An Amicus Brief In Support of Dustin Penner, Part I -- Traditional Stats
Brief of The Copper & Blue as Amicus Ludus In Support of Dustin Penner for Playing on the First Line, Part I.
Table of Contents
1. Introduction
2. Traditional Stats
3. Power Play
4. Microstats
5. Conclusion
1. Introduction
The Edmonton Oilers failed to make the playoffs for the third straight year in 2008-2009. Craig MacTavish, an ineffective and overmatched coach for the entire season, unfairly used Dustin Penner as a scapegoat throughout the year. Because MacTavish was outcoached by virtually the entire league, Penner was made to suffer, when in fact, he should have been on the first line for the entire season, playing with Shawn Horcoff and Ales Hemsky. Three separate sets of statistics confirm the above argument.
2. Traditional stats
Historically, hockey statistics have been based on the simple statistics found in box scores: Goals, assists, points, penalty minutes, and beginning in the 1950's, Plus-Minus. In recent years there has been a movement towards more granular statistics to bring game and player analysis to a new level, similar to the revolution that baseball experienced thirty years ago. In this section, only traditional statistics are used.
Over the course of the last two seasons, Dustin Penner has played on a line with Shawn Horcoff and Ales Hemsky in 72 games. The results of those 72 games are broken out between '07-'08 and '08-'09 below:
When combined as a line, the Penner-Horcoff-Hemsky line produces just better than a goal per game. The Rangers famous GAG line sent two players to the Hall Of Fame by producing at a similar rate. Shawn Horcoff averages just under one point per game and Ales Hemsky averages just over one PPG while on a line with Dustin Penner.
At that scoring rate, the "Horpensky" line is the 3rd most productive goal scoring line in the Western Conference:
Marleau - Thornton - Setoguchi - 94
Hossa - Datsyuk - Holmstrom - 86
Penner - Horcoff - Hemsky - 86
Franzen - Zetterberg - Samuelsson - 84
Sedin - Sedin - Burrows - 81
How do these two perform without Dustin Penner on their line? Because Shawn Horcoff was injured to finish the '07-'08 season, the only available numbers are the '08-'09 season:
In '08-'09, Hemsky suffered a .5 PPG falloff and Horcoff falls by .4 PPG without Penner on the port side. From the two-year average, Hemsky was off by .36 PPG and Horcoff by .43 PPG. Over the course of a season, that amounts to 30 fewer points per season for Hemsky, and 35 fewer points per season for Horcoff.
In a broader sense, while together with Penner and Horcoff, Hemsky's scoring rate is 3rd among right wings behind Jarome Iginla and Daniel Alfredsson. Horcoff's rate is 10th among centers. Without Penner, Hemsky's rate is 15th among right wings and Horcoff's rate is not in the top 45 in the league among centers.
With Horcoff and Hemsky, Penner's season equivalency is 24-32-56, a rate that puts him 20th in goals and 21st in points among left wings for the '08-'09 season.
Dustin Penner may not be good enough to carry a line in the NHL. However, to call him a complimentary player is a slight. He's good enough that his presence allows Shawn Horcoff and Ales Hemsky to produce as world class players and gives the Edmonton Oilers a line to match any in the Western Conference.
0 recs |
15 comments
|
Comments
Good stuff, Coach. I don’t suppose your “traditional” stats include +/-? I suspect there’s a large differential there as well.
Also, is there any way to disentangle powerplay production and just show scoring at evens in those games?
At Level One, these stats support what I have seen by eye, which is that Horcoff and Hemmer have been far more productive playing with Penner than with Player X.
Writer for The Copper & Blue
by Bruce McCurdy on Jul 3, 2009 12:43 PM PDT reply actions 0 recs
Detangling
If I would detangle, that would mean microstats.
These are pure boxcars, Bruce. You’ll have to wait for part III
by Derek Zona on Jul 3, 2009 1:35 PM PDT up reply actions 0 recs
OK, Coach. I’ll look forward to it. Maybe Penner will still be on the team by then. :)
Writer for The Copper & Blue
by Bruce McCurdy on Jul 3, 2009 2:16 PM PDT up reply actions 0 recs
They sure have been. My question though, is Penner that good, or has Player X just been that bad. I mean, look at what else they have to work with. When you don’t have Penner on the left side, your options the last couple years have been Nilsson, Torres, Reddox, Sanderson, Thoreson, Cole, and for about 3 games, O’Sullivan. Not exactly dream wingers for any 1st line.
by ykmisfit on Jul 3, 2009 3:04 PM PDT up reply actions 0 recs
That was in response to this: “At Level One, these stats support what I have seen by eye, which is that Horcoff and Hemmer have been far more productive playing with Penner than with Player X.”
