Article: The Hidden Value of an NBA Steal

[West is Best]

There was an interesting series of articles that came out of ESPN's Five Thirty Eight sports blog, which focuses on number crunching. The premise of the articles is while points and rebounds are often replaceable (eg if Player A doesn't get a defensive rebound, there's a good chance his teammate will) steals are not. The article goes into a look at the value of a steal and whether its actually the most important stat when it comes to determining wins and losses.

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

Two follow up posts talk a little more about the calculation and how it matches up to the eye test in the NBA:
http://fivethirtyeight.com/datalab/are-nba-steals-born-of-bad-defense/
http://fivethirtyeight.com/datalab/wait-the-76ers-get-a-ton-of-steals-and-they-stink/


They're good reads, but the two most interesting (and debated) findings are shown in these charts:
http://espnfivethirtyeight.files.wordpress.com/2014/03/morris-predictive-ability-1.png?w=610&h=291

http://espnfivethirtyeight.files.wordpress.com/2014/03/morris-steals-2-v2.png?w=610&h=268


Where I find this most interesting is in the case of Dee Davis. Several posters and friends of mine have remarked how Dee's stats don't really correlate to the eye test. Dee was team leader in steals, just ahead of Semaj and will likely be the far and away leader on the team next year. Any thoughts on the article, especially in relation to XU? Do you think this analysis would translate over to the college game?

[PMI]

It's interesting, and I don't have the numbers to back my theory up, but I don't think it would necessarily translate to the college game. I think points, assists and rebounds being "replaceable" is more the byproduct of the players at the NBA level being so good. There are certainly elite rebounders, passers, scorers, etc., but generally speaking, someone or some combination of players will account for those numbers. Steals and blocks are obviously solely defensive stats, and there are only so many elite shot blockers and pocket pickers in the league. From the standpoint of a Wizards fan, we don't have a shot blocker, but we have a few guys who are really quick on defense and can create turnovers, and when they're producing, it leads to a lot of fast break points. It's one of the few things we do better than most teams and we're certainly at our best when we're defending that way. Getting steals is sometimes a product of a certain kind of system, I suppose, but guys like John Wall and Trevor Ariza are highly skilled at anticipating passes and using their quick hands. If we had to replace them with most other guys at their respective positions, they wouldn't rack up steals at the same rate, but offensively, the ball would be in someone else's hands who can likely score and at least rebound in an area. Those kinds of stats, from a team standpoint, seem to stay fairly consistent over the long haul.

In college, I think systems and a relative lack of talent across the floor at all times (for most teams) make for a different story. There's a reason you generally don't press an NBA team no matter how good you are at forcing turnovers... the pros are much, much better at protecting the ball, and they create easy offensive looks after breaking pressure. We saw with Xavier this year how quickly turnovers can rack up when you have trouble protecting the ball, and some teams use systems that look to exploit that. You're never going to see a Shaka Smart or Nolan Richardson type pressure defense used in that capacity in the NBA. It just wouldn't work.

[GoMuskies]

Rick Pitino tried....and yeah, it didn't work.

Maybe if it was a 15 second shot clock things would have gone better.