Sunday, 6 December 2015

Martin Marincin, Jake Gardiner, and the sources of Corsi

Jake Gardiner has been somewhat of an enigma during his time with the Maple Leafs. Many stress that his decision making skills are lacking and that his defensive abilities are abysmal.

However, the advanced stats community has taken a loving to Jake Gardiner's defensive ability. This is always a hot button issue within the arguments aginst Corsi. How can a puck moving defenseman like Gardiner be good defensively? He keeps turning the puck over. He's soft. He's not a defensive defenseman at all!

That's kind of the point.

What advanced stats gurus define as "good defensively" is far more broad, and yet, far more simple than what the average fan might. To be good defensively by Corsi, all you have to do is have a low Corsi (shot attempts) Against. That's simple enough, right?

To be good defensively by eye-test definitions, you have to be physical and be in good position. Also pretty simple.

Where advanced stats eclipse the eye-test is that it itself is quantifiable, and that its value to winning is also quantifiable. You can read this article by Steve Burtch to find out why. But other than counting shot attempts, what makes Corsi?

That's where advanced stats gets far more broad than the eye-test. There are a hundred different skills a player can excel or fail at that can contribute to their Corsi. That's what I'd like to discuss today.

Jake Gardiner

So, we know Jake Gardiner is a good puck moving defenseman. How does he have good defensive numbers?

Having a low Corsi against can come from several skills that Gardiner excels at. Making slick breakout passes. Carrying the puck safely out of the corners. But what else? He doesn't really excel at anything else in the defensive zone. So why are his defensive numbers so good? This horizontal evaluative ranking optic (HERO, for short) chart shows just how good he has been over the last 3 years: 

The best defence is a good offence. 

Controlling the puck in the offensive zone is a huge contributor to Corsi Against. If you're in their zone they can't be in yours trying to score on you. That's as simple as it gets. Corsi Against can show this.

We can find other players among the Leafs who, while aren't necessarily defensive stalwarts, have impressively low Corsi Against numbers, and are helping the Leafs be better defensively as a result. Richard Panik is one. Mark Arcobello another. Matt Hunwick (before his movement to the Leafs he was regarded as an awful defensive defeseman, while posting remarkable Corsi Against numbers), yet another.

Martin Marincin

Where does Marincin come into this discussion? We talked about Jake Gardiner's puck skills being his source of success in Corsi Against. Martin Marincin is quite the opposite. He leads the defense in CA60 (Corsi Against per 60 minutes of ice time) and is second behind the aforementioned Jake Gardiner in dCAImpact (the running total of shot attempts a player has theoretically prevented). Compare his HERO chart to Gardiner's above:

You can see Marincin is a bit behind Gardiner in the Possession Impact section, but still performing at almost a top pairing rate. That's really good.

But if Marincin is really good at Corsi, too, then you might think he excels at the skills I described above for Gardiner. That would make sense if you had never watched Martin Marincin play hockey.

I said even further above there are a hundred different skills that can contribute to good possession metrics. So, what are Marincin's skills?

It's kind of a hard question to answer, isn't it? There's nothing really obvious about Marincin's game that we can say definitely contributes to his possession success.

My theory is this: he's really big, and moves really well. Combining his frame to prevent being out muscled with the ability to be directly in the way of what the offense is trying to accomplish has to be a part of what his successes are coming from.

But this is where advanced stats jumps out of the hands of amateurs like me, and into the hands of video coaches, analytics departments, and pro scouts.

What are your theories on why Marincin is good? Or maybe he's terrible and Corsi is hiding something awful about his game? Share your thoughts in the comments!

No comments:

Post a Comment