I thought it was going to quote that part.
by ykmisfit on Jul 3, 2009 3:13 PM PDT up reply actions 0 recs
Player X
My response, and this is not meant in a terse manner, is who cares?
You’re paying a guy first line money and when he’s on that first line, you have one of the best lines in the west. Would it be better with Zetterberg, yes, but why not “settle” for the 3rd best scoring line in the conference as your first line (especially since it outscores as we’ll see in part III) and use the extra money intended for savior Y on fixing the bottom six?
Are you unsatisfied with the performance of the line as indicated above?
by Derek Zona on Jul 3, 2009 3:19 PM PDT up reply actions 0 recs
I know what I see
and I see what I believe
The only line that Penner should be part of is the Doughnut line.
..at Donuts & Dogs – Atlanta’s Finest Donuttery!
by Mr DeBakey on Jul 3, 2009 1:57 PM PDT reply actions 0 recs
Player X
My response, and this is not meant in a terse manner, is who cares?
One person I can think of that cares is you. You did say the following in the original post, right?
How do these two perform without Dustin Penner on their line?
So you invited that discussion yourself and tackling that issue will require a more nuanced approach in order to defend Penner to all four corners of the Earth, if that indeed is your goal. (Although, unless you go all the way back to Smyth, it’s probably plagued by small sample sizes anyway.)
Anyway, I look forward to the next installment. It reads a little like a puff piece so far, but it will be interesting to see if there is some balance introduced. Personally I have no interest in an uncompromising defense of Dustin Penner because it will surely fall as far short of reality as a diatribe against the player.
Speaking for myself, I’m pretty aware of Penner’s contribution to the first line at this point. What I really want to know is why he can’t impress NHL head coaches for more than 10-15 game stretches?
It gets glossed over a lot on the Oilogosphere these days because it doesn’t fit the “MacT is crazy” narrative, but Penner does happen to be 0 for 2 with NHL coaches. Randy Carlyle was all over him in Anaheim as well, saying many of the same things that MacTavish said and he used the bench and the press conference liberally as motivating tools for Penner.
Perhaps it’s just their expectations? I think there’s more to it though. Regardless, we’re about to get another data point in Quinn, and arguably two with Renney on board, so that will help.
(One other thing – Bravo for getting 07/08 in there. Too often players get evaluated on blogs using just one season and that’s rarely enough. Getting the extra data is far more valuable than worrying about the other complicating factors being introduced by the extra seasons.)
by RiversQ on Jul 4, 2009 1:47 PM PDT reply actions 0 recs
I’m looking forward to further installments. I hope you include situational minutes.
We know Horcoff started in the defensive zone 220 more times than Penner, therefore “without” Penner. So of course his GF and boxcar production rates are going to plummet in his ice-time away from Penner because he’s starting 180 feet awat from the opponent’s net.
What would be interesting to know is how many of Penner’s 284, Horcoff’s 310 and Hemsky’s 305 attack zone draws were taken together.
It is likely that 10-83-27 were teamed strategically in the attack zone. So there boxcar production will obviously appear more Alfredsson-esque.
by slipper on Jul 4, 2009 8:21 PM PDT reply actions 0 recs
Well, I did look at this after 50 games (we discussed it at LT’s site IIRC) and I looked at it this summer also. It’s interesting stuff and it definitely makes a difference – at least with respect to looking at Horcoff and/or Hemsky without Penner. I don’t want to steal Derek’s thunder, so I’ll sit on that stuff until after he’s finished. I suspect he’ll cover it all anyway.
by RiversQ on Jul 5, 2009 4:51 AM PDT up reply actions 0 recs
Starting point
Rivers is down there trying to take the air out of Part III, but your premise is mostly correct. Consider though, that this wouldn’t be the case if there was someone else to pull the plow not named Horcoff.
by Derek Zona on Jul 5, 2009 7:10 AM PDT up reply actions 0 recs
Double Check Your Math
If your going to do a statistical analysis, you should ALWAYS double check your math. Hemmer, Penner and Horc only scored 76 goals based on your stats, not 86 goals. 21+26+29=76. I bet that puts them lower than 10th.
While not as strong as you state, it does show the improved point production with them together as linemates compared to Penner not on the line at all..
by toprightcorner on Jul 5, 2009 5:29 AM PDT reply actions 0 recs
Math
The math is based on an 82 game projection.
76/72 = 1.055
1.055 * 82 = 86.51
86 projected goals for a full season, something the new coaching staff should give a whirl.
Thanks for checking up on the math, though.
by Derek Zona on Jul 5, 2009 7:08 AM PDT up reply actions 0 recs
All the lines
Would it be possible for you to post all of the lines for the western conference that you used? Thanks in advance
by Jeffblake on Jul 15, 2009 10:12 AM PDT reply actions 0 recs

by 

























