This past weekend I attended the MIT Sloan Sports Analytics Conference in Boston. While I was there as a member of ESPN’s True Hoop Network, their network of basketball blogs, I was also able to take in a handful of baseball-related panels and presentations.
Sloan generally doesn’t have a huge baseball presence, for a couple of reasons. For one, SABR Analytics 2013 is this coming weekend, and I think a lot of baseball folks try to focus their energy on that equally excellent, baseball-specific event. In fact, three of our BTBers will be heading there, and I’m really looking forward to hearing what they have to say afterwards. Still, Sloan features some sport-neutral analytics panels, a baseball analytics panel and featured a pair of baseball presentations based on research papers. Throughout this week, I’ll review and respond to some of the key events from the conference.
I also [shameless self-promotion alert] wrote this piece for ESPN’s MLB page on the future of analytics in baseball. You should definitely read that piece, which focuses on a panel that involved Jonah Keri, Baseball Info Solutions’ Ben Jedlovec, Farhan Zaidi of the Oakland Athletics, Joe Posnanski and Voros McCracken, who likely needs no introduction around these parts. I won’t spoil that piece, but the main theme to come from the panel was that baseball analytics haven’t nearly peaked or begun to realize diminishing returns.
In fact, this was a major theme of the entire conference -- despite the fact that analytics have made huge strides in breaking down barriers, informing the public and permeating into everyday sports vernacular, there is still a lot of work left to be done. And even if we had all of the data in the world, there are still 30 teams interpreting it all in different ways -- the dissemination of raw data does not mean a loss in competitive differentiation.
For baseball, said advancements have a lot to do with the visual tracking systems that we’re all very excited about, such as Field F/X and Hit /FX. The panel didn’t get into too deep a discussion about the specifics of where these could go, but an overarching curiosity among the panelists was how we will come to evaluate defense. If Brendan Ryan is a Gold Glove-caliber defensive player, how important is that? Since we all put a lot of stock into Defense-Independent Pitching Stats, we have a long ways to go to understanding the defensive impact on the game. Field F/X and Hit F/X are exciting, albeit proprietary (for now), means of studying these elements further.
But beyond this tracking data, there are dozens of ways teams can leverage new analytic tools to improve their ball clubs. One such discussion focused on injury detection and prevention, which I’ll talk about in greater detail later in the week based on a panel that involved Stan Conte of the Dodgers.
Another had to do with how minor league players are scouted, drafted and developed. How is it that Albert Pujols and Dan Straily slip through the cracks? Or did they not slip through, there was just something done in their development to fundamentally change their path? There is so much that we don’t know about how players get from Point A to Point B that we can still try to uncover. A lot of this may end up being qualitative, and an important aspect that Zaidi pointed out was that our focus could shift from outcomes to skills and, while he didn’t use these words, on finding out how players hit their "80th percentile projections" instead of just their Major League Equivalent stats. With all of the development time allotted to baseball players relative to other sports, along with the sheer volume of prospects and players, there is a plethora of work that can be done in this regard. Understanding MLB outcomes better still doesn’t mean any of us understand the baseball ecosystem.
One other interesting thing that came up, and this was really just A’s-specific, happened as Zaidi answered a question from Keri on their roster construction. He said that the the A’s have seemingly taken on a strategy of "managing the roster from the bottom." That is, since their smaller budget makes it difficult to improve at the top and get, say, a seven-win player to replace a three-win player, they’ve instead try to tack on gains at the bottom. So, say, improving the back-up catcher expectation by half a win at a low cost, or improving the fifth arm in the bullpen by a partial win, and so on, to make cumulative gains at a substantially cheaper price. This isn’t really "news" but it is an indication of how the "Moneyball" term gets misconstrued as being about OBP and the like, when really it’s about the continued pursuit of market inefficiencies to exploit.
These five hyper-intelligent and entertaining guys touched on a handful of other topics as well, but I’m not here to simply review, word-for-word, what was said. These were some of the key messages espoused that I thought it worthwhile to pass along. In the coming days, I’ll also be writing on injury detection and prevention, paired pitching and the value of flexibility, so check back.