NHL: Why are season predictions so hard to make?

NEW YORK, NY - NOVEMBER 03: Ryan Pulock #6 of the New York Islanders and Thomas Greiss #1 of the New York Islanders celebrate after game against the New Jersey Devils at Barclays Center on November 3, 2018 in the Brooklyn bourough of New York City. The New York Islanders defeated the New Jersey Devils 3-0. (Photo by Andy Marlin/NHLI via Getty Images)
NEW YORK, NY - NOVEMBER 03: Ryan Pulock #6 of the New York Islanders and Thomas Greiss #1 of the New York Islanders celebrate after game against the New Jersey Devils at Barclays Center on November 3, 2018 in the Brooklyn bourough of New York City. The New York Islanders defeated the New Jersey Devils 3-0. (Photo by Andy Marlin/NHLI via Getty Images)

As the 2018-19 NHL season nears the 15 game mark, several teams have surprised fans early on. Why does this happen?

The NHL is the hardest major league to make predictions for, before the season begins. While some may credit this to puck bounces and deflections, a lot of it has to do with goalies.

Among the four major leagues within North America, the NFL, MLB, NBA and NHL, there is no position like the goaltender. The MLB and NFL come the closest with quarterbacks and pitchers, which is why basketball is so easy to predict. There isn’t any point where one player is going to start having less of an effect.

More from Puck Prose

1In hockey, goaltending means everything. It’s very rare for a team to spontaneously become good as a whole. They can improve from year to year, but when you look at teams who have crazy breakout seasons, a lot if it can be traced back to the goalie.

Why does this make it hard to predict team success? Well, team success is tied to goalie success and goalie success is very difficult to predict. Stats like corsi for % and PDO make it easier to track a skater throughout their career. However, while new goalie advanced stats are still coming out, there is really is no equivalent.

This season, there have already been a few examples of goaltenders making or breaking a team. As one of the most obvious examples, let’s first take a look at the New York Islanders.

New York Islanders

Wow, the New York Islanders are doing surprisingly okay this season. This is quite impressive considering the departure of John Tavares, isn’t it? How are the Islanders managing it?

If you’re an Isles fan, you’ll know the answer to this. They were good scorers during 2017-18, with 261 goals over the entire season, the NHL average being 244. The Islanders’ problem was in the other column.

They allowed 293 goals, which effectively more than canceled out anything that their scoring ability did for them. So then, was the problem the defense or the goalies? Well, you can pretend it’s both, but it was the goalies.

In a league where the average save percentage is .912, a .903 isn’t going to work. You need a lot more than 261 goals to compensate for that. Even when they had John Tavares, the Islanders couldn’t.

But this season, things are different. Thomas Greiss has been spectacular with a .934 save percentage, and Robin Lehner has been almost as good at .928. But, how is a snooty blogger supposed to predict this?

Greiss was awful last season. He didn’t look anything like a legitimate backup. Now he looks completely different, apparently. Lehner, to be fair, has had some great seasons, but not last year.

These goalie swings are near impossible to see coming, yet goalies somehow have to be factored into the mystical equation of playoff predictions. The Islanders are just one example, however.

Florida Panthers

Next on the list are the Florida Panthers. Last season, they were led to the verge of playoff contention by James Reimer and Roberto Luongo, who, for some reason, is not considered as one of the top goalies of all time.

Luongo has been injured for this first bit of the season, leading to a Michael Hutchinson and Reimer duo. They have not excelled. Hutchinson is at a .839, while Reimer is at .885. Look how much this has hurt the Panthers.

Take a .919 from last season and compare it to the .876 of this one. You might as well have a completely different team. Luongo will probably help things, but this is rough start the Panthers are out to.

In addition, it also hurts the idea, that this might be the year for the Panthers. It’s still possible, only 11 games into the season, but it hurts them. To expand on the point, it also makes everything a lot less predictable.

Conclusion

Hockey is hard to predict because of all the goalie variance. A goalie way more of an effect on a game than any other player. But that’s the important part: the goaltender is just one player.

All of this variance, season-to-season, makes it hard to predict who is really going to make the playoffs. When someone who controls the game to the extent that a goalie does, how are us poor bloggers ever supposed to know anything?

Really though, if you’ve ever wondered why hockey is so much harder to predict, it’s not just because of the luck factor. It’s because goalies are individual players, and individuals can fluctuate.