The Goalie-Centric Model of the Universe

As an aside, I really don't understand why Galileo is the famous one. Copernicus figured out the heliocentric model 30 years before Galileo was born. Kepler used Brahe's data to figure out the orbits were elliptical. But Galileo gets a hed (P.E.) song. Go figure. (I'll wait while you figure.) Anyway....

The goalie is the most important player on a hockey team, right? You can't win the Cup without an elite goaltender, right? Wrong and wrong.

A goalie is important. Which goalie is not that important.

Sure, simply having a goalie is important. But the difference between the best goalies in the league and the dime-a-dozen guys up from the AHL is wafer thin. Shot differential plays a bigger role in how many goals a team gives up. Last season, the Devils gave up only 1668 ES shots against. Toronto gave up 2411. The league average was 2004. The best goaltenders in the league have career ESSP of about 0.930. Average is 0.920. The worst regular goalies in the league are around 0.910.  Here's a table of goals against for the various combinations of shots against and save percentage:

No matter where you are in this table, you are better off moving left (cutting down your shots against) than moving up (improving your goaltending). For example, average team with average goaltending gives up 160 goals. Moving left, to elite shots against, is 27 goals better. Moving up, to elite save percentage, is 20 goals better..

You can win a Cup without an elite goaltender

Since the first lock-out, let's look at the teams and goalies who won the Cup.

Osgood was lousy. Ward and Fleury are below-average. Thomas was elite. Everybody else falls into the general average starter category.

Goalies just aren't the center of the hockey universe. Shot differential is far more important. Even within the confines of team defense, shots against has a bigger impact on goals against than goaltending.