Score One for Adjusted +/-

Everyone new that the new Boston Celtics were going to be a good team. But I think a lot of people -- myself included -- have been surprised by just how good they've been. Certainly, I wasn't envisioning this level of defensive domination. One person who's not surprised, though, is schtevie at the APBRmetrics forum who noted on August 3 that if you believe in the adjusted +/- stat that it predicted "that the Celtics will outscore their opponents by between 10 and 13 points per 100 possessions, which would put them in the same league as the greatest Bulls team, which had the highest margin of all time (around 12, if I recall correctly)." He also noted that the estimate might be biased downward because "there is no reason to believe that the Celtics will fill out its roster with replacement level players as there will be an incentive for decent veterans, looking for a ring, to come along for the ride. Also, it was assumed that the current remaining youngsters only had replacement value, and I suspect (at least in the case of Rajon Rondo's) that this isn't the case."

And, well, it's all true. Their efficiency differential is a ridiculous 16 points per hundred possessions. Boston opponents should consider themselves lucky that the Celtics only play 93.6 possessions per game, thus keeping their margin of victory to only 13.8.

I assume the greater accuracy of the adjusted +/- relative to other ways of looking at the situation has something to do with the fact that it can, in principle, do a much better job of capturing the defensive half of the game, which is really where Boston has distinguished itself.

Comments

I disagree. Not everyone new.

Posted by MS | December 14, 2007 1:08 PM

Score one for schtevie.

Posted by Petey | December 14, 2007 1:18 PM

And score one against Berri. As usual.

Berri is the Timberwolves of analysts.

Posted by Petey | December 14, 2007 1:20 PM

John Edwards is the Timberwolves of presidential candidates...

Posted by Oh No You Didn't | December 14, 2007 1:22 PM

The adjusted +/- is my favorite statistic because it is not driven by the box score, but just by the scoreboard.

I am a little surprised by how well the C's have come together so quickly. Mainly because the only things I have seen Ray Allen do in his career is shoot and whine about anyone who defends him agressively.

The biggest problem facing the team right now is the ridiculous number of minutes Rivers is putting them on the floor. I mentioned the other day that if they continue at current paces, Garnett would play something like 12 more games than Duncan. The discrepancy is even worse for Allen and Ginobili.

One tweak of the model that might be interesting is projecting play-off performance based on a play-off type rotation, where teams like the Spurs and Mavs who are consciously sitting their main guys significant minutes bump those minutes up to the 36-40 they will average in the close play-off series.

I wish I didn't have a real job.

Posted by Curtis | December 14, 2007 1:26 PM

I edge closer to the band wagon after I see their western swing (love that steel guitar).

Posted by cw | December 14, 2007 1:35 PM

And score one against Berri. As usual.

Berri is the Timberwolves of analysts.

The dynamics here are everything Berri is stupid/stubborn to grasp. Two biggies:

1) Players with high offensive officiency who also have a high level of utilization are more valuable than those with a low level of utilization. Stick them together and their collective officiency will frequently increase (if the players fit together properly)

2) Defense. Berri pretty much denies that players really bring anything on the defense side of things other than rebounds, I suppose. This is obviously stupid and pretty clearly a problem here.

Posted by mpowell | December 14, 2007 2:01 PM

2) Defense. Berri pretty much denies that players really bring anything on the defense side of things other than rebounds, I suppose. This is obviously stupid and pretty clearly a problem here.

Yeah, this baffles me. Nobody agrees with Berri about (1), but I understand what the argument is supposed to be. But to not believe that players differ in their defensive ability is ridiculous.

In general, stats that incorporate defensive rebounds but not other aspects of defense (PER, too, in short) seem to suffer from a silly and obvious flaw. The best "advanced" stats are things like TS% that are self-consciously offense-only.

Posted by Matthew Yglesias | December 14, 2007 2:18 PM

Gosh, I haven't looked at their minutes played, but with the number of blow-outs they have had, I'm really surprised to hear that Garnett's minutes played are so high. Garnett always played mega-minutes in Minnesota, for obvious reasons, but I had just assumed that his minutes played were down because the Celtics were winning so many games by large margins.

Posted by Will Allen | December 14, 2007 3:13 PM

"The biggest problem facing the team right now is the ridiculous number of minutes Rivers is putting them on the floor."

Doc was playing the Three Horsemen a whole lot of minutes to start the season, but it's starting to become more reasonable (partly because of lots of blowouts).

