Advanced Stats Glossary of Terms

Chris Pronger: .552 Fenwick, .537 Corsi
Braydon Coburn: .538 Fenwick, .536 Corsi
Lukas Krajicek: .471 Fenwick, .461 Corsi

I would normally give you a link, but these numbers were acquired from Vic Ferrari’s Time on Ice site. Gabe Desjardins told me he’s changing his Corsi scores to a percentage (as above) this summer and hopefully adding a Fenwick. So, until that happens, these numbers will only be found a) in the Grading the Flyers series; or b) upon request.

Well, I’m not sure on every series, but Gabe Desjardins did phenomenal work throughout the playoffs over at Behind the Net, and I know for a fact he spent a lot of time on the Canadiens (the Caps and Pens were the better team), the Sharks (specifically the Avalanche series which they dominated), and at least one on the Blackhawks (They won the first two rounds despite being only marginally better).

If you are at all interested in these advanced stats, he’s the guy I get about 80% of my data from. I can’t say enough good things about him and right now he’s doing work on the World Cup.

Maybe a bar graph? Put the team scores at the halfway point of the x axis and show each player’s Corsi and Fenwick along the y axis?

As far as the others go… Not sure.

There’s some really cool stuff out there, but I feel like it’s underappreciated. A good visualization goes a long way towards helping people make sense of numbers. There’s some interesting blogs out there on the topic, and it’s nice to see what goes above and beyond the normal graphs and charts.

I think you should be thinking about your objective before you think of a visualization. Are you trying to compare players against each other? Maybe you want to understand the distribution of numbers of a particular stat (you could put them in a histogram and point out which buckets some key players fall in).

Or perhaps you want to regress one variable against another (often goal differential, if you are trying to figure out how well a stat is correlated with goals). I did a stats post on SCH a little while back where I used some techniques to shade these graphs by probability, and also point out some of the points. I think I went overboard on this, and perhaps it’s more useful to show deviations from what you expect (the regression line) by drawing some lines from that to a point for a player or game.

There’s a whole world of possibilities out there — I’d love to see what you can come up with, because I feel like I’m struggling to figure out how to display data in a digestible manner too.

If you are graphically inclined, I would just have a go at putting the numbers on a spreadsheet and fiddle around with things that look good to you. Simplicity is generally the key for graphs.

I’ve made a few scatter plots with regressions just out of interest’s sake and they look ok. I think the main thing, especially if you want to do a fanpost or something, is that it is easily and quickly read and understood. And that is what I’m having most trouble with