Tuesday 14 October 2014

Scratching Jake Gardiner got the Leafs a win; but was it the right call?

The simple answer is no.

The more convoluted answer is: yes, but it shouldn't happen again.

The Leafs, on Sunday night, had to decide which of their seven defensemen would have to take the press box for the game against the Rangers. According to Randy Carlyle, "Jake drew the short straw tonight," which is to say that it was a tough decision that came down to luck rather than actual skill.

The fact that Randy Carlyle couldn't realize that he could have made the decision based on skill as opposed to drawing straws (figuratively) is baffling.

Going into Sunday's game, this is how the advanced stats looked for the Leafs' defense.




To understand this graph, I'll break each element down individually, then describe what it all means together.

First, the Y-axis is labeled as Fenwick Competition %. What this represents is how strong of competition each player faced. The data range goes from 52% to 59%, which means that a player on the 52% mark faced, on average, opponents with a Fenwick % of 52%. Basically, a player higher on the Y-axis played better players, and players lower on the Y-axis played worse players.

Secondly, the X-axis is labeled as Off ZS%. What this represents is how many of your shifts begun with offensive zone faceoffs, relative to your teammates. High Off ZS% numbers indicate that the coach is giving you more time in the offensive zone to protect you from playing defensive zone time and/or to allow you to control the offensive zone.

Thirdly, the colour of a bubble is labeled as Fenwick Rel %. This is a player's on-ice Fenwick % relative to their teammates. The more blue the dot is, the better a player was at controlling the play, and the more red the dot is, the worse they were. 

Lastly, the size of a bubble is labeled as TOI/Game. This is a simple enough concept, which is there to indicate which players are playing the brunt of the minutes. As you can see, the bubbles are all roughly the same size, which means the minutes distribution was pretty even. This is a good thing, and something that was not done at all the last few seasons by Carlyle. +1 for him there.

There's a couple easy trends you can notice from this graph. The first one is you can see what the defense pairings were quite easily, by looking at which two defenders are near the exact same level of Fenwick Competition %. They were, as we know:

Phaneuf - Robidas
Gardiner - Rielly
Percy - Polak


The second trend you see is with the colours of the bubbles, and that's that Jake Gardiner and Morgan Rielly were much better than their teammates at controlling possession.

The third thing you can notice is that Gardiner and Rielly were pretty significantly sheltered, both in terms of competition and zone starts.

Lastly, the thing you can take from this graph is that Robidas and Polak were pretty significantly buried, Robidas in both competition and zone starts, and Polak with defensive zone starts. They both performed very poorly in this situation, a FF% of only 26% for Robidas (that is all kinds of terrible) and 35.85% for Polak. Those are simply not good enough, and can't be totally justified by the tough minutes they played.

If anyone should have been taken out of the line-up it should have been one of Robidas or Polak, because of their awful performance in those first two games, plus the fact that if anyone could use the rest it's the 37-year-old, two-broken-leg-recovering, Stephane Robidas, and not the young, healthy and more talented Jake Gardiner.

To make this long story short, in hindsight scratching Jake Gardiner may have been the right call for that game, because it got the win and you can't really argue with those short term results. However, if this decision becomes repeated in the long term, it will very seriously affect the Maple Leafs' ability to control the play.

No comments:

Post a Comment