Tuesday, March 24, 2009

Count on one hand

If you have the chance, I would highly recommend watching tonight's Utah-Houston matchup. Unfortunately, Carl Landry is still recovering from a gunshot wound, so we won't be treated to the Paul Milsap - Carl Landry showdown, but both of these teams have been playing so well recently that this game should be a must-watch.

What to watch for: These teams know each other pretty well by now, so they should be pretty prepared for this game. Utah once again has one of the more efficient offenses in the league, and Houston, as usual, has one of the better defenses. The battle of the boards might be telling -- Utah is the fifth best offensive rebounding team in the league, grabbing close to 29% of their misses, while Houston is the fifth best defensive rebounding team in the league, only allowing their opponents to retrieve 25% of their own misses. So, something has to give (in 3 games so far this season, Houston has done an acceptable job of keeping Utah off the offensive glass). The battle should be particularly pronounced at the power forward spot, where Paul Milsap and Carlos Boozer of the Jazz are two of the better offensive rebounders around, while Luis Scola and Chuck Hayes of the Rockets are two of the better defensive rebounders around. At the other end of the court, the Rockets have been mostly average on offense while Utah has improved their defense this year and are top 10 in the league right now. However, Utah continues to be pretty foul-happy, which is a problem since the Rockets as a team are the fourth best free throw shooting team in the league, at 80.6%. The Jazz need to avoid giving up free points to the Rockets. Also, Ron Artest has really improved his offensive game over the last few games, playing within himself and making good decisions. If he continues to do that, and Kyle Lowry continues to give them some fast break opportunities off the bench, then Houston's offense should be fine.

Other: I've been noticing more and more how pronounced the internal narrative of each game I watch is. The story is easier to pick up in some games than it is in others, but I suspect that this Houston-Utah matchup should give us a pretty good storyline, without having to build anything into it from the outside.

I linked to it elsewhere, but this piece about Lamar Odom is worth reading.

I agree with a lot of what Bill Simmons says in this ESPN the Magazine column, in fact he echoes a lot of what I've been saying here (he has a few details wrong -- for instance, what he calls "mega-assists" are already available at 82games.com and Queen City Hoops (as super-assists), and 82games also answers the Dwyane Wade question for us -- his effective field goal percentage is a full 1% higher if you exclue the 10% of shots he takes with 3 or fewer seconds left on the shot clock). I like this line: "You could be feeding us gourmet cheeseburgers, except you're more interested in cloning cows. Let's clear up the small picture before we get to the big one." He's definitely right to say that teams have access to a lot of game-charting information that would be great for us as fans to see, but that they keep a secret. What he's saying about wanting more contextual stats reflects, I think, what is so interesting about what this guy's been doing recently, but it's more than a rejection of boxscore-based value metrics like PER and Wins Produced -- it's philosophically a repudiation of things like adjusted plus-minus, which in its current form assumes that a player's effect on the court is fixed regardless of the context changing around him. His example of the Spurs needing Bruce Bowen more than they would have needed Carmelo Anthony makes sense even though it's problematic -- Bruce Bowen throughout the last several years has consistently performed better in +/- type stats than Carmelo Anthony. The argument Simmons is making, though, isn't that he's a better player, but that he's a better fit (although might have been even more powerful if he used Tayshaun Prince and Anthony, seeing as how the Pistons had the opportunity to draft Anthony and passed). The argument could conceivably be extended to other more extreme cases, and particularly cases where even the adjusted +/- tells us that player B is better but we know player A is a better fit -- although I'd warn that the most recent Olympics showed us that the superstars in the league are much more capable of expertly filling specific and limited roles on talented teams than most of us probably would have imagined.

And, speaking of 82games, there's a good interview of Roland Beech (the guy behind the site) up at Slam Online. Mr. Beech also gives a pretty persuasive argument in favor of more context and less all-encompassing player metrics. He and Simmons together in the same day seem to be good cheerleaders for tracking more basic information in boxscore-like formats as opposed to trying to guage an overall value: "Oddly while I have published a lot of regression based ‘adjusted +/-’ articles on 82games, I am not actually a fan of that approach. I think again, with more data on hand you can really understand a player’s strengths, weaknesses and traits very clearly without having to resort to mathematical techniques to try and extract info that you think is ‘missing.’" I also loved that he ended with this: "It’s ultimately entertainment. I take issue with the notion that teams should be all about a championship or they need to blow things up. It was sad to see the Suns dismantled prematurely to my mind when they were such a great team to watch, and had certainly some significant success with still the hope of finally breaking through."

And finally: The Lakers website has an article up about Luke Walton and Jordan Farmar that maybe sheds a little light on my previously hypothesized "BFF Theory."

1 comment:

  1. how does the BFF theory work with kobe & shaq?
    destructive disdain and gorgeous basketball.

    re: simmons - momentum is such a tricky and all-encompassing part of the game, more so than in baseball. that's what seems to be missing from the stats discussion: when these stats are a collection of things that have already happened and lack the x-factor of momentum in predicting what could/should. how do you factor for that in formulating winning strategies w/ stats? i.e., dwight howard is smiling after he swats the opposing team out of bounds because the deflation of spirit is not tracked on any stat sheet.