Big Ten has the best set of coaches when it comes to tourney performance

This is a guest post from Pete Tiernan, who has been using stats to analyze March Madness for 19 years. His insights into the NCAA basketball tournament can help you build a better bracket. We're hoping he can help us understand why 12 seeds dominate 5 seeds. E-mail him here or visit bracketscience.com.

On the Big Ten Network's telecast of the Michigan State-Penn State game, commentator Tim McCormick said something that got me to thinking. He proclaimed that the Big Ten had the best coaches of any conference in the country.

Strong words. Fighting words, in fact, among fans of the ACC, Big East and SEC. At first I passed it off as just the kind of thing a former Michigan player—and possible “Big Ten homer”—might say. Then I decided to see if McCormick was right, at least as it pertained to performance in the NCAA tournament since 1985, when the tourney expanded to 64 teams. After all, the big dance is where coaches really make their reputations and build their programs.

So here’s what I did: I grouped all the coaches from each of the Big Six conferences and compared their cumulative performance in the tourney based on win/loss records, accomplishments, and, most importantly, performance against seed expectations, or PASE. I wanted to see which conference had the stable of coaches that most exceeded the number of wins that their seeding suggested they should achieve.

A quick word about PASE: We the PASE statistic compares a coach’s actual win total to the amount he should’ve won based on his seed. Over the last 24 years, each seed has recorded an average number of wins per tourney. Top seeds win 3.42 games each tourney, two seeds win 2.41, three seeds win 1.82 and so on. When a top seed wins only two games, it underperforms by 1.42 games—despite a solid 2-1 record and top-ten caliber .667 winning percentage. If you add up a coach’s year-by-year performance, you can arrive at the average games per tourney it deviated from the expected victory total. Let’s take Tom Izzo as an example. In his 11 appearances, Izzo should’ve won 18.3 games based on seeding; he actually won 26. That’s about 7.7 games above expectations for a PASE of +.704. That’s very good. In fact, of coaches with at least 10 tourney appearances, only Rick Pitino has a better PASE (+.766)

So back to Tim’s claim that the Big Ten has the best group of coaches: After crunching the numbers, guess what? I agree with Tim. Check it out:

The 11 coaches in the Big Ten have been the biggest overachievers of any set of Big Six conference coaches. Based on the seeding positions of their 62 tourney appearances, the Big Ten coaches should’ve won 83.3 games. They actually won 105 games, which works out to a hefty +.349 PASE—more than a third of a game per dance. Just as importantly, the Big Ten coaches have beaten seed expectations in 35 of their 62 tourney appearances, for a seed overachievement rate (SOAR) of 56.5 percent.

Heck, all you have to do is look at the individual PASE values of the 11 Big Ten coaches to see that their a strong bunch of overachievers. You might think that Tom Izzo’s +.704 PASE would lead the way, but John Beilein’s +.865 PASE is the highest among all active coaches with at least four tourney appearances. And Todd Lickliter owns a gaudy +1.172 PASE from his two seed-defying runs with Butler. Then there’s Bruce Weber (+.410), Tubby Smith (+.351), Thad Matta (+.328) and Bo Ryan (+.126)—all tourney overachievers. In fact, even the conference’s underachieving coaches aren’t exactly embarrassing. Matt Painter’s PASE is -.135, Tom Crean’s is -.152, and Ed DeChellis’s is -.146. Only Bill Carmody’s -.352 PASE is significantly below seed expectations. Put it all together, and it’s no wonder that the Big Ten has pulled off some surprise wins this season, despite not being the nation’s most talented conference.

Compare the 11 Big Ten coaches to the 16 Big East coaches, the next best overperforming group. Their +.167 PASE constitutes about a sixth-of-a-game per tourney overachievement—nothing to sneeze at, but nowhere near the solid performance of the Big Ten. What’s more, Big East coaches have overachieved in only 43 of their 98 appearances, for a more unreliable ROAR of 43.9 percent. Granted, Big East coaches have reached the Final Four more than Big Ten coaches (12 to eight) and won more championships (four to two). But in terms of their ability to defy seed expectations, the Big Ten is solidly ahead of the Big East.

The same goes for the ACC. Yes, ACC coaches have compiled a better record (.678 to .636), more Final Four appearances (16 to eight) and championships (five to two) than the Big Ten coaches. But their PASE is just +.136. That’s actually lower than the +.160 PASE that the conference itself—including retired coaches—has achieved. Besides, Mike Krzyzewski and Roy Williams have accounted for 16 of the Final Four appearances and four of the crowns all by themselves. And the ACC coaches’ average seed position has been 4.54, almost two seed positions higher than the Big Ten coaches (6.27). So you’d expect them to accomplish more. Finally, ACC coaches have beaten seed expectations only 44.6 percent of the time—12 percent lower than that of their Big Ten counterparts.

The only other stable of coaches that has overachieved as a group is the SEC. In 46 appearances, the SEC coaches are 3.4 wins above seed expectations, for a PASE of +.075. That’s decent…but this isn’t: SEC coaches have overperformed in only 19 of their 46 tourney trips. That works out to a ROAR of just 41.3 percent—the lowest of any coaching group among the power conferences. In fairness to the SEC, it’s a pretty inexperienced bunch. On average, the 12 coaches have just 3.83 tourney appearances under their belts.

The SEC isn’t the greenest group of coaches, however. That distinction belongs to the Big 12. The conference’s 12 coaches have just 41 appearances among them, for an average of just 3.42 per coach. What’s more, Big 12 coaches have failed to live up to seed expectations. Granted, they’re only about one win away from their 48 seed-projected win total. But it’s underachievement nonetheless. Big 12 fans can take solace in one fact, though. Their coaches have overperformed with greater regularity than any other stable of Big Six coaches except the Big Ten. In 20 of their 41 appearances, Big 12 coaches have defied seed expectations, for a solid 48.9 percent ROAR.

And that brings us to the Pac-10 coaches. The conference itself has not fared well in the tourney. The 98 Pac-10 squads that have entered the dance since 1985 own an underachieving PASE of -.176. That’s not just the worst among Big Six conferences; it’s the worst of any conference, including the Mid-Majors and Smalls. This is all a long way around offering Pac-10 fans a little consolation prize. Yes, your current stable of coaches is the weakest tourney performers among the Big Six conference coaches. But the PASE of Pac-10 coaches is just -.083, meaning that this current group is better than the usual Pac-10 group. And here’s one more thing to crow about: Pac-10 coaches have overperformed against seed expectations in 20 of 44 appearances for a ROAR of 45.5 percent. That’s third best among the Big Six, better than the Big East, the ACC and the SEC. Not too shabby. Of course, this stable of Pac-10 coaches has yet to win a championship among them—and no other conference is saddled with that stigma. So let’s not get carried away.

Bottom line: as far as tournament performance goes, Tim McCormick is absolutely right to say that the Big Ten has the best group of coaches of any conference in the country. You can debate whether the tourney ought to be the final word on this debate, but there’s no debating that Big Ten coaches have been the biggest overachievers of the modern tourney era. Let’s see if they can translate their collective performances into greater representation in the late rounds of the 2009 dance.