Anything for wins With the Rockets' choice of Daryl Morey as general manager beginning with the 2007-08 season, and Rockets coach Jeff Van Gundy's contract expiring after next season, Van Gundy was asked about the possibility of an extension. "I don't care about (a contract extension)," he said. "I would trade the last year of my contract next year just to win (tonight). Then we'll start taking limbs. The pinky for (beating) the Lakers." His nose? "That should be worth two wins," Van Gundy said. "It's a sickness. It is. "I may give up parts of Carroll (Dawson), too. Since he's on his way out anyway, maybe he'll give up some limbs for the cause. We'll take his pacemaker." I laughed out loud at this (even tho it probably gives cred to the earlier reports about the hate between JVG and CD ).
Agreed. I should have written more succinctly. The reason I generalized your posts to begin with is I'm not in Houston, didn't see the Fox clip and had absolute faith in not only the credibility of your summary, but also your impression of the clip. I would have withheld an opinion in skepticism if your post had been made by 90% of the people on this board. And I was shocked by your information. Honestly, my too-broad of a stroke with my remark was just expedient and now regrettable, I should have just said "I believe the accuracy of this post". I apologize but sincerely there was zero intended criticism or sarcasm. Just poor writing on my part.
Deckard, that sums up my anxiety, confusion and hopes perfectly. I have to be honest here. If someone had told me last week that Les was going to hire CD's successor and I'd suddenly be worried about JVG's status here I'd have laughed in their face. But that's where I am now.
LOL. There's so much droll wit in those remarks I can't hazard a guess as to the usual underlying truth in his humor. I'll take a stab at CD's possible retort: "My pacemaker is no big deal. I'll refrain from telling you which body part JVG'd give up for a title."
(1) Not sure what you mean by "explanation or just stats". The inference being made is that Scalabrine appears to have had an overall positive impact to Collins, Carter, Kidd, & Kristic as indicated in the fg%, TO's, reb's, & ast's. (2) Regarding the 1 year issue: two things (a) they probably felt confident with the sample size (430 min - 530 min) being looked at. Bball stats guys seem to set the min. sample size to being 250 minutes or more to ge a valid analysis. I don't know the math behind how this rule of thumb is derived but the 430-530min sample is well above the 250 minutes range. (b) to do what they did took a lot of work to pull the data together and folks shouldn't underestimate the effort to do this. It's not just populating stuff but being able to actually synchronize the right windows of game stats to sample in the flow of each game. Not easy to do whatsoever. Pulling another year's worth of stats could easily be double the amount of work. (3) Could the improved #'s be a result of VC, JC, JK, & NK playing against scrubs while Scalabrine is in? That's a possibility and it doesn't look like the analysis tries to isolate that out. However the probability that VC, JC, JK, & NK all have improved #'s while BS is in, it seems unlikely that it's solely a reflection of those guys playing against scrubs while BS is in. 2ndly it's easy enough to do a check and Morey spells it out. The +/-'s were done with VC, JC, JK, & NK + other players subbed in and compared to that of BS subbed in. Assuming the similar caliber of opposition players (scrubs ) subbed in at the time if VC, JC, JK, & NK's stats would show improvement as well - but it doesn't. Instead there is a corelation between BS in with improvements for VC, JC, JK & NK. (4) The per 48 minute approach is a way to normalize analysis of performance so that you're looking at apples to apples vs. apples to oranges. The idea is to look at the ratios comparitively vs. looking at the absolute values because absolute values are distorted by the number of minutes played. So making all the minutes the same (say 48) gives a way of isolating out the distortion due to differences in number of minutes.
(1) Actually I was trying to say if there is any logical explanation why Scalabrine is a +ve impact player besides these stats. Not important anyway. (2) Okay, good explanation. (3) No, I didnt say the Nets played against scrubs when Scalabrine was on the floor. I said there would be an opponents' backup forward in the game so the opponent's interior defense was weaker. Obviously, the opponents' forward needed to stay with Scalabrine around the perimeter. If the +ve impact came from the opponents' player rotations, the role of Scalabrine was replaceable. So to say, we could put S Padgett to replace B Scalabrine and the impact would be similiar. If this case is true, the +ve impact is independent of the player. (4) I understand "per 48 mins" is a common base so they could compare the stats. But when we talk about starters' stats, we commonly use "per 40 mins" base. It is unrealistic and a delusion if we use "per 48 mins" base because the starters' mins are normally 35-40 mins. In Scalabrine's case, they should use "per 15 mins" base. When I saw the stats line: "Carter +7.7% -1.6 +1.8 -0.7", I thought it was a lot. But if I scale it down to 15 mins, the impacts on RD, TO, and Assists were minimal and VC only gained about 1.2 points (+7.7% FG%; VC averaged 21.7 FGA per 38.9 mins). That's why I said "per 48 mins" base is a magnification because it gives me (or just me) a wrong impression. So, Scalabrine is nothing but a minimal +ve impact player. Thanks anyway!
