March 6, 2010

MIT Sloan SAC, Baseball Analytics

Rob Neyer is leading this panel. The conference underestimated the popularity of this session, as it’s standing room only.

Rob is making the analogy to cartography, how the old maps had information, but not complete information. We shouldn’t lose sight of what the old maps had to offer.

Update: Tom Tippett gets the first question on projections. He disagrees that there’s little to do with improving batting projections. Tom thinks projecting players changing contexts needs to be improved.

Update: John Dewan says defensive projections are the next big thing, and Dan Duquette agrees.

Baseball Analytics Session

Standing room only

Update: John Abbamondi and Shiraz Rheman talk about front office work on projections, and both agree it’s getting better but there is still a big “error bar.”

Baseball Panel, MIT Sloan SAC

Baseball Panel. Shiraz Rehman, Diamondbacks. Tom Tippett, Boston Red Sox. Dan Duquette. John Dewan, BIS. John Abbamondi, St. Louis Cardinals. Rob Neyer, ESPN.

Update: John Dewan is talking defense, and how much more we know now than 15 years ago. He notes the Mariners, the Tigers and now the Red Sox are all improving their defenses to win more. John says the Red Sox will gain six to eight wins with their defensive improvements.

Update: John Abbamondi says that at the start of 2009, the Cardinals front office thought they were a good defensive team, but the projections on ESPN said otherwise. John thinks they were right, given Wainwright and Carpenter’s seasons. He acknowledged that the ESPN ranking, however, made them go back and look again. He feels they are still just scratching the surface on defense.

Update: Tippett notes that the shift to defense isn’t a change in philosophy by the Red Sox. It was just the best way to build the club with the money they had this year. They are not valuing defense over offense any more or less.

Baseball Panel, MIT Sloan SAC

John Dewan, Joh Abbamondi and Rob Neyer listen to Tom Tippett.

Update: John Abbamondi is afraid FIELD f/x is going to make defensive analysis too easy, and teams like the Cardinals will lose their competitive edge.

Update: Dan Duquette is talking about keeping pitchers healthy. He concentrating on the advances in technology that are allowing pitchers to learn how to pitch properly at a young age. He thinks this will help, but since throwing overhand is an unnatural motion, he’s not sure how much it will help.

Update: John Abbamondi and Shiraz Rehman talk about internal debates between changing a pitcher’s delivery to keep him healthy or making him effective.

Update: John Dewan is talking about a study he did that suggested that pitchers who throw more than 100 pitches in games early in the season are stronger later. They get a chance to build up their stamina.

Update: How do teams handle all the data? How do they decide what’s important? Cardinals take pride in making all their information easy to access for the decision makers. Sometimes, when someone asks for a day to think about a trade, they know that team is not as organized.

Tippett agrees that good data management is important, especially as PITCH f/x and HIT f/x become more important. Tippett also notes that different parts of the orgranization need different information (like medical reports). It’s the job of his unit to make this work for everyone.

Update: John Abbamondi wants to know how to measure the makeup of a player. Will they play hard, not abuse drugs, etc.

Update: Dan Duquette want to know how to consistently develop great pitching.

Update: John Abbamondi thinks the league wide medical database will make a big difference.

Tom Tippett thinks there are lots of things we don’t know, because every day he gets a question that he doesn’t know. That was my impression working at ESPN. There’s always a new and interesting question.

Update: Shiraz Rehman want to know about bio mechanics.

Update: Cardinals build an option pricing model to help figure out how to pay free agents.

Update: How do you sell decisions based on analytics that are unpopular with the fan base? Tippett takes the question, and I’m not surprised by his answer: Winning. He notes Theo Epstein has a very good track record, which helps.

Update: That’s it, a great panel. It would have been interesting to have an anti-analytics person up there to criticize.

6 thoughts on “MIT Sloan SAC, Baseball Analytics

  1. BenS

    Since you’re adding a lot of updates as the session continues, would you consider tweeting the link again when you’re finished updating posts like this? It would make it easier to be sure we’re not missing some of what you’re covering. Thanks.

    ReplyReply
  2. Tyler Hissey

    David,

    Thanks for the updates. Having just moved to Boston, I would have loved to have attended this event but your updates are the next best thing.

    ReplyReply
  3. BenS

    Re: measuring the makeup of a player
    I wonder if using a psychological assessment test like the MMPI (Minnesota Multiphasic Personality Inventory) might help with that. They could compare results to the general population for now and gradually build up data for baseball players specifically over time. Perhaps MLB could require/request all or certain draft eligible players to take the inventory (like the NFL & Wonderlic). Cultural factors (such as so many Latin American players) could pose some problems, but my guess is that those have already been researched by some cross-cultural psychologists.

    ReplyReply
  4. Pingback: » Big Barrel or regular size baseball bat? and whats the difference?

  5. Pingback: A NEW EMPHASIS ON DEFENSE? | Pater Familias

Leave a Reply

Your email address will not be published. Required fields are marked *