, and that's a considerable difference.
Stanford unfortunately is fighting a curse this year; fortunately they'd probably be good enough at logic to ignore it if they knew about it. The team that entered the tournament ranked #1 in the ISR's has never made it to Omaha, much less won the whole thing -- the last two lost in the supers, which might bode well for South Carolina if it meant anything.
W-L pRPI ISR S Reason Probs
1 Central Florida 49-12 28 33 3 EW 34/ 4/ 0
2 South Carolina 44-17 4 14 1 C 60/15/ 1
3 The Citadel 38-21 56 78 4 ? 6/ 0/ 0
4 Princeton 22-23 117 178 4 - 0/ 0/ 0
I've been disappointed in the TAAC overall this year; I expected better things from a couple of the teams after the good postseason last year. However, since reputation can become poll rankings which can become undeservedly high seedings, maybe their success can become a self-fulfilling prophecy the way East Carolina's has.
W-L pRPI ISR S Reason Probs
1 Tulane 50-10 19 7 1 - 61/37/ 3
2 Mississippi 38-21 16 22 2 - 22/10/ 1
3 Oklahoma State 39-20 41 30 2 C 16/ 6/ 0
4 Southern 41-10 96 133 4 - 1/ 0/ 0
This is a really nice regional, one even I can't find much to say bad about. You could make a case for OSU as a #2 seed, but they're close enough to the bubble there that there's really nothing wrong with this.
W-L pRPI ISR S Reason Probs
1 Louisiana State 40-19 3 12 1 - 45/23/ 1
2 Virginia Commonwealth 38-17 35 60 - RN 7/ 1/ 0
3 California 33-23 46 16 1 RW 40/20/ 1
4 Minnesota 35-19 49 64 4 - 7/ 1/ 0
Well, much like Nebraska, they tried to make it easy for the host, but they messed it up. Between California and Tulane, this is probably the toughest road LSU has had to Omaha in a year they were in the top 10 in the polls in over a decade.
Note to self: Stating in public that LSU has a 1% chance of winning the title going into the tournament isn't likely to make your mailbox a peaceful place this week.
W-L pRPI ISR S Reason Probs
1 Miami, Florida 38-12 2 11 1 - 70/52/ 5
2 Florida 33-25 17 40 3 R 20/11/ 0
3 Stetson 40-15 48 59 - NW 10/ 4/ 0
4 Bucknell 27-14 165 186 4 - 0/ 0/ 0
Some time this summer I'm planning to write a column or two about the mathematical underpinnings of why the RPI's stink so bad (yes, I write columns year-round; don't go away when the season ends). I'll probably use Florida as a sample case. I applaud the committee for taking them; it's nice to see that particular piece of unfairness end, at least for a year. However, they're a #3 seed, not a #2. Stetson's a borderline at large pick at best, so the Hurricanes get off a little easier than they probably deserve this year.
W-L pRPI ISR S Reason Probs
1 South Alabama 44-17 26 34 3 W 49/17/ 0
2 Clemson 38-20 22 42 3 R 40/14/ 0
3 William and Mary 35-18 45 86 4 R 9/ 1/ 0
4 Seton Hall 32-21 74 129 4 - 2/ 0/ 0
Now, this is ugly. I mean, really ugly. Every team in here is over-ranked by the RPI's. Note to the committee: If you set up a regional that Seton Hall has a 2% chance of winning, you're probably doing it wrong.
W-L pRPI ISR S Reason Probs
1 East Carolina 44-11 6 19 2 R 70/37/ 2
2 Winthrop 46-14 25 48 - R 21/ 7/ 0
3 South Florida 32-29 33 69 - R 10/ 2/ 0
4 Maryland-BC 31-19 187 212 4 - 0/ 0/ 0
Winthrop is another team that wouldn't have been in a purely ISR-based field, but they were close, and they're not a bad choice. That said, they're a #3 seed, not a #2. Add that to the annual East Carolina over-seeding (I wonder what the move to C-USA will do to their RPI; that's worth studying) and you have a really unattractive regional. On top of that, you have a balanced but unimpressive field in Knoxville, and you guarantee that somebody yawn-worthy is going to Omaha. Note that nobody in either regional tops a 2% chance of winning it all.
