The authors of this post support a minor Keeper League rule change that would give goalies one point for overtime and shootout losses, similar to the way NHL teams are given the "loser point." Before accusing us of being out to lunch, or insisting that sometimes goalies are simply shit out of luck, we encourage you to read our discussion below.
In addition to being closer to NHL rules, this change would also have the advantage of reducing the effect of chance. Shootouts are essentially random and the difference between zero points and two points [or occasionally five points if it's a 0-0 game] for your team should never be a Mason Raymond spin-o-rama).*
One concern that comes to mind is that loser points would result in slightly increased point totals for the goalie, who is already typically the highest scorer on about half the teams in our format. However, it will not be a lot of points. In 2011-12, the goalie with the most OTL/SOLs would have gained an additional 13 points. On average we would expect a five-to-10 point bump. Moreover, as Greg pointed out in a previous post, he and Eric are sitting pretty at the top despite having few points from their goaltenders. In fact we would argue that our league significantly undervalues goaltenders at just one slot out of 14.
On that note, we have three other goalie suggestions for consideration:
1) We should consider having two goaltender slots on the scoring roster, thus expanding the scoring roster to 15 and the in-season roster to 23, but retaining the keeper roster at 10+2.
2) Goalie goals should be worth more than one point. We propose somewhere in the neighbourhood of ten points. The goalie goal is so rare that it should be truly exciting for the GM who owns that goalie, although it shouldn't make or break a season.
3) A statistic which reflects save percentage and adds points for every x decimal points a goalie is above some save percentage value (perhaps .915 which Kelly Hrudey remarked on HNIC last night was the league average) per y amount of shots (perhaps 100). This would be similar to the statistic that Fy came up with for his Auction Keeper League, and one we were particularly reminded of when we saw that Ben Bishop stopped 41 of 42 shots in his 1-0 OT loss to LA on Friday night. Ben Scrivens followed that up with a 32 for 33 night for the same outcome last night against the Avalanche. These great goaltending performances, we submit, should be rewarded in some form. If there is interest in learning more about this stat, we can provide some sample calculations for some various possible values.
3) A statistic which reflects save percentage and adds points for every x decimal points a goalie is above some save percentage value (perhaps .915 which Kelly Hrudey remarked on HNIC last night was the league average) per y amount of shots (perhaps 100). This would be similar to the statistic that Fy came up with for his Auction Keeper League, and one we were particularly reminded of when we saw that Ben Bishop stopped 41 of 42 shots in his 1-0 OT loss to LA on Friday night. Ben Scrivens followed that up with a 32 for 33 night for the same outcome last night against the Avalanche. These great goaltending performances, we submit, should be rewarded in some form. If there is interest in learning more about this stat, we can provide some sample calculations for some various possible values.
For reference, below is a list of KL goalies and their "loser points" as of November 24, 2013.
G-Phil's Fylers
Halak - 2
Rinne - 1
Powder Rangers
Anderson - 2
Anderson - 2
Harding - 2
Mackhawks
Quick - 0
Backstrom - 2
Milan Micahleks
Rask - 2
Fylanders
Fleury - 0
Thomas - 1
Winter Claassens
Hiller - 2
Bobrovsky - 2
Dicklas Lidstroms
Lundqvist - 0
Ramo - 1
Schizzarks
Price - 2
Markkstrom - 3
Moilers
Luongo - 4
Crawford - 3
Teeyotes
Bishop - 1
Ward - 4
Joshfrey Krupuls
Lehtonen - 2
Miller - 0
Los Amjawors Kings
Niemi - 5
Nabokov - 3
Patrik Stefans
Pavelec - 3
Schneider - 3
Quebec Rordiques
Holtby - 1
Varlamov - 0
Vanrooser Canicks
Howard - 6
Bernier - 1
Wilkes-Benham/Scranton Parkers
Smith - 4
Dubnyk - 1
[Commissioner edit:
Since this may come up for vote, the "loser point" will have a minor effect on the standings. Everyone stays where they are except the:
Patrik Stefans move up to 8th from 9th
Dicklas Lidstroms move down to 9th from 8th
Joshfrey Krupuls move down to 11th from 10th
The Los Amjawors Kings move up from 13th to 14th
The Quebec Rordiques move down to 14th from 13th
Thanks - G]
This has been a joint post by the GMs of the Milan Micahleks and the Patrik Stefans. We invite your comments.
*Since the initial draft of this post, we have made the following observations regarding Mason Raymond: he has actually picked up his scoring pace, currently at the 60 pt mark adjusted for 82 games; he is an unremarkable 1 for 4 in the shoutout this year; his spin-o-rama is no longer a legal move in the shootout. Additionally, we note that a goalie does indeed get awarded for a shutout in the event of a 1-0 shootout loss, BUT NOT for an overtime loss (as was the case for Bishop and Scrivens, mentioned in the sv pct stat suggestions).
