The idea behind advanced stats is to allow us to better understand the value of players' on-ice contributions. There are stats like Corsi and Fenwick that can indicate how a player impacts his team's puck possession. Others (Offensive Zone Start % and Quality of Competition) tell you how challenging a player's ice time is by breaking down where that player's shifts begin and what the level of the player's competition is that he is facing. There are also several metrics that try to assign a total value to describe a player's contributions to his team's success in all phases of the game. Of those the best one, in my opinion at least, is Goals Versus Threshold (GVT).
GVT basically measures how important a player is in terms of goals over a threshold or replacement player. It was created by Tom Awad of Puck Prospectus and is a complex calculation that takes aspects of offense, defense and shootout performance and puts it into terms of goals to determine how valuable that player is. I've exchanged a handful of emails with Mr. Awad over the years to better understand his thought processes to see just how much of the game he factors into his calculations and I was quite impressed. There isn't much he doesn't take into consideration.
We all know advanced stats are commonplace in baseball. And of course baseball and hockey are two very different sports. Baseball is easier to apply advanced metrics to because the game consists of a lot of one-on-one battles between pitcher and batter. Hockey is so much more of a team game. A player's performance can be impacted tremendously one way or the other by the contributions of his teammates. Trying to pinpoint exactly how much one player contributes to his team's success is no easy matter and while GVT is not perfect it is the best we have.
I thought it might be interesting to take a look at the final GVT ratings for the Columbus Blue Jackets players. Let's see how favorably GVT looked upon Columbus. But before we get to that, let's consider what we refer to as the 3-1-1 principal.
Basically, every three GVT is worth about one standings point and it would cost roughly $1M to buy the difference in production between that player and a threshold player. Consequently we can calculate how closely Columbus finished to their projected points total according to GVT and also how efficiently they spent their cap dollars.
Adding up all of the GVT contributions for the Columbus Blue Jackets (I took a fraction of the GVT accrued by players who played a portion of the year with other NHL teams based on games played and applied it to the total) I get 127.9 (I assume Puck Prospectus prorated their figures to reflect a full-season total). Dividing that by three I end up with 42.6. It's been determined a team of all threshold players would record an average of 50 points in the standings. Therefore the Jackets are projected by GVT to have earned 92.6 points in an 82-game schedule.
Based on their 48-game total of 55 points, we can project Columbus to have earned just shy of 94 points. So GVT's projection of 92.6 point is pretty close. The discrepancy could be due to how much of the GVT of Marian Gaborik, Derick Brassard, Steve Mason, etc. I applied to Columbus' total.
As for Columbus' cap spending efficiency, the Blue Jackets cap payroll at the conclusion of the 2012 - 2013 season was $53.893247MM. The minimum player salary for this year was $525K. Given a roster of 24 players (the maximum allowed) the cap payroll of a club with all Threshold players would be $12.6MM.
If you subtract $12.6MM from their actual payroll you get roughly $41.29MM. Using the concept that it costs $1M on the open market to replace three GVT and consequently one point in the standings, Columbus did an adequate job in terms of payroll efficiency. They spent $41.29MM above Threshold and received 42.6 standings points for that expenditure.
Now here's a list of the 2012 - 2013 Columbus Blue Jackets GVT ratings.
Sergei Bobrovsky - 30.4 GVT (Tops in the NHL)
Mark Letestu - 12.7 GVT (56th in the NHL)
Fedor Tyutin - 11.2 GVT (82nd in the NHL)
Vinny Prospal - 9.7
Artem Anisimov - 7.0
Cam Atkinson - 6.8
Jack Johnson - 6.2
Dalton Prout - 5.6
Brandon Dubinsky - 5.5
Nick Foligno - 4.8
Nikita Nikitin - 4.4
James Wisniewski - 4.2
Matt Calvert - 3.4
R.J. Umberger - 3.3
Derek MacKenzie - 2.8
Tim Erixon - 1.7
Jared Boll - 1.3
Derek Dorsett - 0.9
Colton Gillies - 0.4
Nick Holden - 0.2
Cody Goloubef - 0.1
Adrian Aucoin - 0.1
Jonathan Audy-Marchessault - (-0.2)
Nick Drazenovic - (-0.3)
Sean Collins - (-0.5)
David Savard - (-0.5)
Ryan Johansen - (-0.9)
Blake Comeau - approximately 0.9
Marian Gaborik - approximately 1.35
Right away it's easy to see that without Bobrovsky Columbus was a very average team. That's hard to argue with. According to the metric Bob was the best player in the NHL this season.
Mark Letestu was a tremendous bargain for the Jackets. Even under the terms of his extension, Letestu's salary ($1.25M) would be overshadowed by the cost to replace his production this season ($4.23MM).
Fedor Tyutin's production was worth $3.73MM but his actual salary was $4.5MM. It's not a horrible overpayment but an overpayment it is.
I'm a little surprised Ryan Johansen is rated so low but I have to wonder how much of that had to do with his slow start. I'd be curious how well he performed in terms of GVT after his call-up from Springfield.
GVT may not be perfect but I'd be willing to bet many NHL teams have their own metric, similar to GVT, to perform the same value analysis we've done here. If I'm right this gives us a bit of insight as to how the Columbus Blue Jackets roster might be analyzed by the front office.
Columbus Blue Jackets Blogger • 
Join the Discussion:
»