Day Four Roundup

Not a terrible day for me picks-wise, largely because it wasn't a great day for anyone. Bradley and George Mason won again, making my losses of Kansas and Michigan State in the first round largely moot since no one had either team going that far. And then I had West Virginia picked right (where many had had Iowa), got the three gimmes (UConn, Villanova, and Texas), and lost on Ohio State like everyone else because I wasn't confident enough to take the Georgetown upset (though I did select it on Alma's behalf as it was one of the ones I had looked at for a while). All told, nobody got more than five out of the eight games right, and I was one of them (unlike yesterday when I had five right but seven people had at least six).

This is a pretty crazy tournament. Let's talk about it.

Upset of the Day: You might say Bradley, but I'd say it was George Mason over North Carolina. Not a single person of 17 even took George Mason in Round One (one of just three 11-13 seeds of whom that was true, the other two being Air Force and Montana), and they were down early - North Carolina opened the game on a 16-2 run, in fact. George Mason wasn't one of the best defensive teams in the nation for no reason, though, and they held Carolina to just 44 points over the last 35 minutes of play while scoring 63 themselves in the same span. Plus, Roy Williams, who seems like one of the most even-keel guys in the game off the court, looked ready to kill half his team for much of the second half. When the higher seed's coach is doing that, it's the sign of a big-time upset brewing.

Non-Upsets of the Day: Is it time to start worrying about UConn and Villanova? I know the argument is that "champions win close games," but these were some serious second-round struggles against teams that underachieved all year. Connecticut probably still makes the Final Four because of who's left in their region - although I probably have as much confidence in Washington to possibly do it as I would have had in Tennessee or Michigan State, and clearly North Carolina could be rattled more easily than you'd have suspected. Villanova, however, might have to face Boston College and (if they win there) Florida. Think UConn got the easiest bracket?

Some Ketchup for Your Crow?: During the UConn-Kentucky game, Billy Packer described Bradley and Wichita State's wins as "impressive victories." Later, Jim Nantz mentioned that many people thought Bradley should not have been in the tournament, including "two people sitting right here." Packer's only response was a barely audible chuckle, which was probably accompanied by a withering stare and some teeth gnashing.

When you get right down to it, Packer (and Nantz as well, to be fair, but Nantz's face doesn't frighten small children) could hardly have been more wrong. Not only does the Missouri Valley have two teams in the Sweet 16, but that's exactly as many as the ACC has, and one more than the Big 12 has, after Packer made such a huge deal out of how the selection committee was implying that the MVC was as good as the ACC and Big 12 because it got the same number of bids and how clearly wrong that was. Meanwhile, two Big 12 teams went out in the first round to mid-majors. North Carolina, the second ACC team to lose, also went out to a mid-major in George Mason (another team that some thought should not have been included in the field). Can anyone honestly say it would have been better to have had Michigan and Florida State in place of Bradley and George Mason? Enjoy the crow, Billy.

Go, big conferences!: For the record, the Big Ten had six teams in the tournament (Ohio State, Iowa, Wisconsin, Michigan State, Illinois, and Indiana) and not one of them made it to the Sweet 16 (with three losing in the first round). The last time a Sweet 16 didn't feature a single Big Ten team? 1996. (Coincidentally, this was also the last time Bradley made the field.) That year, the Big Ten entrants were #1 Purdue, #5 Penn State, #6 Iowa, #6 Indiana, and #7 Michigan - three of the five lost in Round One, with Purdue nearly joining that club as well as nearly becoming the first ever #1 to lose to a #16; Purdue and Iowa won their first-round games by a combined four points and were both eliminated in Round Two. (1996 was also the last time Duke lost in the first round, when they were a mortal #8 seed and went out to #9 Eastern Michigan.)

Who Wants to Be a Two Seed?: As I noted on the Challenge page, this is the seventh year in eight in which at least two #2 seeds failed to make the Sweet 16; it's also the eighth year in the last ten. Back through the 1997 tournament, there have been 40 #2 seeds, and fully 20 have not made it out of the first weekend! That's a pretty woeful percentage, wouldn't you agree? Especially when you consider that in the same time period, #1 seeds are 34-6 in the second round (and obviously none have lost in the first), and in six of the ten years no #1 lost that early at all, whereas at least one #2 went down every year. I guess there's more of a difference between the #1 and #2 seed line than we're inclined to think. Worse still for the 2s, it's not like they're always falling on buzzer-beaters. As this chart shows, only four of the 20 losses have been decided by one possession (and that includes the Iowa State loss to Hampton in 2001, which you would certainly expect to be a close shave), and eight were decided by double figures. (None of the six losing one seeds lost by more than seven, though only two lost by 1-3 points.)

Had I done that statistical analysis before the tournament, I might have been able to talk myself into Georgetown. (Though there's probably an equal chance I would taken Marquette and been screwed.) Sure, it's a little dangerous to base your picks on stats alone - stats tell us that the 8-9 game is more often won by the 9, but this year three 8 seeds won out of four. (Of course, those 8-9 games are usually a crapshoot when it comes to picking anyway.) Still, it's a pretty compelling trend that two #2 seeds lose no later than Round Two in just about every year nowadays. If you don't think that speaks to increased parity in the game, read this stat: between 1985 and 2005, 31 #2 seeds lost in the first two rounds. 18 of those came in the past nine years, with just 13 in the twelve years before that. In other words, it used to be one a year, and now it's two a year. Seems like increased parity to me.

So maybe we shouldn't be quite so surprised when we see results like this year's. Sure, Bradley is the first #13 seed to make the Sweet 16 since Oklahoma in 1999, and the first non-Big Six #13 to do it since Valparaiso in 1998. But that's not really that odd. What would be odd is if Bradley becomes the first #13 ever to make a regional final. What would be odd is if the national champion doesn't come from the top three seed lines (since the field expansion, only three - #4 Arizona in 1997, #6 Kansas in 1988, and #8 Villanova in 1985 - have done it, and twelve of the other 18 have been one seeds). What would be odd is if three mid-major teams make the Elite Eight (Wichita State and George Mason play each other, so there can't be four). But in the end, you know that the national champ will almost certainly be a #1 or a #2. The tournament is always exciting, but parity hasn't hit its full stride just yet.

Another surprise: this post getting more than a couple comments. Sorry it was so long.