Interesting Article

Any And All Things T-Wolves Related
Post Reply
User avatar
The Rage Monster [enjin:8010341]
Posts: 330
Joined: Fri Nov 29, 2013 12:00 am

Interesting Article

Post by The Rage Monster [enjin:8010341] »

http://fivethirtyeight.com/features/the-hidden-value-of-the-nba-steal/

A lot of great stuff here, I'd recommend reading all or most of it. A real quick summary for people who don't have the time (but I'd still highly recommend reading the whole thing):

First the guy uses stats to compare the value of points, rebounds, assists, turnovers, blocks, and steals. His statistical model found that a rebound = 1.7 points, an assist = 2.2 points, a turnover = 5.4 points, a block = 6.1, and a steal = 9.1.
Second he looks at how easily a certain stat can be replaced by another player. He found that assists are 88% replaceable, points 83% replaceable, rebounds 83% replaceable, blocks 32%, and steals 4%.

Not only does this article use Rubio and Love as it's primary example I think it is highly applicable to our discussions about the center position. Looking at the two most important categories, steals and blocks, per 36 Pek averages 0.6 steals and 0.5 blocks while Dieng averages 1.4 steals and 2.4 blocks. According to the article, in those 2 categories alone, Dieng is worth 18.87 more points than Pek.
User avatar
Camden [enjin:6601484]
Posts: 18065
Joined: Tue Jul 09, 2013 12:00 am

Re: Interesting Article

Post by Camden [enjin:6601484] »

"Interesting" was the right word choice. Not sure we can really take something like this and be objective about it, but it's something to think about nonetheless.
User avatar
The Rage Monster [enjin:8010341]
Posts: 330
Joined: Fri Nov 29, 2013 12:00 am

Re: Interesting Article

Post by The Rage Monster [enjin:8010341] »

Camden wrote:"Interesting" was the right word choice. Not sure we can really take something like this and be objective about it, but it's something to think about nonetheless.


I like the angle he took. Most of the advanced stats aim towards coming up with one overall stat that is then used to compare players. In my opinion these tend to be unreliable and heavily skewed in favor of more offensively minded players. What I liked about it and thought was most interesting is that his numbers are the same regardless of the player. The conclusions are simply what statistics are most important to the and which statistics are the hardest to replace. The only time I think he might have gone a hair far was when he made everything relative to 1 point.
User avatar
Lipoli390
Posts: 15656
Joined: Mon Jul 08, 2013 12:00 am

Re: Interesting Article

Post by Lipoli390 »

I agree with Rage's take on the article. Actually, the most compelling part of the article wasn't the analysis of the relative impact of steals compared to other stats. Most compelling were the author's stats on Rubio's impact on wins and losses. That's not a particularly esoteric analysis and it confirms I think what most of us knew already. Bottom line is this. Love and Rubio (both together and separately) have a substantial net positive impact on our team's ability to win games. In other words, they are impact players. And they are really our only two impact players. So what is the take-away from all this? I think it confirms that salvation for this team lies in the ability of our front office to shore up the team around these two pillars over the summer. It won't be easy, but I think it's doable. It will depend on Flip's use of our lottery pick, the MLE and the few valuable trade assets we have -- primarily Pek but to some extent the expiring contracts of Barea and Luc.
Post Reply