Donatas Motiejunas the 8th best rim protector in the NBA. Dwight 3rd.

What do you think about these stats then? Are they skewed by computers?

12-8 record
19th in NBA in points allowed

Those are Chicago’s numbers so far this year. Without any computer manipulation. What do those numbers tell you about Noah’s defense?

 

Donuts has improved but his presence in the paint doesn't alter shots change the opposing teams offense. Players still go at him and by him.

 

I would like to add that what we see of SportsVU's data is but a tiny tiny subset of the data it generates. SportsVU can actually tell a user the average distance of the nearest defender on every shot by every player in the NBA. It is that good. Another example of a similar technology is the facial recognition searches on the footage of every one of the hundreds of cameras distributed around London. Yes big brother is out there and he is coming to get you.

 

Yes but that data clearly shows that opponents make less fg% at the rim than their average when dmo is guarding him

 

Nothing against you, but between actual SportsVU data and a fans eye test, you know which side I am going to rely on.

 

From one guy with a math degree to another, you know very well that all data no matter how advanced still has an error term. I think heypartner is wrong. But I understand the reservations some people have.

Even still, as we've talked about many times. The sportsvu data is the best in it's class, so any error (you and I both know all about this possibility so no reason going over it) is minimal.

That's all there is to it really.

 

Fair enough. But if there were significant errors in the data I have to believe that NBA front offices would have caught them and had it fixed. Morey and his staff have had more than two years to analyze its efficiency. Of course there will be 50-50 calls that a specific analyst might disagree with. But those are always going to a small percentage of the data set.

I believe people just do not understand the technology yet. Therefore they don't trust it. I looked into SportsVU a few years ago when it was first introduced (at one time it was installed in just 4 arenas). I am doing my best to explain it in layman's terms.

 

I generally agree with this. Especially the second part. And as always with data, interpretation is one of the hardest parts.

 

What I do not want to see in four months time is some CF member with 2000 posts virtually spitting in my face screaming "my fan eye test is better than SportsVU". Do I have any hope whatsoever of getting there?

 

Probably not. Lol. It's the Internet. That just comes with the territory.

 

Let me try to explain my point "in layman's terms."

SportsVU data does not say "best rim protector", where does it say that? It's not like having a filter for Brightest Color or Fasting Object, which can scientifically be measured. SportsVU data does not have a button to show "best rim protector." It is the analyst's interpretation of the data that tries to say that. We want to find that answer in the mounds of data; but have we?

All I'm saying is the definition of "best rim protector" based on that data is wrong. I'm not saying the data is wrong; I'm saying we are wrong in our application of the data.

fair enough?

 

When you say rim protection, do you define it as all the shots regardless from where in the court they were taken or only shots taken at the rim? Then are you concerned about number of attempts, made, fg%, or simply number of blocks?

 

are you asking for my defintion or the definition in the OP? Does the OP even provide a clear definition?

 

I would like to post this find by Commodore.



Yes SportsVU data can be validated using other statistical sources. And by the way that o/c -8.1 is worse than when Harden sits.

 

It does not matter who goes first

For me these are the shots taken at the rim and fg% is the main concern

What's yours?

 

I agree 100%. "best rim protector" is the OPs words not mine. I would never frame an argument so sloppily. You can look at my Bev defense post to see how I did it. Present the data, draw reasonable conclusions and let the reader decide. And then attempt to explain the methodology. I am here attempting to explain SportsVU and not defend bmd. Trust me, I would never defend bmd.

 

We are not questioning whether SportsVU data is valuable. Of course it is. We are merely asking is there a scientific means to turn that data into "Best Rim Protector?"


EDIT: just saw your last post above. looks like we agree .

 

Because the guy he is guarding has a 47.5 FG percentage at the rim. And there are 16 guys who are better than that.

 

I have never ever put D-Mo in the same class as Howard, Davis or Jordan. I have always used Hibbert and Omer as comparables. Davis and D-Mo differ so much in style that I as a fan do not have the knowledge to compare the two types. I am only here to attempt to explain SportsVU data.

 

It actually is a bit more complex than that. Memphis and Houston have different defensive philosophies. Houston attempts to funnel the ball handler to the sidelines and baseline. That limits ball penetration. Memphis on the other hand is freer at letting the ball handler to the middle. They then funnel him into the Gasol-Randolf area. If you watch Houston's defense closely, they let very very few ball handlers penetrate into the paint. McHale's defensive strategy is well considered. Memphis does not have Bev and Ariza level defenders on the perimeter. Tony Allen works, but cannot he sustain big minutes. Things like that make a defense.