Saturday, January 19, 2013

Javale Making it Rain...on the basketball court

I thoroughly enjoyed the G-chat between Jack and Paul where they contemplated which was a better end of quarter strategy with 3 seconds or less left - having your best shooter try a contested 25 footer, or your center (or other worst shooter that would be uncovered in that situation) shoot a wide open 3?

I'm here to address this argument. My first thought was "no effing way Dwight Howard or any other center should ever get a look from 3." Luckily, I have data to make my opinion for me! Using Pro Basketball Reference's shot finder, I found the expected value of just the situation that Paul described. I limited it to all 3's taken in the first three quarters (because only Vinny Del Negro would run a play like this at the end of the game) of games last season, where there was 3 seconds or less left in the quarter. I also filtered by shots inside 30 feet to exclude half courters and other obviously low percentage shots.



The results aren't pretty: players taking these shots average around 22.1%, for an expected value of .663 points per play (nice guess Paul!). So a center would only need to hit 22.2% of these shots for this to be a viable strategy. How do they stack up? Actually while we're here, who are the best and worst players in the league in these situations?

I chose to exclude players with less than 5 attempts to eliminate some small sample bias - I realized my analysis was flawed when I saw my main man Andray Blatche sitting pretty at the top of the list with a 100% shooting percentage on 1 for 1 shooting. As my high school football coach used to say: "Even a blind squirrel finds a nut sometimes." Ahem. Moving on.

The winner, as Knicks fans will be thrilled to hear, is Mr. Discount Double Check Steve Novak. Steve is quickly becoming the favorite token white guy on this blog, if he's not already there. Steve shot 5/8 in this situation for a percentage of 62.5% - probably a little bit of small sample bias in that number but it's clear that Novak is a great shooter so I don't have a problem with that conclusion. Moving down the list, we have Kevin Love (not surprising), Russell Westbrook (a little surprising), DJ Augustin (DJ Augustin!!!), and Gary Neal (not surprising at all given that Popovich probably gets him an open look every time).

Obviously there were a bunch of random players tied at 0%, so I had to give the least valuable player title to the player with the most attempts. The LVP of this list, which thrills me as a Terps fan, is the Doofus from Duke himself, Shane Battier. Shane missed all 12 of his attempts in these situations. I think Joel Anthony or even Spoelstra himself might have hit 1. Honorable mentions to JJ Reddick/Chris Paul (both 0/9), Wesley Matthews (1/11), and Paul's boy Jamal Crawford (2/12).

Back to the reason for this post - how did centers fare? If I include all shots taken by centers, not good - about 14.7%. But it doesn't seem fair to include a 29 footer from Marc Gasol that he had to take because Mike Conley fucked up a pick and roll. If I use my previous split (5 or more attempts), they actually do very well - 27.7%, but that only includes 3 players. Not enough data there. If I include all players that took 2 or more shots, the average is around 20.8%, which is pretty close to the overall average. I think that seems like a fair way to look at it.

So what is my conclusion? I think this is a viable strategy as long as your center isn't a complete jabroni (If Joakim Noah shoots a 3 under any circumstances he deserves to be disemboweled). By running a play to get a center wide open, you have essentially the same expected value as a contested shooter taking the final shot. Plus if that strategy is in your playbook, your center can work on his 3s in practice and your expected value would likely exceed the league average. And there's always the possibility of Dwight Howard doing this. That's a win-win in my book.

Knibb High football rules!

No comments:

Post a Comment