Where did you get the possession numbers? They only include possessions that the players were together on the floor, right?
One problem I have with APM is that the possessions are not weighted. I mean a big second quarter in December at home against the Timberwolves is worth as much as a big 4th quarter during a playoff push against a rival on the road.
I'm neither agreeing nor disagreeing. It just doesn't matter. Some very good coaches use it, and some don't. And a better example would have been Phil Jackson, arguably the greatest NBA coach ever, who is known to care very little about stats. http://espn.go.com/blog/truehoop/po...ses-artest-for-nba-lead-in-overall-plus-minus [rquoter] Tangential note: LeBron James was a league-leading plus-871 last season, and during training camp Pat Riley pointed out to Dwyane Wade that Michael Jordan registered plus-700 in nine consecutive seasons. [/rquoter] http://sportsillustrated.cnn.com/vault/article/magazine/MAG1067216/2/index.htm [rquoter] In implicitly criticizing Johnson, Riley may be engaging in mind games to push Magic to another superior playoff performance—Riley often does that kind of thing—but, to whatever end, he seems determined to make the point that Johnson could be doing more. Sitting in his office in the Forum one day not long ago, Riley glanced at a computer printout that listed a "plus-minus rating" (a statistical analysis made by the Lakers' staff that covers everything from points scored to fouls committed) for every player in the NBA. The Chicago Bulls' Michael Jordan was on top with a .770 rating, Boston's Larry Bird was second at .700 and Johnson was in third place at .680. [/rquoter] Also,this 1996 article describes how many coaches were using new computer technology to evaluate their lineups and strategize for the playoffs. Rudy T and Larry Bird were mentioned specifically as coaches who "couldn't get enough of this stuff". I would bet that Pat Riley had his young future coach Eric Spoelstra pouring through that information and coming up with reports for him. But don't take my word for it: http://espn.go.com/blog/truehoop/mi...6/how-advanced-stats-changed-chris-boshs-game [rquoter] Before being promoted to head coach ahead of the 2008-09 season, Spoelstra was an assistant coach and director of scouting for the team but also helped to build an internal advanced statistics database for then-coach Pat Riley. “As an assistant, I used numbers quite a bit for my reports for Pat,” Spoelstra said. “With Pat, you could never give him too many numbers. He could always digest it. And then when you start working with players, you tell them, ‘Hey, you happen to be really good here, struggling a bit here. Let’s work on this.’ And then you give them a number and they’re surprised.” [/rquoter] Also: http://www.welcometoloudcity.com/2009/12/7/1190090/loud-links-12-07-09 [rquoter] When Jackson arrived in Los Angeles, Lakers trainer Gary Vitti handed Jackson a packet of print outs that included plus-minus, a daily ritual when Vitti worked for Pat Riley. "I told him to throw it away," Jackson said. "I’ve never consulted those. I know there is some value to it. But to me, it’s an instinct by a coach. ‘This group is going pretty good. I think I’ll leave them in there.’ But obviously there’s some merit to it. You can’t throw it all out." Riley, now the Heat’s general manager, said he devised his own plus/minus rating 30 years ago. In Riley’s formula, he grades players in all areas, including effort, then shows players their best five games as a standard to shoot for. "We wanted to quantify an across-the-board evaluation," Riley said.... [/rquoter] Does "intensively quantitative methodologies" including watching tape? If not, then I would choose "traditional approaches". Fortunately, we don't have to limit ourselves to one or the other.
Yeah, I know, but it still inflates guys like Chuck who go hard all the time on D and guys like Kevin who are always looking to get their points. Kevin Martin: Net pts (all minutes): +167 Net pts (clutch): -56 http://www.82games.com/1011/10HOU7.HTM Chuck Hayes: Net pts (all minutes): +213 Net pts (clutch): -12
That's not a problem intrinsic to APM, though. While with Dallas, Wayne Winston used an "Impact Rating" metric that was basically APM that weighted possessions based on how important they were. Some methods (Rosenbaum) will also filter away possessions that are clearly unimportant, like garbage time.
I noticed that Martin is part of 10 combines for the top 30 offensively combos and none of the combinations for top 30 defensive. Nothing new, but it was funny to see.
