While we are eagerly awaiting the arrival of abstracts for the Research Papers Competition ahead of the deadline this Friday, 9/28, we sat down with Philip Maymin, who took first prize in last year’s competition. In his paper, “An Open-Sourced Optical Tracking and Advanced eSports Analytics Platform for League of Legends,” Philip introduced new metrics to track League of Legends performance and used these to quantify an individual’s impact on their team’s chances on winning.
Your paper introduced new data and metrics for esports, can you give us an update about what’s come out of your paper since publishing? What has been the response been like?
Quite a number of professional teams, individual players, universities, and researchers have reached out, so there’s been a huge amount of interest. But esports analytics seems to be about where sports analytics were a decade ago: hardly any budget, and enormous barriers to integrating it into the overall decision making process. And just like sports analytics a decade ago, the first few entities to invest seriously and heavily both in terms of resources and culture into esports analytics will probably become dynasties.
What was your experience like presenting at SSAC18? Any highlights or memorable interactions?
I liked when two of my esteemed judges, Nate Silver and Kirk Goldsberry, both admitted to having just turned 40 and not really knowing much about esports. I realized later that I had only learned about esports when I turned 40 too! A critical age, apparently. Whoever makes the first video game specifically targeting midlife crises will be a billionaire.
You’ve been a frequent participant at SSAC, what keeps bringing you back?
The people! Jessica and Daryl don’t get enough credit for both the quality and the quantity of the community they have created. How can there be so many smart, thoughtful, AND kind people all in one place? There’s got to be some kind of serendipitous selection bias. It inspired my friend Eugene Shen and I to start the Journal of Sports Analytics, where Jessica and Daryl were gracious enough to be our first advisory board members, and in the five years we’ve been running that publication, we’ve noticed a flood of international submissions too, so the community they fostered is even larger and more global than it appears.
I hope I’m not giving away a trade secret here, but Mike Zarren organizes two very special events around the conference primarily for basketball analytics folks: a pre-conference sports bar night and a post-conference basketball game. You can find info on the apbrmetrics forums. Also, listening to Mike speak on any topic at the conference is a treat. For literally any question, his answer is always original, compelling, and surprisingly brief, and immediately forces the moderator to move on to a new topic, because that one is done, totally addressed. It’s a miraculous sight. Before he speaks, you think, “Oh, what an interesting and difficult question.” After he speaks, you think, “Oh, I guess that settles it.” He’s a genius.
The bottom line is this. You stand a little taller at SSAC than you do the rest of the year. And it’s not just because of all the seven footers. Something about the conference is magical and makes you proud to be a part of sports analytics.
What have you been working on since SSAC18?
The soon-to-be-very-legal gambling market in America will have enormous new opportunities, and I’m involved with a few different efforts to create a better future for people in that space. Brett McDonald and I have a patent pending on a new product that will revolutionize sports gambling.
It’s now also the twenty year anniversary of my time at Long-Term Capital Management, so I’ve been writing about some of the ripples from that event that are still with us to this day. I’ve also been doing a lot of corporate education and training in both behavioral decision making and data analytics. One of the biggest problems facing Fortune 500 firms, in my experience, is the communication gap between the business leaders and the data scientists, which has been a theme in sports analytics since probably SSAC1 and SSAC2; the rest of the world is still catching up.
Any advice for someone considering submitting their paper to the competition this year?
1. Pick a sport that you love and that you know intimately. While working on your submission, there will come a time when you’ll find yourself drowning in data and neck deep in algorithms, and you’ll have to make some simplifying assumptions, or reasonable unit tests, or sanity checks, and if you don’t already have a good intuition about the game, you’ll be in hot water, which can lead to mixed metaphors and tears.
2. Build interactive tools as early in your process as possible so you can explore and visualize the data and your assumptions and intuitions.
3. Don’t overcomplicate. Most studies in economics, health, medicine, and psychology do not replicate. Complication usually leads to that kind of trouble. Keep things as simple and as true as possible.
4. Aim higher than high: make a paper so outstanding that if you hadn’t written it, you’d be kicking yourself out of envy, and if it doesn’t get accepted, you’ll feel sorry for the conference on its loss. Get results so clear and broad and important that any team owner can immediately see the value. Don’t just go for low hanging fruit. Harvest the whole damn field.