Individual Point Percentage Part Two
On the weekend I wrote a little bit about individual point percentage (points per goal). Individual point percentage is a describes the number of times an individual player gets a point (either a goal or an assist) relative to the number of total goals scored while he's on the ice. For example, if a player is on the ice for fifty goals-for at even strength over the course of the season and he gets a point on twenty of them, his individual point percentage would be 40%. I first read about the concept in an article by Tyler Dellow, which, if you didn't read already on the weekend, I highly recommend. At any rate, my discussion on the weekend was limited to forwards. After the jump, we'll take a look at defenders.
As with the forwards, I decided to limit my analysis to all defenders who played a minimum of 50 games over the last three seasons. As you might expect, there aren't as many defenders who make the criteria as there are forwards. As with the forwards a lot of the people you'd expect near the top of the list are in fact near the top of the list. However, the variation from year-to-year seems to me to be greater with defenders than it is with forwards which means, in general, defenders would seem to have less influence on the offense (what insight!). This is also borne out by the average individual points percentage for a defender which clocks in around 29% compared to 70% for forwards. Any time a player deviated substantially from his three-year average it's marked in either red (down year) or green (up year). All of the data is at EV. Here we go:
So there you have it! Some notable names that aren't on the list because of a lack of games played but who are above 35% in both qualifying seasons are Scott Niedermayer, Dan Boyle and the Long Island version of Mark Streit (he qualified all three years but his last Montreal season saw a total of over 60% because he played a lot of forward). In addition to those three, the players with a 35%+ average are some of the best offensive blueliners in the game... and Steve Montador. I've checked it and I don't think I screwed up. At any rate, these players, if nothing else, are quite involved in the offence at EV. And down at the bottom of the list we find players who are, unsurprisingly, not really involved in the offence.
As an Oiler fan it's interesting to see Denis Grebeshkov ranking so well, up in that 35%+ category. Part of that is a monster season this year and it seems possible that he might regress somewhat but given the other players up in that range, if I had one in his mid-twenties, I sure wouldn't give up on him. Grebeshkov's replacement, Ryan Whitney, isn't on the list because he missed too many games in 2008-09 with injury but this year he's at 27.5% and in 2007-08 he clocked in at 30.1%. So... he had better be a step up from Grebeshkov defensively. The other offensive blueliner moved this season was Lubomir Visnovsky who didn't do as well as I thought he might but was up over 30% each season which is a nice accomplishment. He's certainly high enough to say that he's a big part of the offence when he's out there at EV. The third member of the Oiler blue who qualifies is Tom Gilbert who also rings in with a number well above average despite a poor number this season. I think there's a good chance we see some of that offence come back in future years. And for those wondering Sheldon Souray only qualified for one season (2008-09) but he did put up a monster number (47.3%). I think it's safe to say "offense from the blueline" was a clear organizational strength coming into this season.
I also just wanted to mention Marc Staal and the New York Rangers. Staal is having a pretty big season by the percentages. His individual shooting percentage is 8.5% (career 4.3%), his PDO number is up at 101.7 and his individual points percentage is 49%. Staal is a restricted free agent at the end of the year. The Rangers already have a lot of money committed to Redden and Rozsival on D but they've been using Staal as their shut-down guy (toughest competition and toughest zone-start ratio) even though he's only 23. He's been outshot slightly but has actually outscored (percentages) his opponents and all things considered has done very well defensively. What a time to have the guy hit the (unsustainable) jackpot on offence. Good for Staal, tough spot for the Rangers.
Finally, Strudwick is so terrible. If I recall he played some forward in there (the season that's green) and yet he's still in the bottom five for offence. And he's poor defensively. But Kevin Lowe sure does know his defenders! Have faith everyone!
0 recs |
13 comments
|
Comments
ES % vs total Pts %
Power Play ice time has an enormous impact on D scoring so some of these D are going to bounce up or down if they get added or removed from the PP unit. The PP set up is much more static and lends itself to touches by the defenseman compared to say ES play (that’s a hunch, I haven’t verified that). So if a D goes from being on the PP unit to off the PP unit, my expectation is that he would be less likely to pick up points at ES than on PP as a share of all Team Goals being scored.
All things Thrashers + stats: www.birdwatchersanonymous.com
I believe all of the above is even strength only.
