The Eighth Annual Sloan Sports Analytics Conference starts Friday morning at the Hynes Convention Center in Boston. And even though it's not as baseball-oriented as other conferences (one of its co-founders is Houston Rockets GM Daryl Morey), there's still lots of interesting panels and research for the baseball fan. Since this is my third conference, I thought I'd write a quick guide telling newcomers what to expect.
If You Have a Ticket
Boston styles itself "America's walking city", mostly because it's hellish to drive in. My advice is to take public transportation: Hynes has a stop right on the MBTA Green Line (B, C, or D trains), and there are a number of restaurants and bars a block or so away*. For those taking the train or a bus into town, the convention center is in walking distance from Back Bay Station, or at least it would be if it weren't Jack London levels of cold this week.
* - Some of my favorite restaurants near there, in no particular order: Tapeo, Jasper White's Summer Shack, Stephanie's on Newbury, Pavement Coffee, and (further away) Eastern Standard. My favorite bars near the conference are probably Bukowski Tavern and the Cactus Club.
And don't forget: the dress code for the conference is listed as "business attire" and yes, everybody wears a tie.
If You Don't Have a Ticket
Even at $600 for a full-price ticket, the conference sold out months ago. If you missed out, there are still ways to get your fix. The finalists for this year's research paper award are available online, and a live webcast will be available -- albeit for a fee -- this weekend.
The meat of the conference is the panel discussions, featuring a number of owners, general managers, media members, and the occasional player. The whole schedule is available here, and bios for the featured speakers are provided here. I've listed here the panels I'm most interested in, in descending order.
- It seems like sensor technology is more prominent than usual at the conference. I counted four panels dealing with companies using wearable sensors and cameras to track players, not to mention the research papers using STATS' SportsVU system. MLB Advanced Media promises big things for MLBAM: Putting the 'D' in Data, centered around the unveiling of the "first-ever complete and reliable measurement of every play on the field". On the sensor side, Athlete Analytics will discuss sensor-based advances in player tracking and injury prevention.
- There was at least one panel last year at the SABR Analytics Conference that was dominated by Bill James spitballing possible innovations, and In-Game Innovations: Genius or Gimmick? looks like it will feature more of that. Recommended for fans of those Effectively Wild podcasts where Sam and Ben debate the merits of putting a fielder 15 feet away from the batter.
- You would think the Baseball Analytics panel would be the star attraction for baseball fans, but remember: just because Astros GM Jeff Luhnow will be in on stage doesn't mean he'll be divulging the secret family recipe. I understand, of course, but it gets frustrating to watch the front office guys on this panel dodge the interesting questions fired at them by the moderator and other panelists.
- Man & Machine: Real Time Data and Referee Analytics is usually where Mark Cuban makes tongue-in-cheek remarks about NBA officiating, but he won't be in attendance this year. Instead, PITCHf/x wizard Dan Brooks of BrooksBaseball.net will participate, along with Sportvision CEO Hank Adams and Hawk-Eye founder Paul Hawkins.
- I put The Science of the Deal: Negotiation Workshop down because it represents a big hole in my knowledge. As someone with an engineering background, I have no idea how contract negotiations work, and it's obviously such an important part of what a front office does.
Inside the War Room: Building Alignment from the Front Office to the Field deals with football, but should still provide an interesting look into how an owner and general manager work together ... If you've taken or plan to take Andy Andres' Sabermetrics 101 course (free on edX this spring!), From the Classroom to the Locker Room: Teaching the Next Generation of Sports Analytics might be worth looking into... I'm a sucker for a good infographic, so naturally I'm going to Sports Data: Seeing, Showing, Making Inferences.
An underrated part of the conference, I think, is reading the posters and listening to the research presentations, so I wanted to highlight that portion of the conference here. Here are my impressions from my quick reads through the three baseball-related finalists. An underrated part of the conference, I think, is reading the posters and listening to the research presentations, so I wanted to highlight that portion of the conference here. Here are my impressions from my quick reads through the three baseball-related finalists.
A Data-Driven Method for In-Game Decision Making in MLB (PDF)
This paper presented a machine learning model to decide whether or not a pitcher would give up a run in the subsequent inning given his performance so far this game. If you listen closely, you can hear a sound as if a million managers cried out in rage. Leaving feasibility aside, I never could figure out how the authors' model chose which pitchers to leave in and which to take out.
Can't Buy Much Love: Why Money Is Not Baseball's Most Valuable Commodity (PDF)
Mike Trout was the best all-around player in baseball two years in a row, and just signed a new contract that will pay him $1 million in 2014. A few lockers down, Albert Pujols earned $16 million last year to contribute between one and two wins to the Angels last year. The recent discrepancy between payroll and performance has been investigated before -- here's a Sam Miller article on it from last summer -- and this paper formalizes that analysis.
What Does It Take To Call a Strike? (PDF)
The fact that the strike zone fluctuates with game situation has become an accepted fact in the PITCHf/x era. The authors confirmed many of the usual suspects: for instance, that the zone expands in three-ball counts and shrinks in two-strike counts. But they also introduced something I'd never seen before: in a three-ball count with runners on, the strike zone changes more than with no runners on. The authors suggest that the presence of the runners necessitates a quick decision from the umpire, which is more likely to be influenced by the count than a more deliberate call.
. . .
Bryan Cole is a featured writer for Beyond the Box Score. You can find him at this weekend's conference, or follow him on Twitter at @Doctor_Bryan.