The only thing that I don't like about the fancy stats is what they pretend to predict. As a means of looking back, I think they are fine.
Corsi and Fenwick pretend to speak to possession, but they really only measure shot attempt differential. Shots do not always equal possession.
PDO pretends to measure "puck luck." But a person that chooses poor shots (low shooting percentage) and allows good shots (low save percentage) will appear to be unlucky.
I was going to do some big post on PDO a little while ago, but I got bored partway through and lost interest. I did, however, make this chart before I stopped.
I took the 2013 and 2014 PDOs of 120 players who played more than 3/4 of the games in each year and played more than 14 minutes or 16 minutes a night at even strength, for forwards and defensemen, respectively. If PDO was repeatable, you'd expect a correlation between the '13 and '14 numbers, but there clearly isn't one. No players were more than 20 points below average in both years, and only four were more than 20 points above average in both years.
Also, shots pretty much equal possession.
There was a Leafs fan who went over half a season of Leafs games with a stopwatch tallying up time on attack to see if it correlated to corsi/fenwick, and he found it was a nearly perfect correlation.
One thing on the fancies is that it does not show bad chemistry.
Nick and MJ defer shooting to Ovi. So they don't score because they don't take quality shots, the entire line one dimensional. Get the puck to Ovi. The shots they take, the goalie is greening with Juan in the crease and makes a crester save with ease time and time again.
Won't the fancies say they are just having horrible horrible bad luck, and that Ovi should really be +10 versus -30 because of it?
Largely based on projecting previous years numbers of theirs under a run and gun system, into the modern day in which those players seem completely unmotivated.
My numbers are a week old, but he'd actually be a -12 (19 point increase) if he had a 1000 PDO. Backstrom would be -8 (12 point increase) and Green would be even (14 point increase). I was only looking closely at the top 10 unluckiest forwards and defensemen, so I don't have data on any other Caps.
