May 20, 2010
From Daigle To Datsyuk
The Scouting Side Of Goaltending Parity, Part One

by Corey Pronman

There’s a saying about goaltending that, “There’s no single more important position in all of sports, than a goalie for a hockey team.” It’s a phrase that can be understood to a certain extent. In this sport with such a small scoring context, where one or two goals can swing the course of a game dramatically, having a goalie that makes a mistake once or twice can be the difference. On the flipside, a goalie who takes away chances that other netminders would give up swings the game dramatically in their respective team’s direction. Thus, you can see why goalies highly contribute to a team’s overall performance level.

If you took Tom Awad’s GVT or Alan Ryder’s Player Contribution and saw the proportion of contributions that are awarded to goaltenders relative to skaters, you would see how influential the contribution level for goaltending is. In Ryder’s PC, around 18-20% of total PC is usually allocated to the goalie, whereas in Awad’s GVT it’s around 8-10%. Regardless, this is significant as this means that usually only one player racks up a large proportion of a team’s total win shares as compared to the rest of the roster. If you look at any overall PC or GVT chart with the top performers listed, there are usually plenty of goalies near the top. The reasons for why they have such a high amount of win shares include that goaltenders play the most minutes of any other player, which is the nature of the position, not to mention that they are directly tied to events that either prevent or allow goals (shots on net), which is the ultimate tie-in to player valuation on the defensive side.

Tom Awad recently produced some fabulous pieces on goaltending, and more importantly took a look at the parity of goaltending in the present day NHL, which has been the result of a lack of separation in true skill level. Take a look at the 2009-10 GVT/Game and you’ll see a major spread between the “elite” goaltenders and the middle of the pack, as well as the steepness in the drop-off. Here are the top 5 goalies:

Name 			GVT/Game
Tukka Rask		.543
Ryan Miller		.482
Jaroslav Halak		.438
Evgeni Nabokov		.391
Jimmy Howard		.374

The difference between Rask and Miller is significant, with essentially a half goal extra per start separating the two. The drop from Miller to Halak is also significant and relatively the same from Halak to Nabokov. However, it’s from that point on where the drop-off becomes starts to level off. From the 5th ranked goalie Howard at .374, to the 14th ranked goalie Thomas at .201, the difference is the same as it is from Rask to Howard. Here’s a graph showing the GVT/Game with a threshold of a minimum 40 games played from Rask at 1 to Fleury at 24:

The mean for this graph is .242, and the closest to that is Kiprusoff with a .241. However one standard deviation is .146 which from that .242 mark barring covers Howard (5th) all the way down to Antero Nittymaki (20th). This is a normal distribution as one standard deviation covers 66% of the distribution, but what’s worth noting is how flat the graph gets after Henrik Lunqvist, basically illustrating the parity of goaltending in the NHL outside of the top goaltenders.

To a degree we already knew of this result thanks to past analysis done at Puck Prospectus. However, the next question to ask is why? Why is it that the goalie situation in the NHL is like this? Well my answer is not a statistical one but rather an analysis of how goalies play right now based on my scouting knowledge of the game. Back in the 1980’s and early 1990’s, goaltending was vastly different than in today’s day and age. Goalies back then stood up straight and relied on pure reflexes to make stops. This wasn’t exactly a successful strategy as Patrick Roy led the league 1988 with a .900 save percentage. The difference between Roy and the 5th ranked goalie Brian Hayward (also a Montreal goalie) was small at a .004 save percentage difference. Between Roy and John Vanbiesbrouck, ranked 10th in save percentage, was a .010 difference and between John and the 20th ranked goalie Greg Millen was also a .010 difference. Regardless to say, the difference between Millen, who at 20th was still considered average and Vanbiesbrouck was significant, and the difference between him and Roy/Hayward was enormous. This goes to show that there actually was a talent separation between tiers of goalies back then that we don’t have today.

So what happened between then and now? Why is the separation between goalies in the NHL today not as significant? Well, it actually has a lot to do with the aforementioned Patrick Roy and the butterfly style of goaltending. It of course started with Jacques Plante, but Roy really brought forth the idea of playing the percentages, cutting off the angles and the idea of making the shooter beat you. This playing style emphasized the use of natural reflexes, a reliance on pure ability and a transition towards the use of the head and a predetermined style of play that gave goalies a better chance at stopping a larger percentage of shots.

This use to be the primary style of play in the “Quebec school of goaltending”, however it has quickly turned into the universal style of goaltending. Kids at the youngest of ages are taught to get down on shots because the percentages dictate they have the best chance at making the save by doing so. It’s bolted into their minds that if there’s any uncertainty as to where the puck is heading, they should get down. Even if it’s around the mid-section, they should still go down to defend the shot. What if a goalie doesn’t have the time to track the puck? The reaction is still the same: go down and defend. It gets to the point where netminders get into their butterfly instinctively. There is so much focus on positioning, angles, how far out or in the crease the goalie is, the positioning of the hands and legs upon entering the butterfly and how square the goalie is with the puck, that everything else is forgotten.