W-L pRPI ISR S Reason Probs
1 Wake Forest 41-16 13 26 2 RT 34/17/ 1
2 Tennessee 41-17 10 13 1 T 50/31/ 2
3 Middle Tenn. State 36-15 57 44 3 - 15/ 6/ 0
4 Tennessee Tech 28-27 143 134 4 - 1/ 0/ 0
As I said above, solid but unimpressive. The top two seeds are flipped due to the committee paying too much attention last week, and the Brazelton factor may make that matter more than it should, but it's a reasonably fair setup.
W-L pRPI ISR S Reason Probs
1 Southern California 38-16 9 3 1 - 64/51/17
2 Pepperdine 41-16 36 6 1 RN 26/17/ 3
3 Fresno State 39-23 77 25 2 R 8/ 3/ 0
4 Oral Roberts 44-11 83 52 3 - 2/ 1/ 0
Sigh. %#@#$@#$%$! Sigh. I was hoping that Pepperdine might somehow luck into the top seed they deserve, getting sent back east to beat up on Ohio State or somebody. As it is, the probable loss to USC means that they end up having a really great season mostly ignored on the national level; after all, if they couldn't win a regional, how good could they have been? By keeping all of the second-tier Western teams, most of whom are better than most of the first-tier Southeastern and Southwestern team this year, out west for the regionals, the committee has assured that there won't be more than three Western teams in Omaha in a season when there should legitimately be five or six. Even UCSB, who escaped to Notre Dame, will have to come back to Southern Cal for a super-regional if they win.
W-L pRPI ISR S Reason Probs
1 Notre Dame 45-11 14 39 3 R 25/ 4/ 0
2 Cal-Santa Barbara 31-15 24 8 1 RW 62/22/ 3
3 Florida International 40-18 30 58 - R 10/ 1/ 0
4 Wisconsin-Milwaukee 36-15 99 106 4 - 3/ 0/ 0
Second note to the committee: If the #1 seed has only a 25% chance of winning, you've probably got something wrong.
W-L pRPI ISR S Reason Probs
1 Georgia 38-17 8 15 1 - 55/30/ 2
2 Georgia Tech 41-18 21 32 2 - 28/11/ 0
3 Coastal Carolina 40-18 40 68 3 - 9/ 2/ 0
4 Georgia Southern 41-18 61 74 - W 7/ 1/ 0
At first glance, this looks legit. All of the teams are seeded correctly, assuming that you let Georgia Southern in for some reason in the first place. Then you notice that this is one of the worst #1 seeds, one of the worst #2 seeds, one of the worst #3 seeds, and a team that shouldn't have been in. It's better than most of the other regionals, but it still leaves a bit to be desired.
W-L pRPI ISR S Reason Probs
1 Florida State 43-17 7 17 2 RN 54/31/ 2
2 Auburn 35-19 12 18 2 - 40/23/ 2
3 Jacksonville 37-23 37 75 - R 5/ 1/ 0
4 Bethune-Cookman 22-29 195 213 4 - 0/ 0/ 0
This is a potentially nice two-horse race. Notice how much getting to play the #4 seed in the first game is worth: Teams that are back-to-back in the ISR's end up with about a 15% chance difference in their potential to win the thing. If Auburn gets by Jacksonville, though, most of that goes away, so the weekend could be interesting.
The Mistreated
Teams that would have gotten in if the ISR's were used to select the field, and the probable reason they were omitted:
W-L pRPI ISR S Reason
UCLA 29-25 43 20 2 WC
Cal State Northridge 28-20 55 23 2 RWC
San Jose State 32-22 86 27 2 R
Oregon State 29-24 68 28 2 RWC
San Diego 30-20 97 35 3 R
Arizona 32-23 75 37 3 R
Cal Poly 27-23 58 38 3 RWC
Alabama 30-23 32 41 3 C
Nevada 30-26 90 43 3 RC
Washington 23-20 123 45 3 RWC
Note that even I would prefer to have Houston and Winthrop in over Nevada and Washington, so I'm not arguing against some discretion being allowed on the part of the committee. Unfortunately, there are no checks on the discretion, and essentially no accountability for bad choices, so we're left with the field we have.
| Boyd's World-> Breadcrumbs Back to Omaha-> The 2001 Tournament Field | About the author, Boyd Nation |