Paul Pierce 37.9 MPG
Ray Allen 38.7 MPG
Kevin Garnett 35.8 MPG

compared to

Dwight Howard 38.7 MPG
Rashard Lewis 39.3 MPG
Hedo Turkoglu 37.0 MPG

Pierce and Allen need to play fewer minutes eventually, but that may happen as Tony Allen continues to recover from his knee injury.

Mike

Posted by MBunge | December 14, 2007 3:38 PM

It's fascinating that the two dominant teams of the last 20 years (assuming the Celtics remain dominant, and the 1996 Bulls), were built without the use of an all-star (or even particularly good) C or PG, the two positions commentators universally describe as vital to a good team.. Both teams were dominant at SG, SF, and PF, but are pretty weak everywhere else (Harper/Kerr were pretty good, Rondo is a notch below them; Longley/Wennington were mediocre; Perkins/Big Baby are a good bit better, particularly in the defense/rebounding/ demonstrating some ability to physically move and not be a complete stiff-areas).

I'm still not sold that the Celtics are in the 1995-6 Bulls class, or will even end up with more than ~65 wins, they haven't played any great teams. But unless you're up against the Washington Generals every night, 18-2 is damn good. And having watched almost every game they've played, they've demonstrated an ability to "turn it on" at will. 20 games without getting blown out (one loss in OT, one by 2) may be even more impressive than 18-2..

I'm still trying to figure out how the Celtics D is as good as they've been. I understand them rebounding well (Garnett is a great rebounder, Perkins/Big Baby are good, and Pierce is pretty good for a SF), I can see them winning the TO battle (with their veterans), and they could actually improve in both of these areas as the season continues. What I don't understand is their FG% defense, 41.3%? Really? 45% when adjusted FG% defense? That's much better than Detroit or San Antonio, which is hard to believe. Is Garnett that dominating inside? Just weird..

Posted by Crusty Dem | December 14, 2007 4:00 PM

Garnett is a great, great defensive player, and Perkins and Rondo and particularly James Posey are in very strong on defense and defensive-minded.

But it's less Garnett's talent/skills that makes the Celtics great on D but more his intensity and focus on defense that seems to have forced other Celtics - like Pierce and Allen, and even Eddie House - to play the best defensive of their careers. They are playing D like they know if they didn't they wouldn't be able to look KG in the eye afterwards. It's an amazing phenomenon.

Anyone watched the Celtics/Kings game the other night saw Pierce play D on Ron Artest with the same focus and intensity that Artest always plays Pierce with. Pierce doesn't have the defensive genius of Artest, but when he plays that hard, he's pretty damn good. Strong, quick, smart. In the past, Pierce would work so hard on offense carrying a bad team and battling Ron Ron, that on D he'd relax. Not anymore. Not when he knows he can actually rest on offense, feed KG and Ray Ray, and make the team better. Plus of course, he seems to not want to let KG down.

I say the Celtics win 64.

But we won't really know how good they are until they play the Spurs and Suns and Detroit on the road.

Posted by The CarpetMuncher | December 14, 2007 4:27 PM

"And score one against Berri. As usual."

Huh? Berri was higher on the C's than almost anyone-- he predicted, following his numbers, that they'd the Spurs in the Finals (with the caveat that the Spurs and C's matched up very well and it was really a coin toss as to who would win).

Posted by PJ | December 14, 2007 4:38 PM

For amusement, take a look at NBA.com's all-time playoff bracket and how these geniuses picked the 1986 Celtics over the 1996 Bulls:

http://www.nba.com/playoffs2004/challenge_bulls1996_celtics1986.html

Talk about not taking defense into account, who on the Celtics is guarding Jordan, Danny Ainge? That would be amusing. Not to mention Bird trying to keep up with Pippen and McHale dealing with Rodman, it's just a matchup nightmare for the C's...

Posted by Crusty Dem | December 14, 2007 4:45 PM

Also, it's not accurate to say that Berri ignores defense. He weighs steals, blocks, and (I think?) fouls committed. That's hardly definitive, of course, but it's possible that those first two categories work as (imperfect, of course) proxies for defensive ability more generally.

Posted by PJ | December 14, 2007 4:48 PM

Just generic steals seems pretty usless as a defensive stat to me.

Posted by cw | December 14, 2007 8:18 PM

Two Words--KG!!!

Posted by danielofbballfame | December 15, 2007 12:44 AM