Tuesday, April 26, 2011
Shaq vs. Darko: Losing ugly
By Henry Abbott
Steve Babineau/NBAE/Getty Images
The wager: Darko Milicic would contribute more than Shaquille O'Neal.
Shaquille O'Neal is big. Huge, in fact. His seven feet of height or untold pounds aren't what overwhelms this story, however.
It's his fame. His celebrity heft dominated the conversation last summer, when he signed with the Celtics. Even though he is old, injury prone, slow, and hasn't been himself in years, his signing was treated as something that would set the NBA on its ear.
Hardly, I thought.
And to make a point, I sought a less heralded player to compare. After rummaging around in a desk drawer, I found Darko Milicic. Hard to name a player heralded less than that dude, and as a nice benefit, he had just signed a laughingstock contract with Minnesota.
Blogger Tom Ziller took the bet, with the winner to write something flowery about the loser's player. The terms were thus:
Tom and I spoke a couple of days ago and I made clear that there would be no free passes for injuries or anything else. It's cumulative regular season-long contributions at both ends of the court. I'm not saying Milicic will score more, play better D, have a better PER, be more efficient or anything else. I'm saying that over the course of this season, smart analysis will show he'll produce more at both ends of the floor, in total. In other words, if O'Neal's lumbering fouls up Boston's defense, that'll matter. If one or the other gets suspended, injured, rested or anything else, tough.
And that may not be fair in judging the better player, but it's fair when we're talking about the value of signing this or that player -- players who can get on the floor are more valuable than players who can't.
As predicted, O'Neal was old all year, and injured more than half of it. He played just 36 games and had a limited impact on the Boston Celtics and the league at large. Milicic, meanwhile, succeeded in being just 25, and playing a fairly normal number of minutes.
At the end of the season, having spent at least seven seconds enjoying the predicted Milicic romp in the cumulative box score stats, I e-mailed Ziller to find out precisely how he'd like to concede -- all at once in a blaze of glory or only after first exercising his right to have the decision referred to a panel of experts?
Ziller is nothing if not chock full of fight, and called for the committee. I sent the question off to the experts in the TrueHoop Stat Geek Smackdown, and sat back and waited for them to add new wrinkles to my genius.
That's where things started to get interesting.
Stephen Ilardi responded first:
Sorry to say, I have to weigh in on Tom's side. My analysis has Shaq, even in limited minutes, contributing more this season on both sides of the ball. Without giving away anything proprietary, I can say that Darko actually rated below the average level of a so-called "replacement player" (i.e., a guy who could be snatched up off of waivers or called up from D-league on a moment's notice) in his overall per-minute contribution -- almost solely due to deficiencies on the offensive side of the ball, as his defensive play wasn't bad.
Shaq, on the other hand, made a net positive contribution to the Celts' bottom line when he played.
Silly Stephen, I thought. He's blatantly a genius, yet misundertands. Of course O'Neal is more efficient, but what about the dead simple question of who did more? (Efficiency is production divided by something, whether minutes, possessions or dollars. This contest would brook no such division.)
I replied to Ilardi, saying, essentially "silly me" for asking a dumb question to a smart guy. He was charming in sympathizing with my plight, but tactfully added that even by basic box score metrics, O'Neal was still preferable.
Whatever. There are seven experts. Surely at least four of the remaining six would see the wisdom of my position.
"I'm sorry to say this, Henry," came the next e-mail, from soon-to-be-former friend Neil Paine, of Basketball-Reference, "but I don't see any way in which Darko could be construed as more valuable." That e-mail was pretty long and had a bunch of charts I will never read.
"The answer," e-mails Ben Morris, who until a few days ago seemed like a great addition to the Smackdown, "is Shaq, and it's not close."
"I'll be shocked if this isn't unanimous," adds Matthew Stahlhut, helpfully. "This is no contest. Shaq had 2.7 win shares and Darko had 0.2 and I can't find anything deeper in the numbers that would lead me to any other conclusion."
About the time all this was happening, revered statistical expert and new ESPN guru Dean Oliver called to see whatever happened with this bet. I sketched out the situation for him. He noodled with his laptop on my behalf, mumbling things like "let's see what happens if I set the value of a replacement player really low ... nope, wow, that still doesn't work."
Basically, he was trying to find some way -- any way -- that I could be construed as correct.
Effort, people, is not always enough.
Oliver failed. Which means Milicic failed. Which means Minnesota GM David Kahn failed. Which means that, even though pretty much exactly what I predicted came to pass, I failed.
Briefly I channeled some of the angriest commenters in TrueHoop history -- longing for a time when basketball analysis was dumber, so that I might feel smarter.
The numbers in that little chart up there scream I'm right!
Or not. Think more, think a moment, and however many rebounds he may have, it's all pretty pointless if, on average, when Milicic was on the floor, the Timberwolves suffered for his presence, while the opposite was true of O'Neal and the Celtics. In such a world, O'Neal for one minute is worth more than Milicic for 2,000, and anything that says otherwise is smoke and mirrors.
And one other lesson: When betting, establish very clear terms.
In the meantime, Mr. Ziller, I have a debt to pay. Look for my ode to Shaquille O'Neal in the days to come.