The problem with advanced hockey statistics

More and more, you start to see more advanced statistics used in hockey, especially on the Internet.

Douglas Murray is the newest Hab, but that might not be a good thing, say some people using advanced statistics.

Most of them are confusing, and the average fan doesn’t need to use or care about them. However, it’s getting to the point when you can’t visit any site on the web without people talking about advanced statistics.

I think they’re massively overblown. Some sites use them as the be all and end all for proving why such and such a player is great or horrible. I’ve seen a lot of it lately with sites trying to figure out if the Montreal Canadiens made a smart move in signing Douglas Murray.

But many of these advanced stats don’t look at the big picture enough. The problem is, taken in a vacuum, all of these stats are useless.

Let’s use two players as example to see who is better. For now, we’ll call them player A and player B.

I’m using the 2012-13 season, all players who played at least 30 games, and these are all in 5 on 5 situations. For the record, 509 players meet these criteria.

Below is a chart that highlights where they stand in regards to some of these advanced stats:

Player A Player B
Quality of Competition 0.004 (211th) 0.043 (103rd)
Relative Corsi Quality of Competition 0.221 (285th) 0.665 (154th)
Relative Corsi Quality of Teammates 5.812 (14th) 6.116 (11th)
Relative Corsi 20.3 (12th) 20.6 (9th)
On-ice Corsi 16.10 (20th) 25.4 (5th)
On-ice team Goals Against per 60 minutes 2.25 (275th) 1.48 (480th)
On-ice team save percentage .924 (190th) .943 (46th)

Looking at that list, most would think that Player B must be better. Plays against tougher competition and is better defensively. And Corsi is better in all sorts of examples.

Maybe Fenwick, which measures puck possession statistics is better. These stats use players who had 500 minutes of 5 on 5 ice time last season (355 eligible players).

Player A Player B
Goals scored against team while player is on the ice per 20 minutes   of ice time 13.08 (153rd) 11.981 (70th)
Hockey Analysis Rating Offence 31.2 (2nd) 30.8 (3rd)
Hockey Analysis Rating Defence -5.7 (202nd) 9.6 (95th)
Hockey Analysis Rating Total 12.8 (55th) 20.2 (14th)

A lot of these stats can be broken down even more, looking at the last two-year period, or three year-period. But suffice to say, you get the point.

By the way, Player B is Tyler Seguin. Player A is Sidney Crosby. Yet these advanced stats say that Seguin is better.

That’s where these stats all fall apart. There’s so many of them now, you can cherry pick what you want to prove your point.

Habs Eyes on the Prize, an excellent Habs blog, has been pretty hard on Murray. They present some of these stats, and say it reflects on a bad signing.

I disagree, and this point was what I took away from two different articles about Murray:

I was going to use goal statistics here to add balance, but considering the goaltending Murray received last season, we know how that affects the numbers and I don’t think it’s necessary. What we can look at though, is strength of opponents.

See, something that showed Murray in a more positive light isn’t used, just mentioned in passing. Instead, the stats used lean toward the signing being a negative one.

Habs Eyes on the Prize also wrote on about whether Murray is a good penalty killer, using all these fancy stats. Here’s all I need to know about Murray:

  • Only 60 defencemen last year had more than 100 shorthanded minutes. Murray was one of them.
  • He was 20th among all defencemen in blocked shots.
  • He was tied for 30th among all defencemen in hits.
  • Only eight defencemen had more hits and blocked shots combined.
  • He was in three fights last year, and won all of them.
  • Despite being slower, he doesn’t take a lot of penalties. Last year, he had just 35 pims. Take out the fights, that equates to only 10 minor penalties (about one every four games).
  • While killing penalties, he starts in the offensive zone four per cent of the time, but finishes there 45.1 per cent.

Here’s the thing: Murray is making only $600,000 over the league minimum. And he’s coming in as a sixth/seventh defenceman. He’s not going to be coming in and playing 20 minutes a game.

Now, don’t get me wrong. There are some important non-traditional stats. Top teammates, PDO, offensive zone starts, etc.

But advanced stats can get out of hand. You can look at them all you want and use them to prove whatever you want. But in the end, they don’t really prove anything.

And that’s the most important thing to remember about advanced stats.


Leave a comment

Filed under Sports blogs I like

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s