For Orioles Hackathon participant, an 'impossibly complex' task: Make sense of data

When I reached out to the Orioles about Friday's Baseball Analytics Hackathon at Camden Yards, I was disappointed to hear back: No comment. Here were some of the country's best and brightest amateur baseball geeks, headed to Baltimore for a possible sabermetric revolution, yet seemingly shrouded in secrecy.

Friday came and went, with tweets from the event shedding some light on the goings-on, but nothing particularly revelatory. I was feeling glum. It was like I'd missed out on a really cool party.


Then Cincinnati's Aaron Lehr read my mind, got to a computer Monday and decided to blog about his day, taking readers inside the coding marathon. His must-read post has a relatively happy ending, too: He and his team finished second at the Hackathon and won Orioles tickets. Here's what else is worth knowing.

1. Almost 300 people applied for an event with an open-ended mission.

Of those 300 applicants, the Orioles and consulting firm Booz Allen Hamilton hosted about 90 participants Friday. Their goal, according to Lehr: Given PITCHf/x data for every at-bat over the past three years, "do something with the data that was creative, actionable, and functional."

Oh, and do it all in eight hours. No big deal.

2. Baseball data can be "impossibly complex."

With a treasure trove of information, you can try to do almost anything. This can be more than a little paralyzing. "Baseball can be impossibly complex," Lehr wrote. "It's always tempting to take a few observations and turn them into some sort of axiom that can be applied to all future situations. But there are so many outcomes and so many variables that unless you look at a ton of data, it's hard to make definitive conclusions."

Lehr and his teammates — his girlfriend and close friend — improvised along the way before settling on a worthwhile focus: context-rich pitcher-hitter matchup information. In other words, don't worry so much about the outcomes of at-bats against Kevin Gausman; scrutinize the location and movement of pitches that preceded them instead.

3. If you can make this in eight hours, imagine what the Orioles (and the rest of baseball) are on the brink of.

Lehr's final project looks like something a college student might take an entire semester to produce. And again, he and his team did it in eight hours.

It's unclear how the Hackathon serves the Orioles' interests, but they do seem to be at the vanguard of this particular movement. After a quick Googling, I couldn't find any other teams advertising similar events, at least publicly.

There are more stories of "bad" hacking in the baseball world than ever, thanks to the St. Louis Cardinals. It's fair to say the game needs accounts of "good" hacking as much as the Orioles need another starting pitcher. Here's hoping that the tide starts to turn in Baltimore.

(H/T to Aaron Lehr)

Recommended on Baltimore Sun