Tuesday, March 23, 2010

Duel of the Jews, Basketball Edition

Before you read Stewart Mandel's claim that the term "mid-major" should be retired, here are four statistics for you:

1. No team from outside of a BCS conference has won the national championship in college basketball since UNLV twenty years ago. (The Tarkanian-coached UNLV teams had, uh, certain attributes that make them distinct from any current program. It appears that the mid-major who came closest to ending the drought - 2008 Memphis - also shared some of those attributes.)

2. Here are the seeds of the last 20 NCAA champions:

2009 North Carolina - 1
2008 Kansas - 1
2007 Florida - 1
2006 Florida - 3
2005 North Carolina - 1
2004 UConn - 2
2003 Syracuse - 3
2002 Maryland - 1
2001 Duke - 1
2000 Michigan State - 1
1999 UConn - 1
1998 Kentucky - 2
1997 Arizona - 4
1996 Kentucky - 1
1995 UCLA - 1
1994 Arkansas - 1
1993 North Carolina - 1
1992 Duke - 1
1991 Duke - 2
1990 UNLV - 1

14 #1 seeds, three #2 seeds, two #3 seeds, and one #4 seed. '88 Kansas, '85 Villanova, and '83 N.C. State are in the distant past and even those three noted Cinderellas come from the Big Eight, the Big East, and the ACC. If you want a good piece of evidence for my notion that the NCAA Tournament should be 16 teams instead of 64, this is it.

3. We are one year removed from an NCAA Tournament in which all of the 1, 2, and 3 seeds made the Sweet Sixteen.

4. We are two years removed from a Final Four that featured four 1 seeds.

So yes, Stewart, in light of that evidence, the distinction between schools that belong to the six BCS conferences and those that don't is "mystical." Mandel shows this bizarre ability to acknowledge reality:

As we move into Stage 2 of this championship, the favorites will remain those schools with the number 1 in front of them -- Syracuse, Kentucky and Duke. None were remotely threatened in their first two contests, and surely, we must assume, the clock will run out at some point for our friends at Saint Mary's, Cornell and Northern Iowa. It always does.

and then forget it a few paragraphs later:

But when the "best teams" aren't all that different than "the next-best teams," upsets happen. Here's guessing we've hardly seen the last of them in this, the nation's most egalitarian sporting event.

Mandel's line of thinking annoys me in two separate ways. First, it is a really good illustration of the recency fallacy. He takes one weekend of results and makes broad conclusions, completely ignoring a bevy of relevant facts that demolish his hypothesis. The clear trend in recent years has been for chalk to prevail in the NCAA Tournament. Northern Iowa beating Kansas and St. Mary's beating Villanova doesn't change that. Second, Mandel is feeding the media narrative that the NCAA Tournament is exciting and unpredictable. It may be exciting for people who want to reduce a season to three weeks, but the unpredictability factor is seriously overrated. Yes, there are upsets every year in the first two rounds (some years more than others), but at the end of the road, there is a team from a major school standing in the falling confetti in a dome somewhere. We like to imagine that the NCAA Tournament embodies the American Dream and that the little guy can go from nothing to glory, but just about all of the time, the little guy ends up face down in a pool with "The World Is Mine" rotating above his head.


Anonymous said...

That's awfully disingenuous of you. Why didn't you go back to the 2006 Final Four, which consisted of a #2, #3, #4, and #11, all of whom beat a #1 seed? Because it completely undermines your point?

If college basketball were college football, Kansas would be in a [dreary] Title Game, and Northern Iowa wouldn't even field a team at Kansas' level.

chg said...

No, it doesn't. The #2, #3, and #4 all reinforce that by the Final Four stage, the teams standing are all top ranked teams from power conferences. The 2006 champion is listed right in there with all the rest.

If you want to hang your hat on 1 of 80, knock yourself out.

Michael said...

