Ryan’s Hopeful

Same calculation of Corsi as before (see previous article), but for the 2013/2014 season.  (Note that my dataset has three or four missing games for the season, but that should not affect the calculations much).

The Anaheim drop is spectacular.  Maybe explains the regular season success and post-season “not so much”.  Also is consistent with LA’s win – they beat up on poor teams, but they also beat the snot out of good teams.  Man.

Team EV Corsi% EV RyanCorsi% Delta
ANAHEIM 50.0% 42.9% -7.1%
PHOENIX 50.5% 47.1% -3.4%
BOSTON 53.6% 51.6% -2.0%
BUFFALO 43.3% 41.1% -2.2%
CALGARY 46.8% 43.7% -3.1%
CAROLINA 50.2% 46.5% -3.7%
COLORADO 47.0% 45.4% -1.6%
COLUMBUS 50.0% 47.5% -2.5%
CHICAGO 55.6% 52.6% -3.0%
DALLAS 50.5% 49.1% -1.4%
DETROIT 51.4% 48.7% -2.7%
EDMONTON 44.8% 42.6% -2.2%
FLORIDA 50.6% 47.9% -2.7%
LA 56.6% 53.4% -3.2%
MINNESOTA 48.3% 45.9% -2.4%
MONTREAL 46.8% 46.5% -0.3%
NYI 49.4% 48.7% -0.7%
NYR 52.2% 50.3% -1.9%
NASHVILLE 48.4% 45.7% -2.7%
NJ 54.1% 51.1% -3.0%
OTTAWA 51.8% 50.2% -1.6%
PHILADELPHIA 50.3% 47.7% -2.6%
PITTSBURGH 48.8% 45.9% -2.9%
SAN JOSE 53.3% 50.1% -3.2%
STL 53.0% 49.8% -3.2%
TAMPA 50.9% 47.2% -3.7%
TORONTO 43.4% 41.2% -2.2%
VANCOUVER 51.3% 47.7% -3.6%
WASHINGTON 47.8% 43.9% -3.9%
WINNIPEG 50.4% 48.3% -2.1%
Advertisements

6 thoughts on “Ryan’s Hopeful

  1. Really interesting stuff… 🙂

    Funny coincidence how 3/4 teams in the conference finals were top 4 in Ryancorsi. The one exception being Montreal who with Price beat the Bruins in a seven games series. The Bruins were 3rd on Ryancorsi.

    While the Anaheim drop is spectacular, the Blues were also team that takes a bit of a hit. Montreal jumped from 16th to 12th, so maybe they were a slightly better team than they looked like outside of Price.

    Either way, whether or you use Corsi or Ryancorsi, the teams at the bottom end of the rankings look like poor bets unless they have spectacular goaltending.

    Like

      1. No, I don’t have a tenable solution to the SPSS problem. Until fairly recently, I used to do academic research with the U of A–so I still had a copy of SPSS on my macbook from that endeavor. Now that it’s gone, it’s gone. SPSS cost something like $5700 for a single user license…

        I could look into R, but it’s probably too complicated for me… If I really need to scratch this itch, I could possibly buy a copy of Stata since it’s fairly reasonable. 🙂

        Like

      2. The one and only time I had/used SPSS is during the Stats course of my MBA. That was also an academic license, and disappeared when that machine fried, like your Mac. Since then, I make do with the “Data Analysis Toolpak” add-on for Excel. I figure if it’s more complicated than what the add-on can handle, best to leave the problem to someone who knows what the heck they’re doing!

        EDIT: Also, I did some poking around. Have you thought about trying PSPP? Supposedly an open source replacement for SPSS. Looks like it’s being actively maintained, and in any case, I doubt this problem is going to push any statistical software too hard!

        Like

  2. Haha, the only reason that SPSS was of any particular importance to me is because it’s the only stats software that I already know how to use. 🙂

    I’ll take a look at PSPP and see how it goes.

    Looking at the data itself, in both instances 10 of the top 16 teams by either metric made the playoffs. We know there’s nothing to chase there for this particular year. However, 3 of the top 10 corsi teams didn’t make the playoffs vs only 2 by Ryan Corsi.

    The first thing I’ll do (If I can figure out PSPP) is see if the ryancorsi metric is better at predicting the conference final teams. It’s still just a one-off, so the next step would be to see if it’s repeatable for other years.

    Like

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s