:format(webp)/cdn.vox-cdn.com/uploads/chorus_image/image/35782420/452666789.0.jpg)
Much of this work has been dealing with passes that generate shot attempts or shots. My belief, and what much of the evidence has supported, is that there is something of value in measuring the efficiency in generating shots and not merely attempts. In this article, I want to take it a step further and looking at the efficiency of those shots going in the net. Today, we’re going to look at which passes generate goals.
Summer Passing Series Links
Part One: Efficiency and Winning
Part Two: The Transition Game
Part Three: Offensive Zone Analysis
What Constitutes a Goal being Generated?
I wanted to look at passes that led to a goal being scored, a quality primary assist. In reviewing each goal the Devils and their opponents scored this season, I decided I only wanted to award a Goal Generated (GG) if it was an intentional pass. So, I did not record a GG if a player took a shot and a teammate scored off of a rebound. In the same manner I tracked passes, shot attempts, and shots, I felt it necessary to apply the same strict guidelines with goals.
I broke these down by zone and will be able to look more at this with another season of data, but I thought I’d introduce it here and see what your thoughts were. Certainly there’s going to be a tremendous amount of skill placed on the actual shooter to score the goal, but I figured I’d post what I have now and see what trends develop throughout the coming season. So, yes, I’m aware these may fluctuate.
In truth, this should be nothing new to this audience. John looked at something similar a few seasons ago. His work on Patrik Elias and Ilya Kovalchuk can be seen here and here. I read through these a few times before tracking this new piece and I encourage you all to do the same to get an idea of the thought process.
The Results
Here’s how to read the chart: the x-axis represents the amount of goals generated by the player, the y-axis represents the shooting percentage each player generates (SH%G), and the size of the bubble represents the volume of shots generated by the player (the number is next to their name as well).
Starting with the defensemen, we once again see the disparity in volume between Marek Zidlicky and the rest of the defensemen. Zidlicky generated ninety-two shots and nine goals, for a shooting percentage of about 10%. We’ll see what another season brings, but a SH%G of 10% is a bit above where the team shot at overall.
Andy Greene had the next highest number of shots generated at fifty-four. Greene only generated three goals for a SH%G of 5.6%. This tied Greene with Bryce Salvador for lowest SH%G on the blue line.
Jon Merrill, Mark Fayne, and Adam Larsson each generated a shooting percentage over 10% on the four, three, and two goals they generated respectively. Fayne’s goals seem more fluky due to his overall lack of efficiency in generating shots compared to the rest of the blue line.
Moving to the forwards, we see Jaromir Jagr and Travis Zajac once again leading the pack as each generated eleven goals. Zajac generated twenty-six fewer shots to get there, so his SH%G is slightly higher than Jagr’s; Zajac’s SH%G was right at 10% because of this, nearly 2% higher than Jagr’s.
Patrik Elias was next in terms of volume, one ahead of Adam Henrique in both shots and goals generated. Most of the forwards had a SH%G in the 6 – 8% range, which make sense given how the team shot for the season.
Reid Boucher was the outlier with four goals generated on fifteen shots for a SH%G of 26.7%. Obviously that’s not sustainable, but in Boucher’s sample size, he’s looked quite good by nearly all of these metrics. We’ll see what the coming season brings.
Dainius Zubrus has the honor of being the forward to generate the lowest shooting percentage at 3.4%. Whether that is strictly luck or the shots Zubrus generates are in less dangerous areas remains to be seen.
What Does This Mean?
If we can identify the true Corsi machines on each team (those that generate shot attempts) and refine that to measure efficiency (those that generate shots), then it seems logical to eventually be able to identify generating goals as involving some level of skill from the passer. I don’t know if what I’ve laid out here is the blueprint for determining that, but everything I do evolves as I continue to track stats. I anticipate this will be something that can be debated more intelligently as I accumulate more data.
So, What’d You Think?
This was the fourth article in a summer series looking at all of this data. What did you think of this data? What methods could be employed to improve the process? Are there different ways of looking at? What is the relationship between passer and shooter and how much skill should we attribute to the player not shooting the puck? Give me your questions, statements, feedback so I can better steer this towards where your interest lies. Sound off below!