Vegas Hockey Knight
Fansided

Las Vegas Hockey: Updating Plus/Minus To A Relevant NHL Statistic

By Nicholas Frost

PHILADELPHIA, PA - JUNE 28: (L-R) Brendan Shanahan and Brian Burke of the Toronto Maple Leafs talk on Day Two of the 2014 NHL Draft at the Wells Fargo Center on June 28, 2014 in Philadelphia, Pennsylvania. (Photo by Bruce Bennett/Getty Images)
PHILADELPHIA, PA - JUNE 28: (L-R) Brendan Shanahan and Brian Burke of the Toronto Maple Leafs talk on Day Two of the 2014 NHL Draft at the Wells Fargo Center on June 28, 2014 in Philadelphia, Pennsylvania. (Photo by Bruce Bennett/Getty Images) /
facebooktwitterreddit

When this era of NHL hockey is looked back on, it will be known as the infancy stage of advanced stats and analytics. Statisticians and hockey experts are coming up with new and improved methods of predicting player’s usefulness on seemingly a monthly basis. But not everyone agrees with the weight placed on these predictors when evaluating players. Lots of pundits still deploy traditional scouting strategies, such as Coach’s Corner host Don Cherry, or President of Hockey Operations for the Calgary Flames, Brian Burke.

There is clearly a divide between the old school way of valuing performance with the “eye test” (basically traditional scouting), and the new way of tracking value through advanced statistics. You need look no further than Twitter to find debates on which is the superior method. Read about the Shea Weber for P.K. Subban trade, you’ll see what I mean.

For the purpose of this article, we are going to focus on the out-dated statistic of plus/minus, and how it could be greatly improved. For those of you who are new to hockey, or the plus-minus rating, it is calculated as such: for every time a player is on the ice when a goal is scored for their team, their plus/minus will go up by 1. For every situation where they are on the ice when a goal is scored against their team, it will go down by 1. Power play and shorthanded goals, as well as shootout goals, do not count. Overtime and empty-netters do. Traditionally, a high plus/minus number was indicative of a strong player. But recently this gauge has come under scrutiny and almost dismissed entirely.

The problem with the plus/minus is that a player could be just stepping on or off the ice, and can be docked a point or awarded one, despite having zero impact on the outcome of the play. The other issue with plus/minus is that a really good player can be dragged down by poor line-mates, as well as the opposite. So I say to myself “Self, how can we improve this archaic method of valuing plus/minus?” and I think I have a basic premise for an upgrade.

Before I get into the logistics of this upgrade, I’d like to point to one of my all-time favourite goals as an example. Mats Sundin’s overtime goal vs. the Ottawa Senators in the 2000-2001 Stanley Cup Playoffs:

If you watch the play directly before the goal, #32 Steve Thomas crashes the net hard. Both #16 Mike Sillinger and #6 Wade Redden cover him, leaving #13 Mats Sundin with all the time in the world to unload a slap shot off the post vs. a defenseless Patrick Lalime.

Now according to plus/minus statistics, all Leafs on the ice would receive a +1 rating, and all Senators would receive a -1. That hardly seems fair, does it? The two key contributors involved in this goal for the Leafs are clearly Sundin and Thomas. Conversely, the two Senators who should be statistically punished the most are Redden and Sillinger. The other skaters on the ice are basically being rewarded or penalized for a play they had little impact on.

So how could this be improved you may be asking? I propose that a number be given to players on the ice for their importance to each and every goal scored. For example, based on this small clip, if it were factored on a 1-5, it would be something like this:

Toronto Players : Mats Sundin 5, Steve Thomas 4, Gary Roberts 2, Bryan McCabe 2, and Tomas Kaberle 1.

Ottawa Players: Wade Redden -4, Mike Sillinger -4, Karel Rachunek -3, Rob Zaumner -1, Unknown Ottawa forward -1.

This is all a very subjective process. As you can see, I have awarded Sundin the top points available on the goal at 5. Not every goal would be a 5, but this shot was surgical. Mike Sillinger, who gravitated to Thomas rather than the puck carrier, received a -4. Steve Thomas gets a 4 for crashing the net and drawing 2 Ottawa players with him. He is the catalyst of this play, and despite not scoring deserves top marks. Wade Redden rounds out the top components of the goal being scored at -4 as well.

An argument can be made against severely docking Redden and Sillinger. Clearly Sillinger went to Thomas assuming Redden would step up and cover Sundin, but Thomas skates directly into Redden. It really is a great play by Stumpy! Also, Karel Rachunek appears to be the defenceman jumping into the rush moments before the goal, and is caught up-ice. This is why he is rated a -3. In said Ottawa rush that preceded the goal, Gary Roberts breaks up the play and Bryan McCabe clears the puck. They both were awarded +2 for this transaction.

I am sure you can now see that as I have stated, this is an entirely subjective process. Those with the strongest grasp on hockey fundamentals/ coaching strategies could benefit the most. For all I know, this process of valuation of goals is already being used, and if so, do I ever have some egg on my face! But if not, all I ask is that if you expand on it, you kindly name it after me. And maybe dinner at a nice restaurant, like Mendy’s (soup doesn’t count).

If you would like to discuss the pros and cons of this system, or have any other comments or suggestions, please feel free to follow me @nickeleyefrost on Twitter, as well as @FSVegasHockey.

Home/Analysis