Monday, June 13, 2011
Erik Spoelstra's passivity with personnel
By Tom Haberstroh
Coaches probably wield less control in games than we think. They offer pep talks during timeouts, call plays and make substitutions, but filling out the lineup card might be the strongest imprint a coach has on the game.
It's safe to say Dallas Mavericks coach Rick Carlisle won the coaching battle in the Finals, thanks in large part to his decision to insert the speedy J.J. Barea into the starting lineup in Game 4 after the Mavs fell behind 2-1 in the series. Barea carved up Mike Bibby and the Heat and zipped into the gut of the defense without resistance. In the coaching chess match, Carlisle brought out the queen early and Erik Spoelstra countered too late.
One way that we can examine the lineup battle is look at 5-man unit plus-minus. Simply put, it answers this question:
How did the scoreboard change with a particular 5-man lineup on the floor?
There are a couple ways of measuring the success of these units. One is just to use raw plus-minus. For example, in the Finals, the lineup of Bibby-Wade-James-Bosh-Anthony was the Heat's most frequently used five-man unit and it was outscored 121-116 in the 66 minutes while it was on the floor against the Mavericks. This group lost by five points over that time, so we mark that down as a raw plus-minus of minus-5.
But, there's more work to be done. In order to make sense of that minus-5, we can translate it to a per-48 minute scale. A minus-5 in 10 minutes of play should be treated differently than minus-5 in 66 minutes, because if the game continued at the latter rate, it would get out of hand very quickly. Now that we standardize it to 48 minutes, every lineup is on a level playing field. The Heat's starting lineup with Bibby was outscored by 3.6 points every 48 minutes.
And then there's a third step -- calculating the adjusted plus-minus -- which we should leave up to the professionals. That's why I call upon Wayne Winston, author of Mathletics and professor at Indiana University, who emailed over his adjusted plus-minus ratings in the Finals.
Why the adjustment? Because strength of opponent matters too. The Heat's starting lineup with Bibby didn't outscore the Mavericks in 66 minutes, but we should also account for the quality of the other team's five-man unit as well. Once we do that we find that they had a plus-5.2 rating, which means that unit is 5.2 points above average.
The problem with adjusted plus-minus is that when you engage in statistical gymnastics of this magnitude, we need really big samples before we can see the light.
But it's fun to look at, nonetheless.
Enough with the primer, let's get to the numbers. Here are the Heat's top ten most used lineups in the Finals and Winston's adjust plus-minus ratings.
The first thing that jumps out, to me at least, is that Spoelstra's most-used lineups were getting beat. Big Three plus Bibby and Anthony? Minus-5. Chalmers and Haslem next to the Big Three? Minus-17. Chalmers and Anthony? Minus-one.
I actually think the Chalmers-Haslem lineup has a chance to be the Heat's best lineup going forward, but it got spanked by Dallas in the 59 minutes on the floor. It isn't until you get the bottom half of this list until you see some positive differentials. If you look at the Mavericks' best lineups, you'll see the opposite. Lots of plus-signs at the top half with some minuses at the bottom.
The numbers and the eye test both tell us that Bibby didn't belong on the court in this series. When Carlisle started Barea in Game 4 and Spoelstra was slow to react, it swung the momentum in the Mavs' favor. Think about this stat: Of the 1,166 times in NBA history that a player has played 400 minutes in a postseason, Bibby's 3.6 player efficiency rating (PER) ranks dead-last. 1,166th! Rick Mahorn stumbles in at second-to-last at 4.4 back in 1987-88.
Here's something else to chew on: How did the Mavericks perform in the 33 minutes with Barea on the floor against Bibby? Plus-17. The Mavericks in the 95 minutes with Barea on the floor and Bibby riding pine? Minus-41. That's a 68-point swing.
The interesting thing is that Spoelstra has made a habit of riding an unproductive starter for a long time and then completely dropping him from the rotation. Spoelstra will start someone for months and without a moment's notice, he'll bury a player on the bench, delete his name from his phone and defriend him on Facebook. He did it to Zydrunas Ilgauskas. He did it to Erick Dampier. But no one knows why he waited until Game 6 to do it to Bibby.
No, Bibby isn't the primary reason why the Heat lost, but he clearly lost his legs months (years) ago. At this age, he couldn't drive past a bar stool, and he shot a cool 28 percent from the floor. Throwing him out there because "he organizes the offense," which Spoelstra maintained, appeared to be a costly case of cognitive dissonance.
Moreover, the James Jones thing continues to be a mystery. Spoelstra stated that Jones, who had suffered a bum toe in the playoffs, was ready and Jones himself insisted that he was healthy enough to play. Instead, Eddie House inexplicably received the nod to play 21 minutes in Game 6 after playing fewer than 10 minutes since April. Apparently, Jones, the best shooter on the team, had been taken out back just like Ilgauskas and Dampier.
At the end of the day, it's mostly true that the Heat will go only as far as the Big Three takes them. But Bibby was clearly outmatched in this series and the NBA's downtown champ sat on the bench as the Heat clanked 3-pointer after 3-pointer. It's also true that LeBron James wasn't the only one who's passivity hurt the Heat.