Seeds of final four participants:

2009 - 1, 1, 2, and 3
2008 - 1, 1, 1, and 1
2007 - 1, 1, 2, and 2
2006 - 2, 3, 4, and 11
2005 - 1, 1, 4, and 5
2004 - 1, 1, 2, and 3
2003 - 1, 2, 3, and 3

Now tell me, which of these groupings is not like the other?

If the Big Dance were a 16-team tournament, there are exactly two Final Four participants that would have missed the field over the past seven years: George Mason and Michigan State. Both lost their semifinal by double digits.

Anonymous said...

CHG, LSU was not a "top ranked team" in 2006. They were the 18th or 19th ranked team when the tournament started.

Aside from that, listing only the champions and not the other participants in the Final Four is misleading: just in the past 7 years we've had a Colonial Athletic Association team and three Conference USA teams in the Final Four. Seedwise, in individual years, even where the winner was a high seed the Final Four has been a hodgepodge (for example 2000: #1, #5, #8, #8).

Having a large field has reduced the influence of inculcated tradition and reputation and opened up the game for fans of hundreds of schools that aren't in the BCS conferences. There's no way that Northern Iowa would have been in a 16-team tournament (they weren't even in the AP top-25 coming into the tourney), yet they were better than the #1 overall seed. Its nuts that people continue to support systems that wouldn't have included them.

Anonymous said...

Indiana (#5 seed) won a semifinal game in 2002, and you can't really assume that the #4 seeds would have been in a 16 team tournament. The selection committee moves teams in the 4-7 range around pretty fluidly based on matchups and location. It's very unlikely that LSU would have been in a 16 team tournament in 2006.

But you should make a killing gambling on the tournament if it's so predictable.

chg said...

Anon, beginning with this year's tourney and going through 2014, lets bet on the wild unpredictability. Track the Final Four teams. You get three quarters of the field - every team seeded below 4th. I get the top four seeds in each region.

We'll play for $500 per Final Four participant and pay out at the end of the 2014 tournament.

I eagerly await your acceptance.

Michael said...


1. What's the objective of a team in the NCAA Tournament? To win, no? That's why I focus on the teams that ultimately win the thing. But even if you go down to the Final Four level, you still have a large majority of chalk making it and the underdogs that do make are almost invariably knocked out quickly.

2. Having a large field has killed interest in college basketball before March. When I was growing up, I lived in Charlottesville during the Sampson era. I remember the December game between UVA and Georgetown (neither of which were traditional powers at the time) in 1983 being on the cover of SI. When is the last time that a December college basketball game registered like that? The last one that I can remember was Wake-UMass in '96 when it was Duncan vs. Camby and UMass was unbeaten.

3. If the comittee moves teams between the 4 and 7 seeds fluidly, then why are four seeds so much more successful in the Tournament than 5-7s? Also, if what you're saying is true (which I doubt), then what the hell is the point of the regular season when teams 13 through 28 are thrown in a hopper and spat out based on commercial considerations?

4. You see, there are these things called point spreads that take into account the fact that favorites are likely to win...

Anonymous said...

College basketball in general is loss popular, and you unwittingly identified the source: the lack of transcendent stars, due to some combination of early entry and style of play changes (slower games, more positional defense and flopping) and officiating. Without looking it up, I can think of at least at least three other regular season cbb SI covers (don't remember if they were December or not): Kevin Durant, Greg Oden, and Jameer Nelson. I think there were almost as many for high school hoopsters (Telfair, LeBron). That doesn't have anything to do with the tournament, though.

Same with December games. Illinois-Wake Forest was huge in December 04 (Paul vs. Williams), North Carolina-Kentucky the same year was massive. In December 2006, Ohio St. played both UNC and Florida in pretty massive games. Aside from that, there were 52 teams in the tournament when Sampson played. The extra 12 haven't altered things that much.

Useful Talk said...

I truly value this superb post that you have accommodated us. I guarantee this would be helpful for a large portion of the general population. basketball legends