Writer for The Copper & Blue and primary shareholder of Zorg Industries
"Never be ashamed of who you are" -- Jean-Baptiste Emanuel Zorg
by Bruce McCurdy on Mar 8, 2010 11:44 AM PST up reply actions
Perhaps I didn’t explain myself clearly. Up above, these are all EV stats and the percentages are only relative to goals while the player is on the ice. So, for example, let’s say the Oilers score 150 goals at EV. Now, they score 50 of those while Tom Gilbert is on the ice and let’s say Gilbert has 20 EV points. His point percentage would then be 40% (20/50). I’m not sure what point you’re making by bringing PP time into it.
by Scott Reynolds on Mar 8, 2010 11:44 AM PST up reply actions
More good work, Scott. This is really interesting. Like almost any metric it’s an imperfect lens and likely to be pretty volatile from season to season for reasons you have mentioned. I’d rather have a guy who is in on 50% of 40 goals than one who is in on 80% of 15 goals. Linemates, role within line, qualcomp, Sh%, luck, and random distribution are all going to play a role.
Grebeshkov’s numbers are sure trending in the right direction, what? Scary.
Good call also on Strudwick who did play quite a chunk of time on LW last year. Needless to say any line with Struds on it wasn’t going to score a whole lot, but when they did plumb one home it would be a little more likely to go off of 43 on its way to the net. Whereas on defence more goals are apt to be scored due to the possibility that there might actually be three competent forwards on the ice with him, and the likelihood that said forwards will create the odd goal without any involvement from the back end.
One suggestion: since the current season is only 80% over, it might be interesting to run your numbers with qualifying GP at 50/50/40 and see if anybody interesting gets added to the list. Alternately, wait until season’s end and keep the parameters. In that instance it would be interesting to see if some of the more extreme outliers tended to regress at least a little but towards the mean.
Writer for The Copper & Blue and primary shareholder of Zorg Industries
"Never be ashamed of who you are" -- Jean-Baptiste Emanuel Zorg
Like almost any metric it’s an imperfect lens
Steve Montador proves that point pretty emphatically I think. Or do I just wildly underrate Steve Montador’s offence? You are right of course that this is no silver bullet of analysis, just one more way to take a peek to see which players are driving offence and maybe flush out some guys having lucky seasons.
As for the GP criteria, I have the data for every player who has played 50 games in at least one season, so I have some idea of what I’m missing (Chris Campoli with two qualifying seasons at 40.2 and 42.3; Ryan Wilson in Colorado with 53.2 as a rookie defensive blueliner; Doughty with a brutal first year (18.8) and a beauty second (41.7); Mark Giordano with two seasons at 35+). If it’s a guy who has only played 40 games this season but doesn’t qualify in the other two, I’m not going to get too worried. I can send you the data if you’d like.
You’re also right that as the season progresses we’ll probably see some of the more extreme numbers come back to the pack (Keith, Enstrom, the aforementioned Staal) but there are some pretty extreme full seasons in there as well. I might take a look at it again in the doldrums of summer when I’ve got the full season to look at.
by Scott Reynolds on Mar 8, 2010 11:56 AM PST up reply actions
Montador didn’t get a ton of icetime in 07-08 (nine and a half mins) and his PTS/60 was a ridiculous 1.39, fourth among defencemen who played at least 20 games. His big year looks like it was just luck. I think your points percentage is great for looking at individual players in context (like the Gagner analysis) rather than putting too much weight in the league leaders. Another tool to build a more complete picture of what a player did on the ice.
Oh definitely. I’m sure Montador’s big year was lucky because, on top of the fact that it’s Steve Montador, it’t the best single season by a defender in the last three seasons. That said, he does have three seasons above average in a row now. Someone needs to get that lucky and maybe it’s him, but then again, maybe he’s not getting enough credit for his offence.
As for the leaders, I do think that the guys at the top generally have more offence than the guys at the bottom. In that sense, I think they do tell us something, though obviously you do need to add some context for this to be really meaningful just as you would with any other statistic.
by Scott Reynolds on Mar 8, 2010 2:19 PM PST up reply actions
This is a great series, Scott. I love this big picture stuff.
A posse ad esse.
The Copper & Blue|OilersNation|Hockey or Die!
Twitter: @JonathanWillis
Mail: jonathan.willis@live.ca
Gilbert
This is very interesting as far as Gilbert goes as he had two above average years on average teams and now this. We should expect a rebound next year.
Editor of The Copper & Blue, and leader of The Cult Of Hartikainen.
Gilbert’s number is really volatile. I would be shocked if we’re not seeing the low end this year so something of a rebound is in order but I don’t know what to predict in terms of extent.
by Scott Reynolds on Mar 8, 2010 3:34 PM PST up reply actions
"Offensive Presence"
The WHA actually coined the phrase “Offensive Presence” for Pts/TGF (total points / total goals scored while on ice). I recall seeing this in a 1970’s Hockey News Yearbook. The WHA was very innovative in some ways.
Wow, that’s cool. I’d never heard of that. Do you know if they have that data listed anywhere online?
by Scott Reynolds on Mar 12, 2010 11:47 PM PST up reply actions

by 
