I feel the problem isn't that we have good, smart, underrated players. I know we have a lot of them, as I'm sure Rockets stat guys have even better measurements to find hidden gems. The problem is that none of this matters if we lack star power. If all Morey has to do is find role players around Durant/Westbrook, Yao/T-Mac(circa 2004), Dwight Howard, etc. then I'm sure Morey can find the right pieces to surround them and build a contender. But we have no centerpiece, so all else matters little at this point.
So we're taking statistics based on some 2008-09 data...and extrapolating it out for the 3 following seasons...and we're ignoring the fact that many of the "pairs" never played together...and the Rockets were in the Lottery 2 of the 3 seasons...and they were collectively 138-108 (a winning % .001 lower than this years' NO Hornets & MEM Grizzlies...one of which was a 1st round exit)... And we're (once again in a durvasa thread) being "sold" the statistical values Hayes and Battier. Folks, you all can dig into those stats all you want. Only one stat matter to me. The Rockets' 2010-11 win %age improved AFTER they got rid of Battier. The % will go up further once Hayes is returned to his natural position. Backup situational PF.
The statistics are 2008-2011. That was a typo on my apart. They are ratings for when they were playing together over the past 4 seasons. So what? Who's arguing that the Rockets were supposed to be a great team based on these statistics? Where are people like you and saleem getting that from? Quit reading into things just so you can get on your tiresome anti-stats soapbox. What was Adelman selling us when he went on and one about the values of Hayes and Battier? How come you never respond when I ask you that? It doesn't quite jive with your argument, now does it? Of course. And if the winning% went down, that would be yet another "$Ball" stat that we should ignore. Much as we shouldn't pay any attention to how much better our win% was with Chuck Hayes starting rather than Jordan Hill. You select the numbers that suit you, but that's nothing new.
The bottom line is to win durvasa. I don't care about anything else. Stats have their value,but have their limitations. Hayes certainly made a difference without a doubt,big time. Battier was important,but we didn't die without him as so many were predicting.
Fine. From now on, if you make a post referencing anything factual other than wins and losses, I'm going to call you out on it. Deal?
If you're trying to win the next game, then relationships like this can help you achieve that. I think that's all you can really make of these numbers. They're not saying these guys give you the best shot to win out of all the others in the NBA. Looking at the big picture, yes, we're not good enough to win it all and need to keep making moves, but while those moves come, or in between them, you still have to find the best way to win with the guys you do have at the moment.
Rick Adelman can praise any player(s) he desires. The sheer fact is this. Unless the critique is being used for effect...as in Phil Jackson trying to motivate Kwame Brown....NO NBA coach publicly criticizes any of their players. Praise is spread on a lot thicker than it's true value. You seem to take the media BS as some sort of gospel that supports your goofy stats. Players like Battier and Hayes are all over the place. You get stars and then fill in the blanks. If you're too geeky to realize this, we have no point of discussion. What in hell are you talking about? The Rockets' win %age went up after they got rid of Battier. Go do the arithmatic. It's a pure fact that's not buried somewhere. Oh...and what else is not buried in a pile of stats BS? Budinger had a higher 2010-11 +/- than the brick-laying Battier. http://www.82games.com/1011/1011HOU.HTM As far as "pairs"...Martin played a miniscule number of games with Yao at the tail end of Yao's career. That alone ought to demonstrate the lack of relevance to your goofy "pairs" crap. Bottomline...if Chuck Hayes is still starting at C next season...their headed back to the Lottery regardless of how much McHale reps him.
Perhaps I skimmed past it, but can anyone show me proof that Adelman doesn't use these metrics (or more advanced ones, assuming they exist, which wouldn't surprise me in the least), and other similar tools as a head coach?
I don't know if it was ever directly mentioned, but I think it was insinuated from some third parties (Chronicle articles). Part of the reason why "the front office and Adelman weren't seeing eye-to-eye" -- he wasn't as receptive to their #s/suggestions as they perhaps hoped.
LTF, I read insinuations here all the time. They are worth as much as MY insinuations. In other words, not a hell of a lot, especially coming from the writers covering the Rocks for the Chronicle. I'd be more likely to pay attention to those from a select number of members here on the BBS, Clutch being at the top of the list.