The purpose of a looking at a player's per-minute numbers is to guage what he does while he's on the court. To that end, it's immaterial whether you project to 48 mintues, 40 minutes, 15 minutes, or even 1 minute. The ratio is going to be the same (per minute), regardless. To consider a "real world" example, take Michael Jordan in his second season. He was injured pretty much all year, but came back for the final several games just before the playoffs. He only played 25 minutes a game in 18 games that season, and he averaged 22.7 ppg. Obviously, just looking at 22.7 ppg and one might think it wasn't that special. So, only considering his per game numbers, we might have projected that he'd averaged around 22 ppg in the playoffs. Guess what? He averaged 43.7 ppg in the playoffs (all against the Celtics), including a 63 point game. Unless one took MJ's per minute numbers more seriously, they may not have expected such an outburst. But if you looked at what he averaged in the regular season per 45 minutes (his playoff mpg average), he scored over 40 points per 45 minutes. You'll find with most players that their per minute numbers from season to season are more indicative of future performance than per game numbers (assuming they play a certain number of minutes in the season -- say, 250). Here's a question: if you had one player who averaged 15 ppg game in 20 minutes per game, another player who averaged 20 ppg in 40 minutes per game, and that's all you knew about either of them -- which player would you rather have? The smart money would be to go with the first player. And you'll find, in fact, that Tracy McGrady averaged 15 points in 20 mpg his rookie year, and Cuttino Mobley averaged 20 playing around 40 mpg a couple seasons with the Rockets.
If this guy does all this quantative analysis for us, he will realize that we need to trade Howard ASAP......yesterday....last year. Maybe Les will bring in Stan to be the assistant GM then to help convince JVG to send Howard out.
I understand that. But when we choose a denominator, we want to have something that reflect the reality. That's why guys like John Hollinger uses "per 40 mins" for players. I was suggesting that Scalabrine is a bench player, "per 15 mins" will better serve the purpose.
Yes if that's what the results showed, but Morey said that they did the comparison of other players paired up with the Nets core as well and compared that to how Scalabrine paired up with the core. This is what is stated in the report referenced: He only displayed the results for Scalabrine. Regarding the choice of a per minute basis, the question is which denominator is going to help you understand the relative difference between the things you are comparing any better or not. For instance, let's say we are comparing two cars traveling at different speeds. MPH is the typical basis for comparison. 70mph, 60mph - which one is faster? Easy to see here and understand what the differences in speeds mean. What if we looked at them in miles per second instead (analogy to reducing the time-segment you want to look at per-15 vs per-48)? .019 MPS vs. .016 MPS. Much harder to see the difference, in fact you might think that the cars are travelling just about the same speed. Per 40 or per 48 min baselines seem to be accepted for some reason. I don't know if there's some mathematical basis for this. Maybe someone can shed some light on the topic.
I think Hollinger uses "per 40 mins" because (a) using the same denominator for every player "levels the playing field, and (b) you end up with numbers that our brains can quickly evaluate when you choose 40 minutes as opposed to, say, 1 minute. For the purposes of analysis, the number you choose for the denominator shouldn't matter as long as it's consistently used for all players you're comparing. How would you compare a player who averaged 15 points in 15 minutes a game to a player who averaged 20 points in 40 minutes per game? If you were going to sign one or the other based on that information, would you take the second player?
"per 48 mins" is obviously the game time of 4 quarters. In this case, we always need to scale down the numbers to get the actual numbers. "per 40 mins" is the closest multiple of 10 to the average starters' mins. And these numbers are also closest to the actual numbers.
http://www.nydailynews.com/sports/basketball/story/404987p-342993c.html i wonder if this is really the reason they are bringing him here
If Morey is such a basketball wizard, he should have picked up Kristic instead of Scalabrini after last year. Oh, but you can't analyze playing ability if the player is young and not playing. Les could have bookmarked 82games.com and received the same stat info as Morey will give him.
What would prompt you to even think this was a possibility? Kristic is under contract until 2008. The nets weren't going to let him go. They have been high on him for years, and he is showing why.
Actually it doesn't make any sense at all. The logic is pretty weak. It makes for great rumor fodder but it's pretty lousy analysis because it totally ignores other information that would lead you to a different conclusion as to why Morey was hired.
How about explaining how you think the logic was faulty? How about referencing this "other information" you speak of. The proof, of course will be in the pudding, but your post was ridiculously weak.
