This past weekend, I had the opportunity to attend the Sloan Sports Analytics Conference in Boston. I got to listen on a variety of excellent panels, meet some of the biggest names in sports analytics and participated in the inaugural Sloan Hackathon (more details on this later). I'm going to give a recap of my experience, and you guys should be able to learn a lot of the same things I did this weekend.
Today was the Hackathon competition. I was one of the final 20 selected among the 300+ undergraduate and graduate students that applied. Being a junior undergraduate at the University of Michigan-Ann Arbor, it was already a tremendous honor to be selected to the top 20 as an undergraduate. We had 3.5 hours and a one minute presentation to present our idea. Mine was a tackling efficiency and range metric, while others provided different visualization based projects. Also, today they chose the finalists for the Hackathon, all of which get invited to a dinner with ESPN. I found out that I had been selected as one of the three finalists!!! I was the only undergraduate student in the top three, the other two students were a first year MBA student and a pHD student. What a crazy start to the weekend! More details to follow after this.
Today was the first day of the actual conference. I had the chance to watch some excellent panels and learned a lot about the current state of analytics. Most of the hockey based panels were on Saturday, but a couple of other panels certainly are relevant for hockey fans. As many readers of this site are certainly aware of, hockey is behind sports like basketball and baseball in terms of analytics and technology. Hockey is the next sport, along with football and soccer, to begin making technological breakthroughs. Recently, NHL.com released an enhanced stats page, which gives writers and fans access to statistics not seen previously. I will be doing more analyses of these statistics, how the Canucks are doing in relation to the rest of the league, and talking about what advanced statistics are currently being used in future posts. The next big breakthrough for hockey, in my opinion, is player and puck tracking technology. With hockey being such a fast and physical game, sportVU cameras are less optimal for hockey, in my opinion. Tracking technology such as Zebra technologies has proven successful for the NFL, and would do well in hockey as well. Currently, without this technology, it is challenging to make meaningful discoveries in hockey because we need larger sample sizes of data. With so many great companies such as SAP, Catapult and Zebra, hockey analytics will have its revolution in the near future.
Today was the best day for me personally and for hockey related panels. In the morning, I had the chance to see Toronto Maple Leafs assistant general manager Kyle Dubas talk about how analytics helps us limit cognitive biases in player evaluation. Even though the Maple Leafs aren't doing very well at the moment, Kyle talked about the process of building a team, like he did with the Soo Greyhounds in the OHL, and said that analytics are not magic and it is a long process to create a great NHL team. Unlike other sports, guys that are drafted rarely contribute until after a few years after they are drafted. Kyle's brief talk mostly mentioned that although hockey still remains an eyeballs business, where you have to rely on what you see on tape, analytics can play a key role in limiting the cognitive biases associated with human observation. Biases such as confirmation bias, which for hockey purposes means, if you liked a prospect a lot before watching an additional game, you are more likely to remember the good things the prospect showed on tape, rather than the bad things. With analytics, teams can limit the effect of these cognitive biases.
Calgary Flames GM offered a more cynical view towards analytics. He said that "the notion that you can sit behind a computer and find athletes is bullshit." Burke did mention that analytics do play a role, but it is certainly not close to being a replacement for the eyeball test.
Personally, I was waiting nervously for the Hackathon award results. After anxiously waiting for Ben Alamar to announce the winner, and he announced "the winner of the Hackathon, is Shawn Farshchi, University of Michigan" (me). WOW! I could not be more grateful for this opportunity and I finally had my chance to make a name for myself in the industry, at least among students. It is so crazy how everything came together. I had not been posting here as often as I wanted to, since I was prepping for the competition, but in the end it was all worth it!
My Biggest Takeaway From The Conference
The biggest theme throughout the conference was the issue of communication of analytics throughout the organization. Often, it is difficult to pass information from analytics guys to the front office, let alone spreading that information from the front office to the coaching staff to the players in a way that everyone can understand. With new technology on the way, it is critical that teams can communicate these new ideas in a clear and succinct manner. Analytics will never replace the eye test, but they can be an important means of gaining a competitive advantage. Overall, this weekend was amazing and my outlook has changed beyond what I could have ever imagined.