Varly & Neuvy Head to Head

by NGreenberg on Apr 8, 2011 6:41 PM EDT

With the polls almost closing on who the fans want to see in net for the Washington Capitals, I thought I would share some data I have on the netminding tandem.

First I want to touch on why goalie wins don't tell you much of the story, if anything.

Goalies are only as good as their team, in the vast majority of circumstances. We can get a sense of the role a team plays in a goalie's W% just by the amount of goal support he receives and the goals allowed by the netminder, also known as the Pythagorean expectation. If we know goals-for and goals-against we can estimate W% by GF^2/(GF^2+GA^2). Try it with a few teams and you will see it is pretty accurate, enough for a good barometer.

Michal Neuvirth has a GAA of 2.49 and was given goal support of about 3.2 per game, for a win expectation of .615. If you play 47 times and win 61.5% of your games played we would expect you to end up with 29 wins. Neuvy has 27. Almost exactly what we would expect.

Semyon Varlamov has a GAA of 2.23 but only gets 2.1 goals of support per game, making his win expectation .472. Based on his 27 games we would expect him to win 12-13 games. He won 11, so again, what we would expect.

If Varlamov had the same goal support as Neuvy he would have won ~65% of his games, or ~17 wins for the season. So you see: goalie wins are mostly driven at the team level. Why does the team support Neuvy with 1+ more goals? Who knows, but Varlamov did a better job at keeping goals out of the net despite not getting the benefit of better goal support.

On to the head to head data.

All data is during even strength, except for the PK stats. When I say save percentage (Sv%) what I really mean is scoring chance save percentage. This helps keep the playing field level since the shots are of the same quality. The GF/GA column shows the ratio of goals scored to goals against. Since I don't (yet) have TOI data for these situations I thought it would be helpful to see just how much more support Neuvy gets.

Goalie & Situation (all during 5v5 except for PK data)
Overall	GF/GA	Sv%
Neuvy	1.1	0.852
Varly	1.0	0.879
First 2 periods	GF/GA	Sv%
Neuvy	1.2	0.856
Varly	1.0	0.869
3rd and OT	GF/GA	Sv%
Neuvy	1.1	0.855
Varly	0.7	0.870
Tied after 2	GF/GA	Sv%
Neuvy	0.8	0.837
Varly	~	1.000
OT	GF/GA	Sv%
Neuvy		0.929
Varly		0.625
Close/Late (within 1 thru 2 or tied in 3rd)	GF/GA	Sv%
Neuvy	1.6	0.874
Varly	0.8	0.881
PK		Sv%
Neuvy		0.797
Varly		0.813

I think the numbers speak for themselves. With the exception of OT (where Neuvy clearly comes out on top) Varlamov has performed better in almost every other situation. I know Neuvy has never lost in the playoffs and that means something, just not enough in my opinion to give the Caps the best chance at winning over Varly.

How about you? Knowing what you know now, would you still vote for Neuvirth?

If this FanPost is written by someone other than one of the blog's editors, the opinions expressed in it do not necessarily reflect those of this blog or SB Nation.

Tweet Comment 8 comments | 6 recs |

Do you like this story?

Comments

Should have seen this post coming after all the Twitters comments today.

I’m technically a Varly guy, but frankly I don’t care at this point. They are both good enough to start. If I want to worry about something in terms of the Caps in the playoffs I’ll worry about the power play.

The Artist Formerly Known as CP2Devil.
Associate Editor at Five For Howling.

by Carl Putnam on Apr 8, 2011 6:45 PM EDT reply actions 4 recs

I voted for Varly before, and I think this makes his case stronger. That said, either one is a fine choice.

Matt Bradley: He has sensitive skin, no?

by timmyv38 on Apr 8, 2011 7:02 PM EDT reply actions

I voted for Varly but would not be upset at either one starting. I think, like many, that both will be called on more than just once in this postseason.

That said, many people noted that Varly can be called upon in relief and fill in well, as evidenced the last two years. My question is, what if we give him a chance to take a stranglehold on a series from the beginning? The only series of the three he’s played in which he started was against PIT in 08-09. If he starts last year and we don’t drop the first game and give up 2 in the first period of the second, maybe we close out that series? (Who knows, we’d probably find a way not to anyway).

"Inglewood Jack! Inglewood Jack!" - Coach Jules

by Alz Well That Ends Well on Apr 8, 2011 7:53 PM EDT reply actions

So you see: goalie wins are mostly driven at the team level.

So it doesn’t matter who we pick.

First Round Exit! (Clap Clap ClapClapClap) First Round Exit!

by Rob Parker on Apr 8, 2011 8:17 PM EDT reply actions 2 recs

Correct. I pick Varly in a Neuvy jersey so the team thinks it’s Neuvy. Better SV% and GAA with better run support. Win-win!

"Inglewood Jack! Inglewood Jack!" - Coach Jules

by Alz Well That Ends Well on Apr 8, 2011 8:20 PM EDT up reply actions

Nice post. Rec’d.

Varly it is. And will be, after Neuvy loses Game 1.

Patron saint of quality footwear.

by fat_daddyo on Apr 8, 2011 8:22 PM EDT reply actions

Rec’d, but I still think the difference between the two is razor thin. Which means non-save-percentage considerations can legitimately make the difference between the two. You’re choosing between good and good.

Atta dinnin stick a who!

by Gould Old Days on Apr 8, 2011 10:27 PM EDT reply actions

Why don't you put Holtby's numbers up.....

just for giggles. I know he has a more limited model of work (14 games)….but that is enough to get a comparison in most of the situations you are looking at! ( I already knew he wasn’t included bc he isn’t in the discussion barring crippling goalie injuries).

Yet another offseason......another few months for the Skins to "shine".

by shvd98z24 on Apr 10, 2011 9:12 PM EDT reply actions