March Madness Primer: Observations Based on Advanced Stats

Rise and shine: We’re close to March Madness time. With conference championships beginning this week, and Selection Sunday only a Sunday away, I figured I was due for my first college basketball article in…well, ever.

Admittedly I’m not a big college basketball fan, yet I’m almost always engaged come tournament time. This year, however, is a bit different. I don’t really care. Three reasons why.

For starters, I don’t have a team. Kind of hard to get overly invested in college basketball without a specific rooting interest. This brings me to reason number two: It’s even harder for a Bostonian  to get overly invested in college basketball without a specific rooting interest. I mean, I’m still on a high from four weeks ago. How was I supposed to set aside time on Saturday’s this year when the Patriots-revenge tour was taking place every weekend? Plus, with Isaiah Thomas energizing the Celtics and the Bruins starting to peak at the right time, my rooting obligations continue to steer me away from the college game.

I’m also in London right now. Based on early indications of college basketball’s relevancy over here, I hardly expect British sports fans to stay up past midnight at bars watching the likes of Kentucky or Louisville capture a soon-to-be-vacated national title–if it’s even on TV. For that reason, I don’t expect to stay up well into the early morning to see how this college basketball season shakes out. It just doesn’t interest me enough. I’m praying, though, that I at least won’t miss out on moments as epic as this:

 

 

Or this, of course:

 

 

 

I can miss out on watching March Madness for one year, though, right? OK, based on those videos, maybe not. I’m still at least going to follow  it, though–let that be clear. So with that in mind, I figured I’d do my own investigation as to which teams might take home this year’s title–and which teams might not–as well as some ideas for the most likely sleepers by combing through the advanced stats at our disposal.

Here are six things I learned.

 

Kansas is overrated

The number one ranked Jayhawks seem to have that “it” factor. After all, they’re a staggering 12-2 in games decided by two possessions or less this season.

Or, maybe they’ve just been getting lucky. That’s what Ken Pomeroy’s trusted analytic system seems to think, as he has Kansas rated as only the ninth best team in the country, in large part due to their high ranking in Pomeroy’s “Luck” category. (Quick note on the legitimacy of Pomeroy’s college basketball power rankings: Villanova and UNC, last year’s finalists, finished 1-2 in his rating system last year, while he has correctly identified undervalued teams such as George Mason (2006) and Wichita State (2013) that would later go to the Final Four.)  

Back to Kansas and luck. Pomeroy defines luck as “a measure of the deviation between a team’s actual winning percentage and what one would expect from its game-by-game efficiencies.” The Jayhawks are obviously a very good team, but their success in close games is likely unsustainable. I also think casual fans might be biased toward Kansas based on their awareness of a couple standout players, namely senior point guard Frank Mason III and freshman phenom Josh Jackson. Taken together, I think Kansas is a shakier favorite than people realize. And despite securing 59 out of 65 first place votes in the latest Top 25, they could be one of the more upset-prone overall number one seed’s entering the tournament in recent memory.

 

Gonzaga is great…because they haven’t played anybody 

Perched atop this year’s Top 25 for most of the season was that little school from Spokane, that is until they fell for the first time to an unlikely source, the BYU Cougars. Nonetheless, Gonzaga–ranked fourth in this week’s poll–is still a worthy team. Another Sweet Sixteen, or maybe even an Elite Eight appearance, looks to be in the cards.

That first trip to the Final Four that everyone has been waiting for, however, is once again a stretch. The reason is simple: Gonzaga hasn’t played anybody. Of course  you’re going to have an inflated record when you play the 146th toughest schedule in the country. And the reason why this is concerning for the ‘Zags–like it is ever year–is because the averaged Strength of Schedule (SOS) ranking for Final Four teams over the past five years is 13.3 out of 351. Translation: It pays to be tested. Feel free to pencil Gonzaga in for at least the Sweet Sixteen, but find another team to come out of their region.

