Goaltender Save Percentage by Age

by Hawerchuk on Apr 28, 2010 12:57 PM EDT

Let me add another chart to the discussion:

It's hardly overwhelming evidence, but it would *seem* that the very best goaltenders peak at age 25. Weaker goalies get cut as soon as they have a bad season, which is why save percentage seems pretty constant for the 5000-20000 minute group.

Tweet Comment 8 comments | 0 recs |

Do you like this story?

Comments

The line keeps dropping as the goalies get older.

And we in Oil Country think one dark thought: Khabibulin, three more years.

by David Staples @ The Cult of Hockey on Apr 28, 2010 2:24 PM EDT reply actions

A sobering thought.

by Mogen_david on Apr 28, 2010 2:26 PM EDT up reply actions

Similar to the PPG theory.

Blueshirt Banter - End the Sather Era

US Soccer - The Yanks Are Coming

Red Sox Fan behind Enemy lines.

by Rob L on Apr 28, 2010 2:24 PM EDT reply actions

How much noise does era and the rule changes throw into this data or is that what you normalized. That may have thrown some pretty loud noise into the last analysis.

by Mogen_david on Apr 28, 2010 2:24 PM EDT reply actions

I used 1983-84 to 2008-09. I normalized every goalie’s goals against to an .896 league-wide save percentage. The results – if you don’t normalize – show a completely different trend.

by Hawerchuk on Apr 28, 2010 3:08 PM EDT up reply actions

Nicely done, sir.

Excellent work, by the way, Gabe. I’ve always wondered how goalies tend to perform as they age, and whether they drop off faster than position players, and this is a step towards answering that, a big step.

And for the non-math guy in the audience (me) can you briefly explain what you mean when you “normalize” the results. That league-wide .896 save percentage, is that the average save percentage from 1983-2009.

by David Staples @ The Cult of Hockey on Apr 28, 2010 4:09 PM EDT reply actions

I'm not Gabe...

But I assume when he says normalize that he treats adjusts the save percentages to correct for the fact that the league wide save percentage has increased. For example, a .900 save percentage was really good 20 years ago; now it’s bad. If you don’t correct for that, your answer is going to screwy because any trends will be altered by the rising tide in save percentage.

by Tyler Dellow on Apr 28, 2010 5:39 PM EDT up reply actions

Good stuff Gabe. Am I right in assuming that this is overall save percentage and not EV save percentage? Regardless, I’m still somewhat surprised that the change isn’t more pronounced. .899 to .904 isn’t exactly a big leap over the development years and that effect is only present for the guys who make it. It looks to me like betting on potential over established ability is a bad move with goalies based on this chart. I find that really surprising. I would have thought that most goalies show substantial improvement from age 18 to age 25.

by Scott Reynolds on Apr 28, 2010 4:19 PM EDT reply actions