Want to win in the NHL? Score first—but not until the third

Apr 30, 2014 by Robb Frederick
Samantha Key, a senior from Freeport, analyzed the scoring patterns in 12,000 NHL games. She found that the team that scores first wins 66.7 percent of the time. Credit: Penn State Behrend

(Phys.org) —The team that scores first has an advantage in a National Hockey League game. That should be obvious: They are, at that point, winning.

The extent of that advantage, however, depends on when the goal is scored. If it comes at any point in the first two periods of a game, the probability that the team first will win is 66.7 percent, said Samantha Key, a mathematics major at Penn State Erie, The Behrend College.

"You don't have to rush it," said Key, a senior from Freeport. "Time isn't really a factor until the third period."

If the first goal comes in the third and final period, the likelihood of the team winning is much better: 79.4 percent, Key said. She determined that by analyzing the scoring patterns in 12,000 NHL games – nearly every game played by every team in the league over the last 10 years.

Key expected a standard straight line graph, with the first-scoring team slightly more likely to win as every minute passed. When she didn't get that – when the advantage in the first two periods was essentially level, regardless of the time left on the clock – she plotted the data again.

What she saw was two trends: a relatively flat section in the first 40 minutes, followed, late in the game, by a sharp increase in the scoring team's chances of winning. Statisticians call that – no kidding – a "hockey stick regression."

There are reasons for the spike in the final 20 minutes. Teams that score late are likely playing a defensive game. At the same time, the players are tiring. That can lead to mistakes, which can change the outcome of a close game.

The flat portion of the graph is still something of a puzzle. The team scoring first should, in theory, have a greater advantage as the game progresses.

Coaches know better, however. "They always say, 'There's plenty of time to get it back,'" said Michael Rutter, associate professor of statistics at Penn State Behrend. He supervised Key's research. "And the statistics support that. You don't have to get your own goal right away. If you play well, and you play within the system, there's a good chance you'll still be okay."

For Key, a fan of the Pittsburgh Penguins, who ousted the Columbus Blue Jackets in the first round of the Stanley Cup playoffs, the first goal nonetheless is a source of comfort, whenever it comes.

"Once they get that first , I'm okay," she said. "Then I can enjoy the ."

Explore further: Game-winning 'momentum' illusion is but a delusion

add to favorites email to friend print save as pdf

Related Stories

Game-winning 'momentum' illusion is but a delusion

Feb 13, 2014

(Phys.org) —A hot hand may be hokum: Cornell researchers have examined the concept of "winning momentum" with varsity college hockey teams, and they conclude that momentum advantages don't exist, says a ...

Football displays fractal dynamics

Mar 19, 2014

Football fascinates millions of fans, almost all of them unaware that the game is subject to the laws of physics. Despite their seemingly arbitrary decisions, players obey certain rules, as they constantly ...

Recommended for you

Precarious work schedules common among younger workers

Aug 29, 2014

One wish many workers may have this Labor Day is for more control and predictability of their work schedules. A new report finds that unpredictability is widespread in many workers' schedules—one reason ...

User comments : 0