Thread: Proposal: Detroit - Vancouver
View Single Post
11-02-2012, 06:46 AM
Registered User
palindrom's Avatar
Join Date: Apr 2011
Posts: 4,148
vCash: 500
Originally Posted by eklunds source View Post
Just to address Howard vs. Luongo a little bit here..

- Goalies have a lower save percentage when their team is shorthanded versus when they're at even strength. That's not rocket surgery or brain science. As such, teams who take more penalties are 'artificially' lowering their goaltenders' save percentage. The same effect is more pronounced when a team goes down 5-on-3; some years it just happens more often than others so comparing a two goaltenders' overall save percentage on different teams is useless.

- Goalies who face less shots allow less goals. That's pretty much common sense too. A good defensive team limits the number of shots they give up, so measuring goals allowed per unit of time (goals/season, GAA, etc) is useless between teams too.

- Going back to even strength save percentage... It's also very variable year-to-year. A goalie with a talent level of, let's say, 0.925sv% at even strength might post anywhere from 0.920 to 0.930 in any given year, or maybe even just outside of that. There are so many goals that occur that are completely out of a goalie's control that it's pretty much random luck. Sometimes the puck goes in when it really shouldn't, and sometimes it doesn't go in when it really should. See below:

- There's so much variance in goaltending, as it were, that it takes about 3,000 shots at even strength before you really know how good a goalie is. If you don't believe me, look at Steve Mason - he posted a 0.925sv% at even strength over his first 61 games, and since then he's been under 0.900.

Over the past 5 seasons, Luongo has had even strength save percentages of:

So we can probably safely assume that his 'talent level' is somewhere in the ballpark of a 0.930sv%. As it turns out, Luongo's even strength save percentage has been between 0.925 and 0.936 every single year of his career, including when he was in Florida.

Howard (going back 3 years):


So, probably somewhere in the ballpark of 0.923, 0.924. That could change, it's still pretty early in his career, but goaltenders don't typically improve radically after age 27..

The difference between a 0.930sv% and a 0.924sv% at even strength, if we assume a goalie faces ~1300 shots (that's about 65 games), is 8 goals a season. Goalies who are better at even strength are also better on the penalty kill, so let's say 9 goals.

A goal differential of +3 goals gives you ~1 point in the NHL standings, so the difference between Luongo and Howard as your starting goaltender is somewhere around 1.5 wins per season, or 3 points in the standings.

Of course, that doesn't take into account length of service (Howard probably has 5-6 years more than Luongo left in him) or other factors, but as a rough guide it's probably pretty accurate..
Very nice analysis, we use the the same method.

Now the real question is How much does this 3 point in the standing worth in term of cap hit / lost of asset to make the upgrade ?

will downgrading the defense to upgrade the goal-tending have a greater positive or negative impact?

Last edited by palindrom: 11-02-2012 at 07:10 AM.
palindrom is offline   Reply With Quote