*Since the initial draft of this post, we have made the following observations regarding Mason Raymond: he has actually picked up his scoring pace, currently at the 60 pt mark adjusted for 82 games; he is an unremarkable 1 for 4 in the shoutout this year; his spin-o-rama is no longer a legal move in the shootout. Additionally, we note that a goalie does indeed get awarded for a shutout in the event of a 1-0 shootout loss, BUT NOT for an overtime loss (as was the case for Bishop and Scrivens, mentioned in the sv pct stat suggestions).
long overdue; I fully support this.
ReplyDeleteWhen did they outlaw the spin-o-rama? (which I still think should just be called the spin move or spinning deke; it's not a carnival ride). I believe it's still allowed but reviewable to make sure the puck doesn't stop, the player maintains forward momentum and doesn't interfere with the goalie.
ReplyDeleteI am in favour of the loser point.
ReplyDeleteI'm intrigued by some of the other proposals but one of the things I love about the KL is its simplicity. I like being able to look at a box score and see how many points I got - I'm not sure moving to a "formula" for goalies is the right move. That said I'm open minded and would be curious to see how that might look.
I'm not in favour of a loser point because I feel sv% would be a much more accurate measure. A goalie who stops 40/41 and loses 1-0, whether in OT or not should have some value.
ReplyDeleteThe Auction Keeper League implements this formula:
-0.5 for each goal vs.
+.05 for each save
1 goal against equates to 10 saves. Goalies with a sv% over .909 would gain points in the long run and subsequently goalies with a sv% under .909 would lose points in the long run.
MAF + Josh Harding currently have 13 wins each = 26 points (ignoring shutouts/assists), however Harding boasts a .939 sv% to MAFs .923%.
With the suggested system Harding would have 3 points more than MAF instead of being equal accurately reflecting the difference in sv% (probably the most important goalie statistic)
While the OT point is far more simplistic, it is far less accurate of each goalie's individual value.
Despite being a co-author of this post, at this point I am not convinced that a formula for save % is necessary. That's Stefan's baby. I am also 95% certain it couldn't be done in pickuphockey and would have to be calculated separately.
ReplyDeleteI'm assuming the vote would impact next year's stats (if we get that far, and we will)… I can't imagine a rule change that retroactively changes this year's stats…
ReplyDeleteWe can measure a goalie's "worth" however we like. Sv% more accurately represents real-world goalie value, but this isn't the real world, its the Keeper League. We value goalies who will get the most wins and shutouts (and draft accordingly). Its all completely arbitrary - the KL isn't designed to be an NHL simulator.
ReplyDeleteAnd as for doing it retroactively, we can - if people vote in favour of it.
No need to get defensive Greg. I'm just pointing out why I wouldn't vote in favour of the OTL/SO point if there was a better option such as including sv% as a stat. Unfortunately it sounds like pickuphockey as a medium won't allow this.
ReplyDeleteAll pools in some way are designed to be an NHL simulator. Crosby is the most valued player in the KL because he gets the most points, similar to the NHL.
The KL goaltending stats are by far the most arbitrary, inaccurate stat we have to measure the actual worth of a player. I'm in favour of tweaking the stat and presented sv % as what I felt as the best way to do that. I'm sure we'll do the OT/SOL point instead. Just expressing an opinion which is what I thought the thread was for
I sincerely appreciate all the commentary, so by no means is this meant as any thoughts on any of the suggestions as I think being reflective of parity within the system we choose is a really admirable goal.
ReplyDeleteAll that said, while I know that pickuphockey doesn't allow for super sophisticated point evaluations, I really really like their format. I'd be open to any of the suggestions from Fy or the Stefamicahs if we can keep it within the pick up hockey frame.
I guess i'm basically advocating for all of the above ideas, as long as I can keep my nostalgia and arbitrary attachment.
Fy, I'm not in the least bit defensive (well, I am now, but only because you're seeing a battle of wills where I just see discussion). I'm just pointing out the other side of the same coin. I'm all in favour of opinions and options.
ReplyDeleteMy point is merely that "more accurate" is not necessarily the same thing as "better". I think "more enjoyable" is the benchmark we should strive for. If that means Sv% is more fun, or if people find SOL/OTL points more fun, or whatever else we can come up with, then that is better for the league.
I gave (and give) no value judgment on either option.
I see the merits in a SVP system, as it does most closely approximate real NHL action and is undoubtedly a better stat for determining the worth of a goalie in the real world. That said, I do not see the need to change our system. Whether W/L is "better" or "worse" than a more complex SVP system better approximates which goalie I want on a real team is sort of beside the point to me.
