Why Building from the Net Out is a Horrible Idea II: The Reckoning

So, my first serious post was not enjoyed... I sort of saw that coming. Mostly because I'm not a particularly smart individual. However, one thing that I am is stubborn and as it were, I'm here to defend my terrible ideas... Partly because I'm as stubborn as a mule, partly because I really do believe that the return one gets comparatively from a forward is greater than the comparative return from a goalie.

The major knock in my analysis is that I'd compared the in game impact of one individual to a group of twenty. Please understand that that was, in no way, the case. In fact, I agree: the goalie is the single most important player on a team. That is not what I am arguing. What I am arguing is that there is so much parity between goalies, that all starters are so good at the NHL level, that their relative impact from one to another is somewhat negligible.

Moreover, this allows me to compare the impact of one forward to one goalie and their comparative impact on the game.

The league leader in GAA is Annti Niemi (his gaa is skewed by a great defense, mind you) and sits at 1.98. The average GAA is 2.64. This accounts for a difference of.66 goals per game. Projected over a 82 game season this translates to 54 goals.

The league leader in goals is Alex Ovechkin with 42 goals, the average is 7 (his point totals provide an even larger difference compared to the average-89 to 17). This is a 35 goal difference after 53 games. Coincidentally, this translates to a goal differential (when compared to the average) of .66.

So right there, we are comparing the impact of two singular players, fully neglecting the impact AO has on his linemates (and his 40+ assists). More importantly, a star's presence provides a ripple effect for all of his teammates. The quality of opposition falls for all other lines. This is evidenced by Mike Knubles jump in point scoring despite his age

By the way, Aaron Asham recently scored his 7th goal of the year.... Who falls on the average for GAA and SV% respectively? Marc Andre Fleury and Marty Turco... Just some food for thought about comparative impact

Why Building from the Net Out is a Horrible Mistake

There's an old adage in hockey that says: You build from the net out. This adage points to the idea that Goalies are the single most important position in the NHL, and while the goaltender is a very important position, it is hard to for me to fathom the importance placed on the position. In fact, I'd say that , given the choice between a star position player and a Goalie one should just about always choose the position player as goalies are overvalued at the NHL level. This may sound odd but let me explain:

Before we get into any of this I'd like to present some numbers to you. These are the save percentages of goalies that get the majority of starts for a team. Two ends of the spectrum are shown high sv% to the left, low sv% to the right (Vesa Toskala is not a starter, I don't care what anyone says):

.931-- .908
.931-- .906
.927-- .905
.925-- .904
.924-- .903
.927/913-- .894

The first thing that you may notice is that there is not a drastic difference in these numbers. Granted, .931 to .894 looks incredible but that's 37 more goals out of 1000 shots. To put into perspective, a Sv% of .931 is only 4.1% more effective than a Sv% of .894. Again, the most effective starting goalie in the NHL is 3% better than the least effective starting goalie in the NHL and will have 28 fewer goals against.

Now, Here are the teams corresponding with the Sv%s listed above:
Florida-- Washington
Buffalo-- Pittsburgh
San Jose-- Philadelphia
Calgary-- Tampa Bay
Colorado-- Chicago
Montreal-- Columbus


The left hand column has only 1 team not in a playoff position (Florida). The right hand column also has 1 team on the outside looking in (Columbus). I guess what I'm pointing to here is that there is little correlation between Sv% and team success.

Still, when you think about it it shouldn't be that shocking. As I stated before, the most effective starter is only 4.1% better than the least effective starter. Still, the case against the importance of a goaltender gets even more damning when you consider the offensive numbers of a team and how they relate to success. We just looked at 6 of the top and bottom teams in terms of Sv%. Now we'll look at the top and bottom 6 teams in terms of shooting%.

11.9%-- 8.5%
10.6%-- 8.4%
10.5%-- 8.1%
10.4%-- 8%
10%-- 7.7%
9.9%-- 7.5%

Consider the highest shooting % compared to the lowest: .119 to .075. The highest shooting % is 58% more effective than the lowest and will score 44 more goals per 1000 shots. Remember the Goalie comparison indicated a 4.1% increase in effectiveness and a decrease of 37 goals against. Long story short, 44>37

Now here are the teams that correspond with those numbers:

Washington-- New York Rangers
Colorado-- New Jersey
Vancouver-- Toronto
San Jose-- New York Islanders
Los Angeles-- Detroit
Atlanta-- Boston


The left hand side has the teams with the 6 highest shooting%s. Of those 6 only 1 (the Atlanta Thrashers) is currently not in a playoff position. On the right there are the 6 lowest teams in terms of shooting%. Only one team is currently in a playoff position. The correlation here is incredibly strong. This isn't a phenomenon exclusive to the top 6 either. 11 of the top 15 teams in terms of shooting % are in playoff positions, 10 of the bottom 15 are not.

The real issue here isn't whether or not a goalie is important (it is important). The issue is that the disparity in effectiveness between a group of forwards, on average, is going to be greater than the disparity between two goalies. The issue is that building from the net out offers, at most, a 4.1% increase in effectiveness whereas taking the opposite route offers a 58% increase. This is why I see the goalie position as being overvalued in the NHL. This is why building from the net out is a horrible mistake.

What say you?