SABR Analytics Conference: Day 1

So, for those of you who don't know, I am currently at the 1st ever SABR (Society of American Baseball Research) analytics conference in Mesa, Arizona.  The event is held in the Hilton hotel beginning Thursday March 15th ending Saturday the 17th.  This is obviously a niche event.  Great baseball minds who believe in and continue to progress sabermetrics.

The conference began at 1:30pm Thursday afternoon.  Unfortunately, hotel registration began at 3:00pm so most of us attendees left our luggage behind the front desk and scurried off to the conference registration table.  After receiving my "official" badge and conference schedule I walked into the ballroom for the opening event.  I walked into the ballroom noticing the oddly decorated carpet and (performing quick multiplication in my head) determined that the hotel had set up about 220 chairs with a podium, stage, and projector screen in the front of the room.  Knowing nobody, I moved along the aisles trying to find an open seat with some room for my backpack.

 I sat down and took in my surroundings.  The average age of the conference participants had to be around 35 years old.  It surprised me to see a significant number of older men, some of whom I assumed had been fans of sabermetrics since the 1970's when the guru of the craft, Bill James, began publishing his findings.  On the other hand, with information becoming so rapidly available, many of the older attendees might have recently picked up the analytics side of baseball because of its growing importance to the game.  In addition to the "dinosaurs"of the room scattered around, like sprinkles on a cupcake, were younger participants like myself.  I checked them out, figuring that most of them attended the conference in order to participate in the case competition in which a number of college age student teams compete against each other regarding a hypothetical baseball operations decision.  Going along with the demographics theme, I also noticed one or two female attendees, not something a bunch of baseball nerds would expect, but a welcomed sight amidst a room full of men.

The first panel featured John Dewan, Dave Cameron, and Cory Schwartz.  Dewan works for Baseball Info Solutions and has been a sabermetrician since the founding of the movement in the 1970's.  To his right sat Schwarz, a middle aged goatee sporting guy who wouldn't have been recognizable on any street in any city save his official MLB.com polo shirt.  Cameron, the youngest of the three, sat to Dewan's left.  The moderator for the panel, which focused on the "Changing Face of Baseball Data" was none other than Sean Forman, creator of www.baseball-reference.com.  Search any MLB player and almost assuredly the first link will be to Forman's website.  I won't bore you with the details of the panel, but overall I found it fascinating.  The highlight of the discussion occurred during the question and answer portion when a women in her 40's stepped up to the microphone to ask a question.  She prefaced her query by saying that she trained as an astrophysicist but loves baseball and attempts to use her background in physics to perform research on baseball.  I can't even remember her question, mostly because her presence at a meeting such as this astounded me.

Following this panel I remained glued to my seat waiting for the next talk.  In this discussion, Vince Gennaro, the president of SABR, spoke about his top 10 value plays for building a roster.  He spoke about inefficiencies in baseball and ways to exploit them in order to build the cheapest and most productive roster possible.  Although some of his analysis sounded like statistical mumbo jumbo, two points of his resonated with me.  First, he explained that some teams use a platoon advantage when building their roster.  The best players cost the most money, so when replacing the value of a top player, the cheapest option may be to sign two players to the same position and have them switch off different games.  A current example of this phenomenon is the Phillies attempt to replace Ryan Howard for the first few months of the 2012 season (see Cory Seidman's article on the issue).  Gennaro demonstrated that often times it can be cheaper and more efficient to sign two players to play a position, one left-handed and one right-handed, instead of doling out large sums of cash to a bigger name player.  Gennaro's next point concerned looking at players as assets.  Why shouldn't a team, Gennaro explained, sign/draft a player and convert them to a position that the market overvalues in order to then trade them for a player(s) that the market undervalues.  He gave examples such as Oakland recently trading closers Andrew Baily and Huston Street as well as the Padres trading Matt Latos.

Following Gennaro's talk I listened to economist J.C. Bradbury speak about pitch counts and days of rest for starters.  His paper demonstrated that a pitcher who throws more than 99 pitches in an outing is not likely to have those extra pitches affect his performance the next time out.  On the other hand, he did find that a pitchers' earned run average will rise if said pitcher throws more than 99 pitches in at least 5 and then 10 consecutive starts.  In other words, consistent overuse causes performance and sometimes medical issues, while sporadic overexertion has little effect.  Having recently written about young arms and overuse I found his findings concerning adult pitchers interesting and pertinent.  To give you some perspective, Bradbury looks like a middle-aged college professor, who speaks with a southern drawl.  At times he spoke too quickly, but he also recovered well and repeated anything in his presentation that had extra significance.

Currently it is 8:10am on Friday morning and the first panel of the day (the GM panel) begins in 20 minutes.  I leave you to digest some of yesterdays events.  Later today I will rehash yesterday evening's events including a presentation by Bloomberg Sports and the player panel including Oakland A's starting pitcher Brandon McCarthy.