Tuesday, November 17, 2015

Kind of Amazing


That's an animation of NBA movement data, which apparently you can get via a free API.  Savvas Tjortjoglou goes into detail here.  Who knows what sort of predictive model you could build exploiting this data.  Thankfully the NCAA doesn't have anything of the sort, or I'd have to quit my day job.

4 comments:

  1. I'd be curious to see if there's bias towards one side of the court for teams.. it'd be interesting to classify offenses as L/R biased, as well as team defenses who allow more points from right/left. Tons of stuff one could do with that data.. but you're right, I start to get overwhelmed just thinking about all the ways to implement that data.

    ReplyDelete
  2. That's awesome! I wonder how they generate that data? Overhead cameras and some good computer vision?

    The potentially awesome part about it is that it might overcome the fact that team stats are generally 'outputs' in the same sense that the score is an output. At least that's my perception from my own NCAA stats analysis. With this level of fidelity, the team match-ups of skills that generate the stats and the scores might actually be able to be measured. Even just for examination of missed opportunities this data is awesome. You could grab all the plays with a break-away steal (for example) and find similar play set ups that didn't take advantage of it and see what there is to learn.

    ReplyDelete
    Replies
    1. I just saw that the stats.nba.com explains the camera tracking here: http://stats.nba.com/tracking/#!/player/

      Delete
    2. Yeah, this is the Sports Vu stuff. People with access to the data have been doing some pretty cool things with it. I'm not sure if the NBA is making all this data available over the Web (somehow I doubt it).

      Delete