Possible that randomly by chance that many of the games where the hack a Dwight strategy has been used are games that Lin has started? Kind of late so I don't want to sift through the box scores and I could be totally wrong, but for some reason it just feels like this has been the case.
It's odd frankly, just guessing but I think Jones activity around the court and Lin Lin alleviating some pressure off Harden helps them, Parsons has the most rediculous drop in efficiency though and maybe seeing as we don't run enough PnR as desired with Dwight could explain the dropoff.
The reasoning behind such performances would lend to thinking that Dmo and Beverley should start, in order to maximise the effectiveness of Jones/Lin/Dwight/Parsons. Lin/Jones off the bench with the starting 5 of Beverley Harden Parsons Motiejunas Howard Would seem ideal based on maximising everyones effectiveness.
Makes a lot of sense! I think so highly of Jones that it literally never occurred to me to bring him off the bench. But with this lineup, harden, howard, parsons could run the offense. DMO could be a stretch four, play solid D, get rebounds... etc. And Lin, Jones, Casspi... + backup center and backup sg we still need. Man... we really need a backup center and SG.
Rockets are 12-5 with Lin and Harden starting in the backcourt. Last year, the team made the playoffs for the first time in 4 years with Lin playing alongside Harden for 70+ games. Lin has played in 13 games as a backup to Beverley (the team's record is 8-5). Beverley and Harden have played together w/out Lin in 12 games and the team's record is 7-5 (total record w/Bev-Harden backcourt: 15-10). So the team has a better winning percentage with a Harden-Lin backcourt even though Beverley has not been available in those games. Harden has put up several big games with Lin as his backcourt partner. Having another playmaking guard alongside him has not hurt Harden one bit. Lin has had strong games and some lower scoring/less active games, but his overall averages have been pretty consistent. If Harden is your dominant ball handler and scorer, the guard playing alongside him isn't going to have a 15-20 point night every night or get 15 shots a game. The question is whether that guard hold his own and generate offense when needed and can he complement the primary scorer. The stats show that Lin can do the job. From a matchup perspective, opposing teams have a very hard time guarding both of these players when the ball is moving and finding the best matchup to exploit or the hot hand.
It's odd, I can't really process why Parsons see's such a massive drop off with Lin on the floor, Jones make senses given his activity, running the break, decent pick/pop/roll threat, Lin alleviates some of the pressure off harden this the slight bump, but Parsons/Dwight see a noticeable dropoff
Are we trying to analyze ramdom fluctuation. Good read: Fooled by Ramdomnees by Nessim Taleb. It's in paperback at Amazon.
Let's face it people, the fact that harden is a ballstopper that makes our team play suffer. It's only effective when harden is getting the calls, (which is not happening anymore) and hitting treys.
When Lin/Howard are on the floor, I notice Howard shoots TONS of FT's because Lin always finds him, then he proceeds to get hacked. True story.
So the new spin is that Lin "finds" Howard more, who gets to the line more, who then misses his free throws, LOL. All of this easily measured by Panda's data which shows a clear drop in PPS and PPP by Howard when Lin is on the floor.
Well in the isolated case of just Lin or Bev - I would think Howard would function better on the offensive end with Lin. Not sure how it fits with the whole 5 man unit though. Seem like we can conjecture and theorize forever on these small sample size. Can we just agree that the stats are inconclusive? I think we can all agree though that Lin is pretty damn good as the primary ballhandler. So we should all just be happy that we are not hopeless if Harden is hurt. Hopefully the coaches and players can get everyone to maximize their abilities (that onus is on Lin too). At worst, we can feel better about giving Harden a few more minutes of rest per game and have Lin come off the bench and run the show for 10-15 minutes a game.
Not posting these numbers to show that Lin should leave Houston or anything like that. Just interested to know how consistent Lin can be when he is given the ability to dominate the ball. Because we know he is a ball dominant guard. And (acknowledging the small sample size) the answer is: pretty damn consistent. So I say, let Lin run the offense at least for those minutes when Harden is sitting (or needs a breather even when he is on the floor). What frustrates me is not seeing Harden handle the ball when Lin is playing. It is using Lin as a spot up shooter, and having everyone but Lin handle the ball, when Lin is clearly the second best play maker on the team -- and had shown he is able to get other players, as well as himself, going when he can dominate the ball.
Harden has had some nagging injury problems this year and hopefully we can limit him to around 36 minutes a game. 12 minutes of Linsanity!
Do a bootstrap before you try to draw any conclusions. I tried to explain this in my response to larsv8's analysis of which the conclusions are really drawn from noise - the difference he saw in the numbers is not statistically significant at al. It is simple: instead of trying get one number, you do random sampling with replicates of the game that you use to derive the numbers. But you might need to go beyond excel sheet and use a programming tool such as R to get that. Or you can send me the raw data and I will be happy to do the assessment for you. Stats without confidence interval is junk. I can derive any conclusion from random noise.
Here is what I can agree on: 1.) In theory, since Lin is probably better on offense than Beverley, the lineups should perform better offensively. However in practice, this has not happened, and by looking at the numbers, it is because Lin takes more shots than Beverley which means shots are being taken away from Parsons, who is a better offensive player than Lin. In taking those extra shots, Lin has a tendency to turn the ball over more, which is turn makes the offense even less effective. 2.) For whatever reason, Howard does not score as efficiently with Lin on the court as he does with Beverley on the court. Why? I have no idea. What I do know is that it isn't the silly "more free throw" theory proposed above. 3.) Lin is absolutely the best option to replace Harden in the starting lineup if he is not available. In fact it is almost identical to the Asik/Howard situation. Asik is a Howard lite and a clearly superior player to Jones, Asik alongside Howard just doesn't work. Lin is basically a Harden lite, probably better than Beverley, but alongside Harden it just doesn't work. And I certainly will not agree that these stats are inconclusive based on a small sample size. We are looking 500 possession here, which is a significant amount to draw conclusions from. Is it perfect? No, but there is certainly value here because those 500 possessions for each lineup resulted in 15-25 wins or losses. I think they provide a very nice insight into what is happening on the floor.
If Lin is a much better off the ball player or if Harden plays off the ball more, ideally that can mesh and work fine. We have all seem glimpses of it. Honestly though, unless our ace is willing to change his game a bit, Lin needs to change his or ride the pine when the game starts. Either way, if Lin does start, the new question becomes, what do we do about our bench?
Right on target. Must run tests of signif. And indindicate confidence levels. We may be trying to explain noise
