Saturday, March 06, 2010

MIT Sloan Sports Analytics Conference - Part 1

I got back from the MIT Sloan Sports Analytics Conference in Boston a while ago. I almost didn't make it. I missed the exit and once you miss an exit or take a wrong turn in Boston, you can plan on being lost for the next half hour or so. I was looking for exit 20A. I went by exit 24. About 200 yards later, I saw exit 20 A but couldn't pull into the right lane in time. What happened to exits 23, 22 and 21? Boston is a great walking city and has excellent public transportation but driving there is not a good idea. And this is coming from someone who worked in the city for 13 years.

Anyway, I made it to the conference just in time to see the baseball analytics discussion. Analytics is the modern term for sabermetrics and seems to be getting used more and more. The panelists were:

John Abbamondi - Cardinals Assistant GM
John Dewan - Owner of Baseball Info Solutions
Dan Duquette - former Red Sox GM
Shiraz Rehman - Director of Baseball Operations for the Arizona Diamondbacks
Tom Tippett - Director of Baseball Information Services for the Boston Red Sox
Rob Neyer - writer

Much of what they talked about was not new to me or to many of you but it was interesting to hear it from a team perspective. I'm including some notes below on various topics.


They all agreed that defensive statistics have made great strides in recent years but also that there is more work to done. Dewan says that defensive statistics tell us about 60% of what we need to know about a player's defense. Additional factors must be taken into account. For example, how much of defense is due to good coaching rather than player skills? A player might be in good position to make plays more often than another player simply because his coaches are good at anticipating where balls will be hit. It could also be that teams that pay more attention to defensive data have their players positioned better. The new Field f/x cameras to be used at all MLB games this year should improve defensive evaluation.

Dewan talked about how team's got away from fielding a little bit during the high powered home run derby era. He repeated a theme that we've heard frequently lately - defensive players have become undervalued and some teams are taking advantage of it. They have replaced low average high OBP hitters as the new Moneyball players. He mentioned how the 2008 Tigers were built on hitting and had some players playing out position. They then added defense at a relatively low cost the following year. The Mariners and Rays used a similar approach in recent years.

This year, the Red Sox have added defense at the expense of offene. Tippett was quick to point out though that it wasn't a change in philosophy as was reported by the media. The Red Sox didn't suddenly discover that defense was important. There were just more opportunities to acquire defensive players at the right price compared to offensive players.


There was a general consensus that more needs to done in the area of projecting player performance. Tippett thinks that player projection work fairly well for hitters but that it's important to learn more about how to identify exceptions. Player projections are built off the performance of similar players in past seasons. However, there are always some players who don't fit a mold very well. How can we better identify these players?

Dewan estimates that the Red Sox have improved their team by 60 to 80 runs over the winter considering both offense and defensive. That, of course, assumes that players perform as expected. Rehman brought up the point that defensive statistics are harder to project than offensive statistics.

More later.

1 comment:


Blog Archive


My Sabermetrics Book

My Sabermetrics Book
One of Baseball America's top ten books of 2010

Other Sabermetrics Books

Stat Counter