juwan needs plays run for him. he's not a guy who can get garbage buckets like battier. with yao out, there are more offensive opportunities to maximize what he does well. its the exact opposite of chuck hayes, who can be yao's clean up man. but he can never have any play run for him.
Hayes just seems to gel with Yao so well except for the occasional matchup such as Garnett where Garnett gets it rolling. And then nobody can stop him. Hayes though, has had good games against both Nowitzki and Duncan this year. I like the Juwan Deke combo with the second unit. They have played well together since Yao has been out. And I think they could become even more potent as Bonzi gets into better shape and becomes the effective post up option on the 2nd unit.
I recalculated some of these +/- numbers, and it looks like I might have screwed up the first time. Anyways, here are the numbers since Yao went out (per 48 minutes): Code: [B] MIN ON OFF NET[/B] Alston 1024 6.8 4.1 2.7 McGrady 991 7.7 2.1 5.6 Battier 1030 5.6 7.8 -2.2 Howard 909 6.2 5.9 0.3 Mutombo 797 7.3 4.5 2.8 Head 778 5.1 7.5 -2.3 Hayes 558 5.5 6.6 -1.1 Wells 331 7.7 5.6 2.1 And, specifically, in the month of February: Code: [B] MIN On Off Net[/B] Alston 352 +6.5 -5.4 +11.9 McGrady 370 +1.7 +8.4 -6.7 Battier 351 +1.4 +8.2 -6.8 Howard 306 +4.4 +1.3 +3.1 Mutombo 282 +2.2 +4.7 -2.5 Head 256 +9.4 -3.6 +13.0 Hayes 196 +0.7 +5.0 -4.3 Wells 151 -3.5 +6.3 -9.8 One interesting development is the Rafer-Luther pair seem to be much more effective lately compared to the first month without Yao. Since Yao's injury and through January, we were -2.4 with them both on the court versus +13.6 otherwise. But since February 1st, we're +19.7 with them both on the court and -3.7 otherwise. It looks like that is a combination we're going to rely on a lot, so hopefully the recent trend holds. On the other hand, we've actually had a much better +/- when McGrady or Battier have been off the court. With them both on in February, we're only +0.5. With one or both of them off, we're +7.2. Strange.
Since we are just talking about a month's worth of games, perhaps it's attributable to luck. Doesn't seem to me from watching TV that the team really plays better with McGrady or Battier on the court... Maybe the Rockets just had a few very good 2nd qtrs, when teams tend to play the bench the most.
Problem with this stuff is that it doesn't take into account the game situation (are we leading, behind, playing superior opponents...). For example, if we're playing Dallas in a close game, then certain players will be in, and since we're more likely to lose and fall behind, they get more negative numbers. Against a weaker team we'll be scoring and extending the lead more often and it's less important who's in or not. In fact, once in control of the game, say up by 10, we might have bench players come in, and if it goes to a 25 point lead, that person just got a +15 for say 10 minutes of play, which translates to +15 for a five game stretch or +5 for an entire month. So in other words, looking at this stuff for the month of Feb - you're likely to get a bunch of statistical noise....and you can't read too much into it unless you take very long periods of sample and/or look at it by opposing team. If you really wanted to do the analysis right, you'd need to do a multivariate analysis - include everything and anything, and you might find some interesting stuff. But this - interesting but probably nothing more then directional. By the way, of course Juwon and Mutombo, and T-mac will become more important after Yao's injury. And off course guys like Head, Hayes, and Battier will suffer.
Except we know very well that JVG doesn't really modify his rotations based on margin, unless we're the ones that are losing. At any rate, I could look at +/- only considering stints where we aren't leading by more than, say, 10 points. But I agree there can be significant limitations to these numbers, particularly if we're trying to draw strong conclusions from them for a 1 of 2 month period. I set up a spreadsheet, with which I can refine the constraints to look at a specific set of games, specific periods within games, and/or situations in games where the margin is within a certain amount. But with more constraints, you get a smaller sample.
What? Why you now come up with a different story? I just thought Luther and Alston are poisons to this team when both on the court. Then I saw their magnificent +/- numbers...
