The NCAA tournament said goodbye to one Giant Killer last Friday night, and our statistical model was ready to pack its bags and go home on Sunday, but VCU just blew up our spreadsheets.
Before we get to the details, remember to vote for the 2011 All-Giant Killers team. You can nominate any athlete who played for a potential or actual Killer in this year's tournament; cast your ballots in the comments or via the GK Mailbag.
And now, let's give props to the David who has slain Goliaths like no other underdog before, beating five favored teams as an 11-seed, the Virginia Commonwealth Rams.
Given its regular-season statistics, we estimate the odds of VCU getting this far were a staggering 3,058 to 1. And that raises the question: Should we be evaluating the Rams as a qualitatively different team than the one that entered March Madness?
See, Butler (VCU's next opponent) isn't actually playing all that differently from a month or three ago, even though the Bulldogs have also overcome huge odds to make it to the Final Four. In the regular season, Butler scored 109.7 points per 100 possessions and gave up 98.8. In this tournament, against stiffer competition, that margin is down to 108.0-102.1, which has been just good enough to get by; Butler has won its four games by a total of 13 points.
The Bulldogs have dominated in their favorite phase of the game, preventing opponents from getting second chances. Butler has collected 77 defensive rebounds in the tournament, while allowing other teams to grab just 35 offensive boards. But they've been outrebounded even more heavily (91-42) at the other end of the floor, and haven't been especially impressive elsewhere. It's no insult to say Butler has been fully equal to the task of knocking off three super-Giants in a row while also noting that it enjoyed some favorable bounces (and calls) against Old Dominion, Pitt and Florida. The Bulldogs have been as good as their numbers and lucky, too, which is how most teams advance.
Something different is happening with VCU. To make it from the First Four to the Final Four, the Rams have maintained their own signature strengths while adjusting their style of play to exploit opponents' weaknesses. Throughout the regular season, VCU did an outstanding job of protecting the ball while forcing turnovers, and the Rams have been even better at ball control during their tournament run, giving up the ball on just 13.8 percent of possessions while generating turnovers on 20.5 percent of opponents' possessions. VCU had also been excellent from outside the arc all year, and has stepped up its long-range shooting and perimeter defense to ridonkulous levels over the past 10 days. The Rams are hitting on 43.8 percent of their bombs in the tournament, while holding opponents to just 23 percent from downtown.
But the Rams have also shifted styles according to matchups. Against USC, they crashed the boards and collected 15 offensive rebounds. Against Georgetown, they built a lead, then drew fouls and attempted 39 free throws. Against Purdue, they went inside and shot 29-for-44, or 65.9 percent, on 2-point FGs. Against Florida State, they attempted nearly half their shots from behind the arc, going 12-for-26 on 3s. And against Kansas, they maintained that torrid pace, hitting on 12 of 25 bombs, while shutting the Jayhawks down so completely from outside that Kansas compounded bad shots with panicky heaves and wound up just 2-for-21 (that's 9.5 percent) from long range. When Kansas got close, VCU caught a second wind and put away one of the two best statistical teams in the country.
VCU's huge turnover differential means the Rams end nearly 90 percent of their possessions with shots, as John Gasaway explained in this excellent analysis of how the Rams have become so dominant, versus about 80 percent for their opponents. And their extreme efficiency on the perimeter means VCU's shots are worth much more, on average, than their opponents' shots. That combination explains how VCU was able to get by Florida State even though the Seminoles torched the Rams on the glass -- Florida State grabbed an astounding 45 rebounds, including 20 on the offensive boards last Friday night. Shockingly, the same formula worked against Kansas, a team pretty much built to kill Giant Killers.
Kansas was an even better offensive rebounding team than Florida State, and all season long, the Jayhawks knew what to do with all the extra possessions they collected: jam them down opponents' throats. Except VCU didn't let the Jayhawks keep the ball (Kansas had 14 turnovers), or shoot from the outside or from the inside (where the Jayhawks, after ranking No. 1 in the NCAA in 2-point FG shooting this year, shot under 50 percent against the Rams). In the end, all Kansas had left was its rebounding, but that wasn't enough. Marcus and Markieff Morris had a combined 13 offensive boards, but they missed 18 shots.
As for any ideas that Kansas was rattled by the pressure of avenging last year's loss to 9-seed Northern Iowa, please. The Jayhawks had been playing focused yet chippy hoops until they ran into VCU. In particular, they had limited opponents to shooting just 29.1 percent, third worst in the NCAA, on 3-point attempts. On Friday night, they held Richmond to a horrid 4-for-26 from downtown, smoking the Spiders out of the Alamodome in a matter of minutes.
VCU found a way to go high-risk/high-reward even against the Jayhawks. And if the Rams beat Kansas, who can't they beat? They hit their 3s, they maintain their pressure and, beyond that, they'll match up against you in whatever style fuels your nightmares. Shaka Smart & Co. have taken the classic Giant Killing concept of underdog variability and gone meta with it.
Our model is set up only to predict upset chances by underdogs seeded at least five slots below their opponents. But before Selection Sunday, it called Butler one of the 10 most vulnerable potential Giants. And while the Bulldogs are a fine team, they have holes for VCU to exploit. Butler allows opponents to shoot 48.5 percent on 2-point attempts -- and we've seen Joey Rodriguez make tougher opponents look positively immobile inside all tournament long. Further, the Bulldogs don't force turnovers, and they're one of the worst teams in the NCAA at blocking shots, which means VCU will get its shots despite Butler's strength on the defensive boards.
So assessing the Rams' odds of continuing their Giant Killing spree comes down to figuring out how well they will continue to execute. Which is another way of asking how much faith we can have in the near-crazy numbers they have put up since March 16. Look at the season-long stats for both teams, and you'll come to conclusions similar to those reached by Bracket Predictor or Ken Pomeroy, who both give VCU chances of about 45 percent against Butler, or Vegas, where the Rams are a 2.5-point underdog. But if VCU is playing at a genuinely new level represented by its recent stats, we estimate its odds of moving on to the national championship are actually 58 to 61 percent.
Our model doesn't have vocal cords, but if it did, it would ask: Doesn't VCU's adaptability seem to signal extraordinary intelligence of play, and isn't that a quality that's likely to persist?