Statistics in sports can be used in two general ways: to describe what has happened and to predict what will happen. While the former is certainly simpler, the latter is far more interesting, as successful prediction can guide strategic decisions by coaches; inform a fan how likely their favorite team is to bring home the championship; or provide a tidy sum to bettors. In this column, I use the Elo model to make predictions for the Yale men’s and women’s hockey teams as they look to conclude their seasons.

The Yale Undergraduate Sports Analytics Group’s prediction model utilizes the Elo ranking system, which is most notable for its use in chess. Elo has been applied to a variety of sports by FiveThirtyEight and other analytics platforms; when a team wins a game, its rating increases and the rating of its opponent decreases, taking into account the initial ratings of the teams. If a very good team defeats a very poor team, the result will not likely do much to change the system’s opinion of the relative strength of the teams and thus the ratings change will be small. In an upset, however, the underdog’s rating will go up by a larger amount.

Our model also takes into account two factors that the traditional Elo system does not. First, accounts for the home-field advantage in each sport. Second, it considers the score differential, or magnitude of victory, of each game analyzed. Research has consistently shown, in just about every sport, that considering a team’s point differential is more predictive than just considering wins and losses.

As with any model, Elo has its limitations. The most notable is that it only takes game scores as inputs. It will not take into account, for example, a recent injury to a star player. That said, a redeeming quality is that Elo can quickly adjust; if a team underperforms for a few games, its rating will decrease to reflect that performance.

Yale men’s hockey

As of this writing, the Yale men’s hockey team sits sixth in the ECAC standings. Our model is somewhat more bullish on the Bulldogs, ranking them the fifth-best team in the conference. To make our predictions, we simulated the remainder of the ECAC season 10,000 times. Union, featuring the highest Elo rating in the conference, emerged as the favorite, with greater than a one-in-four chance of winning the ECAC tournament. Due in part to being in strong positions for first-round byes, either No. 4 Union, No. 14 St. Lawrence, No. 5 Harvard or No. 16 Cornell win the title in almost 90 percent of our simulations. While the Elis have only around a 5 percent chance of winning the ECAC tournament, their strong record in Ivy League play gives them a decent chance of emerging with at least a share of the Ancient Eight crown.

Our model has the Bulldogs as the favorites in their upcoming matchups this weekend at home against No. 19 Quinnipiac and Ivy-rival Princeton. Yale has a 66.9 percent chance of defeating Princeton and a 59.1 percent chance of defeating Quinnipiac, with ties weighted as half a win.

Yale women’s hockey

By sweeping Brown this weekend, the Bulldogs solidified their place in the ECAC playoff race. Yale sits seventh in the standings — also their ranking according to our Elo model. When we simulated the remainder of the conference season 10,000 times, No. 5 St. Lawrence, the top team according to our model, and No. 3 Clarkson, the current leader in the standings, emerged as the favorites. With a decent amount of separation between the eighth and ninth place teams, the playoff field appears relatively set, with the current top eight reaching the playoffs over 80 percent of the time. While a near-lock for the playoffs, Yale has just over a 1 percent chance of winning the tournament by our model’s prediction.

As we see it, the Elis are underdogs in both of this weekend’s road matchups, with just over a 31 percent chance against both the Tigers and the Bobcats.

Michael Bogaty is a member of the Yale Undergraduate Sports Analytics Group. Contact him at .

We are undergraduates passionate about the field of sports analytics. Members discuss the current state of advanced statistics in a variety of sports and attempt to contribute to the field through original research. Check out more great content at our website,!