 

West Virginia is criminally underrated

Checking in at number three overall in Pomeroy’s aforementioned analytical power rankings is West Virginia. At 24-7, however, you’ll find that they are only ranked #11 in the latest poll. Despite this discrepancy, West Virginia is a great sleeper pick to reach the Final Four. The main reason why is because they boast one of the best defenses in the country. As customary with Bob Huggins’ teams, West Virginia has forced the most turnovers per game in the nation this season (21.2). Remarkably, the team that has forced the second most turnovers in the country, South Carolina, is equally close to 105th ranked St. Francis (NY)–is that even a school, by the way?— as they are to West Virginia. Their excellence in creating turnovers largely explains why the Mountaineers are ranked 5th in Pomeroy’s defensive efficiency rankings. And it’s their high ranking in that category alone that is very revealing: National champions over the past five years have finished an average of 9th overall in Pomeroy’s defensive efficiency metric, including last year’s Villanova squad, which also finished 5th.

Other underrated squads: Virginia (1st in defensive efficiency and 5th overall despite being ranked 21st in the

AP poll); Wichita State (projected as an 8 seed in Joe Lunardi’s latest bracket projection, but is top 10 in overall efficiency)

 

Keep an eye on Oklahoma State: They have the best offense in the country

Another thing that caught my attention when analyzing Pomeroy’s statistics: Oklahoma State, who is currently unranked but projects to make the NCAA tournament, has the nation’s top adjusted offensive efficiency rating despite playing the  hardest slate of opposing defenses. To make a cross-sport analogy, perhaps this stat might bode as well for the Cowboys as it did for the Atlanta Falcons, who also had the NFL’s most efficient offense despite playing one of the toughest defensive schedules last year.

It also appears that Oklahoma State is peaking at the right time. Looking back through their schedule this year, they clearly hit a rough patch in early January when they lost six consecutive Big 12 contests. But they’re 10-3 since then, with a win over West Virginia on the road under their belt. Even their losses haven’t been too bad: Each defeat has come by no more than five points, and all three were against ranked opponents such as Kansas and Baylor. Take note of the Cowboys when filling out your bracket.

 

It's not a secret anymore, Lonzo. You and your Pac-12 friends aren't that great.

Sell your Pac-12 stock: UCLA, Oregon, and Arizona aren’t who you think they are

Three Pac-12 teams are clustered in the top seven of the latest poll: #3 UCLA, #5 Oregon, and #7 Arizona. But all are suspect, as far as higher seeds go. Lonzo Ball’s UCLA squad stands out in particular. The good news for the Bruins is that they can score: They lead the nation in both points per game and true shooting percentage. The bad news is that they can’t play defense: They rank just 238th in opponent points per game. That’s OK, though, because the Bruins are so good offensively that they’ll just out-score people, right? Well, that brings me to the even worse  news: UCLA’s high offensive rankings are deceptive because they have played a mediocre crop of opposing defenses (109th toughest, in fact). A quick run-down of some of their Pac-12 competitors’ defensive rankings: Arizona 27th, Utah 66th, USC 95th, Washington 251st–you get the picture.

Oregon and Arizona are in similar positions to UCLA in the sense that they haven’t played great competition, either. The difference between those teams, however, is that while they aren’t as potent as UCLA, they play better defense. In all, each are highly questionable national title contenders. Each may be awarded a 1 or 2 seed by the selection committee, but they’re overall rankings according to Pomeroy’s efficiency metrics (UCLA is 16th, Oregon 17th, and Arizona 21st) suggest that they will be overvalued by the selection committee. Perhaps one of them will make a run to the Final Four, but I’m selling the group as a whole.

 

So who’s the favorite?

Tough call, given that there are no juggernauts this year. But little has changed since Jay Wright’s squad emerged last March: Villanova is still the team to beat. The Big East might not be the nation’s strongest conference anymore, but the Wildcats have had another successful season replete with numerous impressive, double-digit victories. Villanova is also one of the more balanced teams in the country, as they rank a terrific 3rd and 16th in offensive and defensive efficiency respectively. They project to face stiff competition from heavyweights such as North Carolina and Kentucky as well as from sleepers like the aforementioned West Virginia Mountaineers in the tournament, but I think Villanova is a decent bet to be one of the rare teams to win back-to-back national titles.

Posted by Mando

Co-Founder of Check Down Sports. Die-hard Boston sports fan (Patriots, Celtics, Bruins, Red Sox -- in that order). Expert on all things related to the Super Bowl. Proudest life achievement: four-time fantasy baseball champion.

This article has 1 Comment

Leave a Reply