Perspective from the Sloan Conference
Written by Justin Zovas (follow him on twitter @JustinZovas)
As I watch a college basketball game with friends and casually calculate Louisville’s expected points on a particular possession, I won’t deny it, I’m the weird one. But for two days this past weekend at the MIT Sloan Sports Analytics Conference, I was incredibly normal. For us sports geeks who worship the intersection of quantitative analysis and athletic competition, this was our Super Bowl.
Over 2,700 like-minded enthusiasts congregated in the Boston Convention Center to celebrate the strengthening marriage between sports and analytics. Much to my pleasure, attendees from professionals to students were very approachable and eager to discuss new ideas and theories.
One seemingly universal theme throughout the conference that warrants further discussion is the challenge to differentiate processes and outcomes.
When the former NFL coach turned ESPN analyst was asked whether the Raven’s aggressive move to perform a fake field goal in the Super Bowl was a smart decision, Herm Edwards responded by asking, “Well, did they win the game?” Humans, even the most elite decision makers in an industry, have a persistent fixation on outcomes. In reality, the process in coming to a conclusion or decision is far more important than the outcome. The merit of a particular decision should be determined by evaluating the process, not the outcome.
For example, let’s take gambling against the spread in any sport. When we pick a side we theoretically have a 50% chance of being correct. Statistically speaking, the outcomes of 50/50 propositions have the greatest variability. That is, the closer an event is to 50/50, the more random the results are. We can perform a spot-on analysis of a match-up and there is still a good probability that the game plays out different than expected. It may seem obvious but it’s still important to remember, if you are a successful handicapper and can win 54% of the time, you will still be wrong the other 46% of those instances. The point is not to get too elated or discouraged during runs; stay disciplined and understand that streakiness and randomness in sports is far more common than people think. And most importantly, we must not let individual outcomes cloud our assessment of decision making.
This is just one theme in a conference that boasted a wealth of innovative ideas and stimulating discussion. At the end of the day, quantitative analysis and objective reasoning are two extremely powerful tools in evaluating and predicting sports. I feel fortunate to have been even just a small voice in this continually growing conversation and cannot wait to attend the conference again next year.
