PDO Numbers
There's a theory out there (which I subscribe to, with certain caveats) which states that NHL players have no noticeable impact on on-ice save percentage, and a small impact on on-ice shooting percentage.
Therefore, since these things fluctuate quite a bit between players on the same team, we can chalk much of it up to luck, bounces, and other unrepeatable stuff - and these things can make us regard players as either better or worse than they really are.
Enter PDO numbers. The PDO number is a simple measure of even-strength on-ice shooting percentage and save percentage. The league average, for obvious reasons, is 100. After the jump: those numbers for the Oilers so far this season.
Forwards
- Dustin Penner: 111.8
- Andrew Cogliano: 111.0
- Zack Stortini: 108.5
- Ethan Moreau: 108.3
- Gilbert Brule: 108.1
- Sam Gagner: 106.9
- Ales Hemsky: 105.8
- TEAM AVERAGE: 105.4
- J-F Jacques: 104.7
- Patrick O'Sullivan: 103.5
- Robert Nilsson: 102.0
- Mike Comrie: 101.5
- LEAGUE AVERAGE: 100.0
- Shawn Horcoff: 96.4
- Ryan Stone: 94.0
Defence
- Ladislav Smid: 110.6
- Taylor Chorney: 109.5
- Lubomir Visnovsky: 108.8
- Jason Strudwick: 105.4
- TEAM AVERAGE: 105.4
- Steve Staios: 104.3
- Tom Gilbert: 101.6
- LEAGUE AVERAGE: 100.0
- Denis Grebeshkov: 99.6
- Sheldon Souray: 99.4
- Theo Peckham: 89.5
0 recs |
12 comments
|
Comments
Right
Specifically, it’s team Save% + team Shooting% when a given player is on the ice.
Writer for The Copper & Blue
by Bruce McCurdy on Oct 29, 2009 9:05 AM PDT up reply actions 0 recs
PDO numbers have always struck me as an oversimplified number based on oversimplified analysis. But they can still tell you something, and a team PDO in the 105 range when our players are as historically average-to-below-average as they are is trouble.
by Benjamin Massey on Oct 29, 2009 9:27 AM PDT reply actions 0 recs
Well, that’s quite the comment.
What exactly are you looking for? It’s just giving a rough idea of puck luck and an indicator of which way the regression will point as the sample size builds.
It’s never been presented any other way to my knowledge and I’m not sure how that can be termed as an oversimplification.
by RiversQ on Oct 29, 2009 9:43 AM PDT via mobile up reply actions 0 recs
Well, I’m quite a guy.
I think the line between “rough look at the puck luck” and “oversimplification” is a very, very narrow one.
by Benjamin Massey on Oct 29, 2009 9:50 AM PDT up reply actions 0 recs
More importantly you need to check a distribution over the team. How a pplayer is performing over the curve….
Also a team can have a higher PDO number than the other one. There are teams like TOR whose PDO number shouldlower than 100(right now)
by SumOil on Oct 29, 2009 10:04 AM PDT up reply actions 0 recs
The problem I think is the expectation that PDO will regress to 100. For forwards with great or poor finish, this can obviously swing the mean value upwards, and if your team plays behind a great or awful goaltender the same will happen. Even at the team level when SH% evens out more, SV% won’t as much.
I think it’s only meaningful when the points above or below 100 are significant (e.g. 5 or higher). Personally I like to look at SH% and SV% separately as I think combining them is mixing two distinct effects.
by R O on Oct 29, 2009 1:30 PM PDT up reply actions 0 recs
H1N1
So swine flu improves PDO?
Dustin Penner: 111.8
Gilbert Brule: 108.1
Ales Hemsky: 105.8
Ladislav Smid: 110.6
Lubomir Visnovsky: 108.8
Editor of The Copper & Blue, and leader of The Cult Of Hartikainen.
by Derek Zona on Oct 29, 2009 10:31 AM PDT reply actions 0 recs
Think about it. Would you lay a hard body check on a guy with some swine flu?
by Benjamin Massey on Oct 29, 2009 10:39 AM PDT up reply actions 0 recs
Sure. I just need some Echinacea :)
Editor of The Copper & Blue, and leader of The Cult Of Hartikainen.
by Derek Zona on Oct 29, 2009 11:22 AM PDT up reply actions 0 recs
No, what you need is Cold FX: straight from the hands of Don Cherry!
by Scott Reynolds on Oct 29, 2009 1:03 PM PDT up reply actions 0 recs

by 



























