FanPost

Regression is Inevitable - Example 124030: Evgeni Nabokov

A little more than a month ago, I wrote the following:

1. Nabokov's high SV% is not REAL - it's a facade caused by luck on special teams that is likely to crumble. Nabokov has essentially been the 22nd best goalie in the league this year - NOT the 7th best as his ordinary SV% suggests. He is NOT an elite goalie (though he is certainly average) and the longer he plays for the Isles, the more likely his true talent will cause his results to drop, and lower his trade value.
......

As a result, Hockey statisticians rely instead upon Even Strength Save % (ESSV%) to judge goalies, as that way we eliminate the problems caused by including Special teams Data. And by this measure, Nabokov is actually 22nd in the league among NHL goalies with 20 games or more played (.925 SV%). The Reason why Nabokov is 7th in overall SV% is simple: He's gotten insanely lucky on the Islanders PK, with a .951 PKSV% (2nd in the league)! Before anyone asks, this is nowhere near anything Nabokov has done in his career, where he usually has PK SV%s of .880 or so.

None of this should be surprising to any of you - Nabokov's current SV% would be his career high at age 36 and far above his .913 carer average.

Long story short, Nabokov's #s are a good deal smoke and mirrors - and are far far more likely to get worse than go up

-Garik16, February 12, 2012

And here were a few of the responses (Names withheld to protect the non-believers):

I don't think he's been lucky...
Thats unfair to say...What I see is a totally relaxed goalie who has closed the holes in his game...
.....
But to say hes simply been lucky is ****ing ridiculous...


-Poster 1

Luck is just a variable that the statisticians have failed to define.
You cannot base an argument on empirical evidence AND luck, since luck is not quantifiable. Just my two cents when discussing a player isn't as good as their stats because they are "lucky". All along I thought stats were defined to determine actual performance.

-Poster 2

--Just hypothesizing on why his numbers are at a career high this year: Nabby was able to take some time off last year and really get his head on straight after bouncing to Russia, then back to the NHL and being unhappy with what team he was going to have to play for. I think taking the time to realize that hockey is what's important to him was a good thing. So being rested with his head on straight is one possible reason for the numbers spike.
--Another possible reason for the numbers spike is that he's playing in a new conference where his stand up style of goaltending isn't often seen. It's going to take some time for guys to get used to shooting on that style of goaltending. This would suggest his numbers aren't going to last forever if that's the case.

-Poster 3

"He's gotten insanely lucky on the Islanders PK"
While there is probably a small luck factor there, the Islander PK has been very strong this year. They've limited the amount of qaulity scoring chances and done a good job of intercepting passes, clogging shooting lanes and forcing shots from the perimeter.

-Poster 4

So if there's always a small element of luck, you can't call him lucky. I.e. if everyone is special, no one is.

-Poster 5


Evgeni Nabokov's SV% on Feb 12, 2012: .927
Evgeni Nabokov's EVSV% on Feb 12, 2012: .925
Evgeni Nabokov's PKSV% on Feb 12, 2012: .951

Evgeni Nabokov's SV% on March 19, 2012: .913 (-.014)
Evgeni Nabokov's EVSV% on Mar.19, 2012: .917 (-.008)
Evgeni Nabokov's PKSV% on Mar.19, 2012: .894 (-.057)

Luck does not last forever. Regression Comes for all.
.
.
.
.
.
.
.

Okay, not to be totally obnoxious here, what we have here is a fairly typical lesson on Regression. We can tell through statistics when players are getting particularly lucky. What happens with such players is that, if they continue to play, their #s will REGRESS back to their true talent. We saw this with the SV% of goalies with Milan Jurcina on the ice last year (which was originally really high before coming down to earth) and now we see it with Nabokov.

Nabokov was getting extraordinarily lucky on the PK. That luck ended, and his numbers collapsed (No longer the #7 goalie in SV% but instead he's down to the #25 goalie.)*

*I didn't make a point on this in the other post (I did talk about this on twitter), but there was a good case that Nabokov's EV SV% was high in February and indeed that's dropped as well (his post-lockout EVSV% in February was .9186, now this year it's .917 instead of .925).

The end result of this has been that Nabokov no longer looks that super star goalie you have to lock up.* In fact, this does once again point out that signing Nabby to an extension, particularly one that lasts more than one extra year, is at best a bit silly and at worst (a multi-year deal) ridiculous.

*It's possible there was no market for Nabby (or that he wouldn't waive his NTC), which made trading him impossible. But not even talking about "should have traded him" anymore.

<em>Submitted FanPosts do not necessarily reflect the views of this blog or SB Nation. If you're reading this statement, you pass the fine print legalese test. Four stars for you.</em>

X
Log In Sign Up

forgot?
Log In Sign Up

Forgot password?

We'll email you a reset link.

If you signed up using a 3rd party account like Facebook or Twitter, please login with it instead.

Forgot password?

Try another email?

Almost done,

Join Lighthouse Hockey

You must be a member of Lighthouse Hockey to participate.

We have our own Community Guidelines at Lighthouse Hockey. You should read them.

Join Lighthouse Hockey

You must be a member of Lighthouse Hockey to participate.

We have our own Community Guidelines at Lighthouse Hockey. You should read them.

Spinner

Authenticating

Great!

Choose an available username to complete sign up.

In order to provide our users with a better overall experience, we ask for more information from Facebook when using it to login so that we can learn more about our audience and provide you with the best possible experience. We do not store specific user data and the sharing of it is not required to login with Facebook.

tracking_pixel_9355_tracker