ReplyDeleteThe way it works now is that the goalie pick is, for the most part, a proxy bet on which team is going to do best (subject to starting goalie uncertainty). At least in terms of scoring, we have striven to keep the KL as simple as possible (no difference between G and A, no PIMs and other categories, etc.). Theoretically we could also add player possession or other advanced stats to measure the true worth of skaters (I JUST GAVE MICAH AN ERECTION) but we have to find balance between simplicity and accuracy. Fy I think you're right that W/L is a pretty big compromise on accuracy in favour of simplicity, but I'm ok with it.
Stefan, I do not think that the scoring roster should have a second goalie. There are simply too many people in the league for this in my view when you consider injuries.
I frankly don't care one way or the other whether we incorporate a loser point like the NHL does.
What if you had two goalies but at the start of the year had to designate one as a back-up, and his points were roughly halved (say one for a win and two for a shutout)?
ReplyDeleteI would like to see more goalies involved because look at all the young goalies, back-ups, and legendary-but-aging veterans with interesting storylines that our league renders irrelevant: Brodeur is not going away without a fight; Mason is reviving his career in Philly; Scrivens and Reimer are looking better than the starters they are backing up; Andersen looked good in relief. I want to get involved in all of that.
Save percentage formulas are a terrible idea - way too complicated, and not actually a better evaluator of worth in terms of calculating standings, because only wins matter in the NHL, not how you get them. No one cares if you win 7-5 with a .730 save percentage or if you win 1-0 with a 1.000 save percentage, just that you win.
ReplyDeleteI am totally indifferent on the loser point issue.
First off, Fy... this has to be said... as a neutral observer, Greg hasn't been the slightest bit defensive. Stop being ridiculous.
ReplyDeleteSecond, and this is to Fy's substantive point with which I agree, there is certainly an aspect to the goalie stat which is unsatisfactory, whether it's appropriately referred to as arbitrariness or something else. The reality is, and this is the extent to which the blog was inspired, there is something that "just doesn't feel right" about the way points are awarded, especially when Bishop and Scrivens put up nights like they did and get unrewarded. We are proposing SOME way to address this and reward goalies where reward is merited. I'm not necessarily convinced that the out to lunch/shit out of luck single point awardance is the best option, but it certainly works it some cases.
If pickuphockey can't accommodate an advanced stat like sv pct, then this whole discussion is either a complete non-starter OR we could actually talk about changing hosts. We've discussed this in the past considering the limitations around trades and specifically the requirement of going slot for slot (for what I believe, this is the single biggest problem with the KL right now), so if we had further reason to change host, it might be worthwhile.
In the end, I do not wish to transform this into anything much more complicated. There is beauty in the current simplicity of our format. But I believe there is still room for improvement. Hockey is about winning and losing, but to get there it's about scoring more goals than the other team. We reward players for their contributions toward scoring goals, why not reward goalies for their contributions toward preventing them?
Lastly, Schizz, sober up.
"We can measure a goalies worth however we want" - I took that as defensive given that I never said that he couldn't… What's ridiculous is you actually think your team is good ;)
ReplyDeleteThis comment has been removed by the author.
ReplyDeleteOnly kidding Stef (you are only 5 points behind me after all)… I'm all for change but I must say I do love the simplicity and the ease of our medium (this after dealing with complexities on Yahoo)…
ReplyDeletePowder hit the nail on the head. Goalies are more of a team stat in our pool because of wins.
As for your point Greg I agree we should do what's more fun. 10 points for a goalie goal for instance is a lot of fun but not accurate at all and something I'd be in favour of.
Ok, sorry for being glib but i, m blasted after wu tang clan
ReplyDelete1) 16 teams in KL and not enough starting goalies in the league. I think two goalies would be create disparity
2) the fun part of me says yeah fuck it but then I think woah that's gimmicky
3) really really really against save percentage. If we do this, we might as well do CORSI and then we might as well all go jump in a volcano #fancystats
Ok, sorry for being glib but i, m blasted after wu tang clan
ReplyDelete1) 16 teams in KL and not enough starting goalies in the league. I think two goalies would be create disparity
2) the fun part of me says yeah fuck it but then I think woah that's gimmicky
3) really really really against save percentage. If we do this, we might as well do CORSI and then we might as well all go jump in a volcano #fancystats
The fun part of Dickie should make all the rules.
ReplyDeleteDrunk Steve Dickie for President… or maybe the Mayor of Toronto
ReplyDeleteIf we all had two goalie slots, somebody would be considering picking up Brett Leonhardt right now. http://www.rotoworld.com/player/nhl/2909/brett-leonhardt
ReplyDelete