Which Games Matter Most in a Best-of-7 Series?

Not all playoff games are created equal. Here’s which matchups truly swing a series, based on the numbers.

A videoboard outside Madison Square Garden before Game 7 of the 1994 Stanley Cup Final. (Source: YouTube)

As the NBA and NHL playoffs roll on, the tension in a new batch of series is set to ratchet up with each passing contest. But not every game in a series is created equal. Some nights feel like they carry the weight of the entire season; others, in retrospect, were just preludes to what ended up being the real turning point.

So which games in a best-of-7 series truly decide who advances? (Or, in the Finals, who wins the championship?) It’s a question that we all have an intuitive feel for as we take in the postseason action. But some simple math can also more precisely measure the pressure of each game — in terms of the potential swing in a team’s odds of winning the series, depending on whether they win or lose that night — to tell us which games moves the needle the most.

This will be an informative reference post as much as anything, building off of work I’ve done in the past. But I’ve long thought it was also an interesting subject with some counter-intuitive lessons and cool applications as well. The basic idea is that each situation in a best-of-seven series — there are 16 possible configurations — carries an intrinsic win probability for each team, depending on what the current margin of the series is and who has home advantage (and also how that varies by sport). This also tells us how much the pregame odds will change based on which team wins, the average absolute value of which¹ is equivalent to the “leverage index” for the game — its potential to drastically change the state of the series in one direction or another.