The Even-Strength Shutout

[Mike]

We have a new post on the front page where we explore the idea of tracking even-strength shutouts. I hope you guys enjoy it. It is the first part of an interesting goalie statistics series which eventually culminates into some pretty cool fantasy hockey advice. Stay tuned for the future pieces due out this week.

http://www.leftwinglock.com/articles...ength-shutout/

[Mike]

And now, part 2: Penalty Kill Save Percentage (PKSV%)

http://www.leftwinglock.com/articles...rcentage-pksv/

[internalprime8]

Interesting articles. You guys have really dissected how to predict goalie statistics over a period of time.

[Mike]

Interesting articles. You guys have really dissected how to predict goalie statistics over a period of time.

Part 3 comes out today and we get into the meat of how this relates to fantasy hockey.

[internalprime8]

Can't wait. The graph for PKSV% was particularly eye-opening.

[Mike]

Spent an extra day on this and part three is now ready. It outlines a method for how to project a goalie's save percentage:

http://www.leftwinglock.com/articles...antasy-hockey/

This is the article most relevant (of the three) to fantasy hockey.

[Mike]

We've added a Part IV to this series on the penalty kill save percentage. In this article, we examine which goalies from the 2011-2012 season were hurt or helped the most by their PKSV% numbers. This gives you a quick look at which goalies are ripe for a rebound or drop in stats for the upcoming season:

http://www.leftwinglock.com/articles...-the-pk-boost/

[Tim]

We've added a Part IV to this series on the penalty kill save percentage. In this article, we examine which goalies from the 2011-2012 season were hurt or helped the most by their PKSV% numbers. This gives you a quick look at which goalies are ripe for a rebound or drop in stats for the upcoming season:

http://www.leftwinglock.com/articles...-the-pk-boost/

I have a question about this. I was looking through the article (great read btw) and I just noticed that in that graph almost all the goalies you have being inflated by their pk save percentage play on the better penalty kills in the league and all the goalies you have being deflated play on the worse penalty kills (most being near the very top or very bottom). Of course they are going to be somewhat connected but it looks like they are almost 100% connected. If that's true then wouldn't their pk save percentage be a predictable stat?

[Mike]

I have a question about this. I was looking through the article (great read btw) and I just noticed that in that graph almost all the goalies you have being inflated by their pk save percentage play on the better penalty kills in the league and all the goalies you have being deflated play on the worse penalty kills (most being near the very top or very bottom). Of course they are going to be somewhat connected but it looks like they are almost 100% connected. If that's true then wouldn't their pk save percentage be a predictable stat?

That's a really good question and it stumped me for a bit (and I'm still thinking about it more deeply). I did just now look at the Top 15 PK teams from 2010-2011 (by %). I then checked that list for 2011-2012. 7 of the teams that were in the Top 15 for 2010-2011 did not appear in the Top 15 for 2011-2012. This is by no means robust, but it provides more evidence for the little to no correlation theory. If all 15 had dropped out, you'd expect a strong negative correlation. If all 15 had stayed in, a strong positive. But, half dropped out.

It is possible, when you consider the role of backups, that individual goalies would show no correlation in year-to-year PKSV% numbers but that their effects could combine to reveal some non-zero level of team correlation. But I can't imagine it would be very strong.

[Mike]

I eyeballed a few teams over the years and SJ, LA, and CHI all seem to bounce around a bit in the PK% standings. But, Toronto has remained near the bottom of the list for years - so that might be an indicator of a team problem and not a statistical fluctuation.