Mile High Sticking
Fansided

Why Advanced Stats May Hurt the Game

By Janik Beichler

facebooktwitterreddit

Advanced stats are the newest revolution in the NHL. It all started slowly, with people doubting that they could really help. But by now, there is a huge number of statistical analysts, every hockey writer uses them, and even NHL teams are starting to hire staff called “Hockey Analyst“. But do advanced stats really help the game — or do they even hurt it?

So, what are NHL teams hoping to get from those analysts? Here’s an excerpt from the Carolina Hurricanes’ news release on the hiring of Eric Tulsky.

"As Hockey Analyst, Tulsky will provide and analyze data to assist the hockey operations department and coaching staff. Tulsky worked for the team during the 2014-15 season on a part-time basis.“Eric is an extremely intelligent person and has been an innovator in the field of advanced analytics,” said (Hurricanes GM Ron) Francis. “We found the information he provided last season to be extremely valuable and are glad that he will be with us on a full-time basis.”"

It sounds great, doesn’t it? Instead of simply using your hockey knowledge and experience to evaluate players, you now have the help of mathematicians. You can track every step and every movement a player makes on the ice.

Let’s just use this fictional example. Goalie A has a .90 save percentage, while Goalie B’s is .92. Scouts watch both of them, and Goalie A looks better, but seems to make too many mistakes — or whatever the reason may be — and lets in too many shots. Now, the good scouts, managers and coaches, will figure out the following, without looking at advanced stats.

More from Mile High Sticking

Goalie B has played behind a defense with very good positional play over the past three years. Therefore, the team didn’t allow their opponents to get too many scoring chances from the slot or other high-scoring areas. Most of the shots against goalie B came from rather difficult angles.

Goalie A on the other hand, got lots of shots against from perfect shooting angles. Right from the slot, the center blue line, and directly around the net. His defense didn’t do a great job blocking shots, or closing down shooting lanes. Goalie A may be the better goalie, but he had much harder saves to make than Goalie B.

That can be proven by using the adjusted save percentage. If both goalies had gotten the exact same percentage of shots from each area of the ice, how would they compare? Not surprisingly, Goalie A would have had a better save percentage than Goalie B, even though his standard save percentage is two percent worse than Goalie B’s.

The good scouts, coaches and other personnel would have seen that without the use of advanced stats. Now, everybody sees it. You, I and every hockey fan out there.

It is a rather extreme example, and likely wouldn’t happen that way. But it shows the point I am getting at. Advanced stats are a huge deal for player evaluations, and they show a lot of things people may have missed otherwise.

The Colorado Avalanche currently has a little bit of a goalie controversy. Not the Canucks kind, but a little one about the backup position. I, myself, used advanced stats — e.g. the above-mentioned adjusted save percentage — to show that Calvin Pickard may be a better goaltender than Reto Berra. In this case, it is something we knew all along, but it gives us assurance.

More from Avalanche News

The point is, advanced stats help us — the fans and writers — and they help the teams as well. It seems like new statistics come out every day, presenting us more and more ways to evaluate players. Isn’t that a good thing?

Well, some coaches are better than others. Some scouts are better than others. Some managers are better than others. It is the same in every part of life — everyone has their talents, and even there, some are better than others.

Advanced stats seem to bring that closer together. They may tell coaches things that other coaches knew all along. Some coaches may need them more, others need them less.

As a comparison, let’s look at another sport: Formula 1 racing. Remember that time when the best drivers were the ones that would win the races? That time is long gone. Today, it is all about who has the brightest brains on their development teams, about who has the best mechanics, and who makes the best car. Put a mediocre driver in the best car, and he’ll win the championship.

Back to hockey. There are coaches we think are great at what they do — Mike Babcock for example — and others that don’t seem to be that good — say, Dallas Eakins? But who knows, if Dallas Eakins was paired with the best hockey analyst there is, maybe he would look like a much better coach.

The more popular advanced stats get, and the more NHL teams rely on them, the more they will change the game. Advanced stats are a cool thing, especially for us hockey writers. But maybe hockey would be more exciting if you still had to trust your eyes 100 percent, without any other ways of evaluation.

What do you think about advanced stats? Do you think the NHL would be more interesting without them? Let us know in the comments. 

Next: A Post-Expansion Realignment Proposal

More from Mile High Sticking

Home/Avalanche News