What the Number Geeks Would Say About Duke and Mercer

Photo By Norm Olson, copyright 2013

Photo By Norm Olson, copyright 2013

Don’t think too deeply about why Mercer beat Duke. They were close enough in talent for a Mercer win to be as probable as catching a cold on a coast to coast flight, which isn’t likely, but hardly improbable. End of discussion, the stat geeks would say. Move on.

Today’s article doesn’t try to unearth a story buried on Page Seven, rather it provides a minority viewpoint on Duke’s elimination by “lowly” Mercer. The perspective is of a statistician, or numbers geek. On the heels of Money Ball and Adam Silver’s bravura performance during the 2012 elections, that viewpoint intrigues many people.

Suppose Mercer and Duke played in a 10 team conference, with all teams of top,  25 coaches’ poll caliber, each playing exactly the same 27 game schedule. They play every other team 9 times.  Suppose, further, that Duke had a .700 season record in that league, and Mercer, .500. That’s my wild, but,  I think,  plausible, guess of how much better Duke is than Mercer, if they played identical schedules in a good league. That’s clearly a supposition, but I need this device to help me provide a few observations. The basic math behind this can be found in any good stat 101 course, and here in a watered down, but accessible, discussion.

Under this scenario, the stat geeks would tell us that Duke has about a 70% chance of winning any particular game against Mercer, on a neutral court. That may sound high, but it also means that if Duke and Mercer played 9 times in this imaginary conference, you would expect Mercer to win three of the games. Those are really not bad odds for Mercer. So, Mercer’s win – if you accept my .700 versus .500 premise — doesn’t mean Mercer is better than Duke. It merely means that the March 21st game was one of those three occasions out of nine that Mercer would be expected to win.  End of discussion.  Move on. Good thing for ESPN that it doesn’t hire stat geeks to do the color analysis.

This is of course little consolation to Coach K or the Duke players; and even less solace for the alumni and boosters. The next time Coach K talks with the University President, we can be sure he won’t use the stat geek argument as an alibi. But it’s still a good explanation.

Ten or twenty years ago, the difference between Duke and a Mercer-like school in the tournament would have been more like .800 versus .300. That, according to the stat geeks, would translate roughly into more than a 90% chance of Duke winning any particular game; or Mercer winning one game, at most, out of 9, against Duke. Yes, a decade or two ago, it would have been very unlikely for a Mercer to win a game against the Blue Devils.

What has changed? Some in Duke nation are thinking Coach K has a lost a step or two, or become a bit (sub consciously) complacent. It’s called losing the edge; not being hungry enough anymore. Maybe his half time exhortations have lost some zip; or his attention to fine detail has withered. It doesn’t take much to move the needle by a lot in the fast lane.

The stat geek would look at this differently. It all has to do with the gap between Duke and Mercer closing from (perhaps) .800/.300 to .700/.500. This hasn’t happened because Coach K is a worse recruiter or has lost a few IQ points. A lot of the change has, of course, to do with the “one and done” trend in college basketball, with the top teams losing players to the NBA after one or two seasons. This affects Duke much more than Mercer.

Less obvious is the impact of the major media coverage that a Mercer and its players receives today, compared with ten or twenty years ago; from multiple ESPN channels and radio sports talk shows, to a raft of blogs that scout and rate players from every top 100 school. There is an amazing amount of detail that’s easily found on just about any player or team in the country.

In the past, a high school kid with offers from Duke or UCLA wouldn’t give a thought to a small, niche program, even if it meant riding the bench at Duke for a few years. If the kid was worried about getting lost in a large pond, he’d at least wind up at a Marquette; small, yes, but an established basketball school that’s been respected for a long time. Today the opportunities for attention are vast. The kid, and his parents, with NBA dreams would look seriously at Mercer, Butler, Gonzaga, George Mason, and others, knowing he will be show up on the NBA’s radar screen.

What about the fate of Coach K? The college basketball bloggers all say Duke’s latest recruiting class is first class. The program is in fine shape, as far as we can tell. Coach K has a lot of integrity, so I would not be the least bit surprised if he raised the possibility with the College President of moving on (Coach K, that is, not the Prez). I can’t imagine the President raising it.

I know a few ardent and knowledgeable Duke fans who argue the team was not well prepared for Mercer and that Coach K was unimaginative and lackadaisical. So, if Coach K offers to move on, the Prez could bite, if he’s heard enough (sacrilegious) barbs about the Coach from influential boosters.  As a stat geek, I am putting a probability around that of about 10 or 15 percent. That’s low, but not negligible. Then, what happens to Coach K? Maybe Phil Jackson, the new New York Knicks GM, goes for a different “K” than predicted – Krzyzewski instead of Kerr. Not very likely, but fun to think about. It would give fans and writers enough material to keep themselves busy and amused during the five years it takes to turn things around in New York.


2 thoughts on “What the Number Geeks Would Say About Duke and Mercer

  1. Alex MacLachlan

    That’s the great thing about March Madness. There are so many college basketball teams, you never know when you’ll get a “Hoosiers” moment, when an unknown ” David” takes down a “Goliath”. They are all 20 yr old kids dreaming of their “Big Chance” that may only turn into their 15 minutes of fame, but what a beautiful 15 minutes it was. BTW: Go Aztecs


    1. Irv Lefberg

      Absolutely . I agree. The Tourney is just about the best thing going in sports. The stat geek might agree too, if he chilled out. But he’d add that the winner of the Tourney isn’t necessarily the”best” team in the country. Indeed, it hardly ever is, except maybe in the old days when we had truly dominant teams, like UCLA. But, even if the winner isn’t the best, the winner is just about always a very very good, maybe even, great team — because you can’t win six in a row against stiff competion by luck. Probability of mediocre team doing that is extremely low.



Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s