Friday, August 05, 2011

Habs Scoring Chance Breakdown

We now all know what a scoring chance is. We now all understand its potential and are ready to play with stats.

Scoring chances should help us look into the games of Michael Cammalleri, Scott Gomez, Andrei Kostitsyn and others in a new way, to get beyond the very helpful but very limited Corsi and to add some new paint to the picture.


Scoring chance analyses

One of the great things about the way the scoring chance trackers do their work is that they build in a timestamp.

Because of this very crucial early decision (and the commitment from trackers to publishing when every scoring chance happened) we have the opportunity to cross reference the whole thing with the NHL time-stamped version of events. From there, we can see not only which shots were scoring chances, but also which scoring chances missed the net or were scored as non-shots by the league.

The composite of the two records is quite exciting because it offers the chance to run calculations that describe the shooting, positioning and offensive decision-making of the players.

I have a nearly full record of the whole 2010-11 season done in this way, but after the very informative Q&A with Olivier of En Attendant les Nordiques, I realised that I needed to go back to the drawing board on some of my analyses. So for now, I give you the analysis if scoring chances in the 2011 playoffs.


Starting from the beginning: the conventional playoff shooting stats.

As you can see below, some players took a lot of shots. Gionta led the team in total shots and shots per 60 minutes, Pouliot was completely AWOL. From these simple numbers, we can also run the simple Scott Cullen 'Josh Gorges' analysis and see that Spacek and Hamrlik could use some time on the practice green.



Moving on, we can look at Olivier's scored chances for the series by player. Gionta still led the team in scoring chances per ice time, but note the fallers. David Desharnais who came out looking like a star on NHL stats, suddenly doesn't look so good. Why? Because 75% of his shots were from outside Olivier's home base.

When people fall, others rise. Kostitsyn and Pyatt jump up because their shots tended to be from within the homebase (and by consensus definition more dangerous) -- though we're having trouble remembering a dangerous Pyat shot just now...



OK, so we knew all that. Or we could have known it if we looked. But the crossrefencing comes in now. Below you will see a column called "Target". Target means a scoring chance (from Olivier's record) that corresponded to a shot (on the NHL's record) -- a shot from within the homeplate area.

This is where it gets interesting, because we can start to see not only who waits to take shots from within the homeplate, but also who can make sure an attempt from there (where coverage is tightest) is on net. The skills of patience, timing, positioning and finally accuracy combined.

Here we see the cream (and Weber) rising to the top. Forwards are now on average better than Dmen and the forwards we thought of as threats at the time show up as the biggest threats in the record. Of greatest note here (apart from Weber who had some very nice shots) is Scott Gomez (half the number of on target attempts per game vs. the elite forwards). Pouliot really did his contract dreams no favours, did he?



So what we'd really like to know is how dangerous were the shots of each player on the team? That way before we go headlong down the Corsi results road we can see if perhaps we'd rather some of those Corsi-helping attempts had been held back in favour of a pass or a dump in.

That's what we can see below. Here are the on-target shots form the homeplate area as percentages of total shots and attempts. Weber again is top, and Andrei Kostitsyn creeps in with some scary accuracy. Scott Gomez, once again, is shown to be a very cheap shot.

If I'm defending this what I see is one player (Kostitsyn), who I'd better stick to like glue and another (Gomez) who I should probably actively open shooting lanes for from the outside boards.



Of course, we could counter by saying that Kostitsyn holds it too much and Gomez is just playing the odds. But is this more than three-fold difference really just a regression to mean waiting to happen? We all watched the games. I don't remember it that way.


Team metrics

Nowadays people who go on too long about individual stats like I just did are eaten by the wolves of Corsi. A player's value you see, can't be shown merely by looking at individual stats alone. i rather agree. And luckily for me, the script that scoring chance trackers use also shoots out all the players that were on the ice at the time the chance occurred.

So now we can see which players were on the ice for the most scoring chances by the team.



This is where things get interesting isn't it?

Scott Gomez, who clearly has no control over or confidence in his own shot is now squarely in the top 5 on the list -- ahead of his linemate Gionta! We aren't at all surprised by those ahead of him (stars of the playoff series that they were), but are a little dismayed that David Desharnais seemed to be a chance sink.


Goals scored

Finally, a look at who converts. So you've won yourself a shot in the slot, Tom Pyatt, what are you going to do with it?



Beyond anomalies Weber and Sopel (and I hesitate to say Subban), we can see that the guys who convert their chances as often as they should (more than 20%) are Cammalleri, Kostitsyn and Gionta. Sopel, who barely played spun some magic as well, because when he was on the ice, the team shot over 20%.


Players of note

So what does all this tell us about the players we wanted to know about?

Brian Gionta: The boy believes in the power of odds, shooting frequently. Fortunately for him, his shot is of above average accuracy

Tomas Plekanec: Mr. Steady. He's up there in most offensive categories. he shoots a lot, but also clearly knows how to make it count.

Mike Cammalleri: Potentiator. Sure his individual record is good, and he scores the goals, but he also is on the ice for the most chances by his teammates. Who said he couldn't pass?

Andrei Kostitsyn: Good positioning, good shot, maybe too patient/picky. Probably a good thing he wasn't let go.

Scott Gomez: Should wait until the teammate becomes free, there seems to be some redeeming bit of magic when he can do that. Anything is better than watching his shot.

Benoit Pouliot: Played himself off the team in perplexing fashion.

David Desharnais/Lars Eller: Our third line might be in trouble once checking gets tight.

PK Subban: What a playoffs.

No comments: