Why it’s so much harder to predict winners in hockey than basketball

So far this year in the NBA playoffs, only two out of 14 series were won by the team with the worse record during the regular season. Compare that to the Stanley Cup playoffs, where the underdog has won fully half of the series.

The Nashville Predators, currently battling the Pittsburgh Penguins in the final, had the 18th best win-loss record during the season, and the Washington Capitals, who won more games than any other team this year, were knocked out two rounds ago. Why? (Or as Caps fans put it: “WHYYYYYYYY.”)

One answer is that the design of hockey introduces more randomness into the results. Investment strategist and sports fan Michael Mauboussin demonstrates this in his book The Success Equation, with a statistical technique that ranks sports by the relative contributions of skill and luck to the regular season standings. He looked at five seasons for five sports leagues and ended up with a continuum that looks like this:

Michael Mauboussin’s skill-luck continuum.

That’s not to say hockey players are any less skilled, but that outcomes in hockey are less representative of true differences in skill between teams, compared to basketball — in large part because the NHL is more competitive. There are many reasons for that, from the design of the leagues to the fundamentals of the gameplay. Check out the video above to learn about some of the factors involved, and head over to our YouTube channel to leave your own theories in the comments.

Back to top ↑