In the coming days when scouts are looking at young goalies in preparation for the 2010 NHL Entry Draft, they are looking for a few key things:

• Size: The ability to cover a large part of the net, especially in the butterfly position, as the shoulders cannot allow too much room between them and the crossbar.
• Rebound Control: When a goalie enters the butterfly, he makes himself susceptible to rebounds by how the pads tend to deflect the puck away, and how a shot to the goalie’s chest would deflect a puck away. It’s important for a goalie to be able to have the lower body strength and ability to push the puck out of the danger areas or be able to stop the puck on a dime and freeze the play. This area is pretty coachable.
• Skating Ability/Movement: It’s funny to think of looking for skating ability in a goalie, but it actually is important, especially in the butterfly era. How well a goalie can move side to side, come out to challenge a shooter and retreat back into the crease amongst other things is significant because the modern day goaltending style requires plenty of movement.
• Positioning: Essentially, this is a goalie’s ability to read the game, which is a very coachable part of a goaltender’s arsenal. A goalie’s knowledge of what to do and where to be when the puck is at one point and the play is developing at another is vital. It’s the ability to stay square, while managing the proper angles and knowing the proper distance between the goalie and the net in his crease at every moment. It’s also about knowing how to place your body in the perfect position to up the percentages in your favor. In essence, it comes down to how well a goalie thinks and sees the game, which is by far the most complex factor and the finest aspect of goaltending.
• Reflexes/Athleticism: This covers how well a goalie reacts to all situations. Does he have quick hands and feet? When he’s down or caught out of position can he make acrobatic saves? Can he snag a hard wrister out of the air or make a big stretch for the puck? What’s his recovery ability like? There are many other things of course that I’m omitting, such as puck-handling, the mental game, etc. but those are just some main areas that scouts evaluate.

So what happens when you take a man with decent to good size, teach him how to play the percentages as a goalie, coach him to play the proper butterfly, and sprinkle in some reflexes, athleticism and rebound control? You get the modern-day goaltender and it’s gotten to the point where that formula for producing a good goalie is easier than it was decades ago because the main ingredients for being a goaltender are now more coachable than they were in the past. This doesn’t mean that anyone can be a goalie; you do need talent and not everyone can be coached if you don’t have the mental focus, in addition to putting in the practice time and showing the dedication. However, there is a reason why so many goalies in the AHL and NHL look so similar in their styles of play to the point where it looks nearly robotic. The style doesn’t require natural talent fully to succeed and most importantly, it works! Look at save percentages from years past to the modern day, and you’ll notice a significant increase. The game may have changed and pads have gotten bigger, but goalies have indeed improved collectively.

We still have outliers, however, from this big group of goaltenders who are all close to one another talent-wise, such as taking into account Miller’s outstanding year and Thomas’ great season in 2008-09. Some will remember Luongo’s brilliant years, Theodore’s MVP season and the Dominator… well… dominating. My theory on this is that these goalies with great seasons are able to save the high percentage shots with a higher volume of frequency. Today’s modern way of goaltending eliminates the overwhelming majority of low percentage shots, and to a degree, the medium quality shots as well. The goalies who can frequently and consistently make the big save, or rather the most difficult saves are the ones who become outliers from the main part of the distribution.

There’s a trend with the goaltenders I listed in the graph above- these goalies tend to be very acrobatic, have good athleticism and have great reflexes. However, when combined with a good butterfly style and an ability to play the angles and percentages properly, you have a goalie who has a good shot at producing an unearthly save percentage. Some were unable to sustain their level of play, which goes to natural regression and goaltending in its nature and the true talent level coming to the forefront. However, aside from Vokoun, I rarely see a goalie post a save percentage over .925 without being able to make a lot of outstanding and difficult saves. This doesn’t mean goalie’s with great reflexes will automatically be great even if they adopt the butterfly style, such as Fleury, as there has to be a balance between having the skill and the modern day style of play. However, it gives goaltenders a solid shot at having an impact in the NHL.

Tomorrow, we’ll take a deeper look at the parity of goaltending, plus the contracts and salaries of goaltenders as well as where to draft netminders.

Follow Corey on Twitter at @coreypronman.

Corey Pronman is a contributor to Puck Prospectus and runs the statistical hockey site The Hock Project. You can contact him at CPronman@fau.edu.

Corey Pronman is an author of Hockey Prospectus. You can contact Corey by clicking here or click here to see Corey's other articles.