Last year the top 12 centres accumulated 957 pts, for an average of 79.75.
Using the old 4.0 stat we get an average goaltender value of 112 (I think obviously out of whack).
Using the 3.75 scenario we get an average Top 12 goaltender of 95.00.
Taking the average top 12 goaltenders from last year, and using 3.5 the average scoring is 78.06.
Which means that on average your first line centre and your starting goaltender would be expected to produce the same amount of points.
I'd say there is a case for making sure goaltenders are equivalent to the highest scoring position in the league, but by what logic do those who favour 3.75 argue that we should value goaltenders an average of 16 pts more than a first line centre?
Subscribe to:
Post Comments (Atom)
13 comments:
Vaguely reading all this.
Two thoughts come to mind:
1) Cam hosed me last year by trading my one phenomenal goalie for two good Ds. And won. Could he have done it if Nabokov wasn't such a formula-based point machine? Is Nabokov really the best asset in the NHL right now?
2) I don't like change to the rules. It makes it hard to predict how to draft. That said, I'd prefer dramatic change to pussy-footing gradual change.
3) I'm with Chris. The problem isn't goalies are over-pointing--the problem is a few goalies are over-pointing. Reducing the common denominator isn't really going to fix that problem, just hide it. Changing the number of goalies you have to play will fix the problem. (But isn't up for discussion, clearly)
All said, and considered, I'm going to back Cam on this one.
3.75 is a really fishy number, considering I can barely calculate the stats as it is, and seems unfairly biased to those who have entrenched goalie FPs (a lot of whom turn up on that top 12 list I notice). Either we leave it at 4, and my next FP will be a goalie, or we take it down to 3.5 (or 3) and next FP will be a forward. 3.75 is just a case of wanting to have and eat cake, and I won't back it.
Richard.
Lost Boys GM.
(Or, or, someone could sway my vote by giving me my next FP--then I'll vote however you want.)
Hi Richard,
Welcome to the party.
The big thing here for the DC - from Richard's comments - is that I read this as him viewing a .25pt move as a fundamental change. DC, please note this! Whatever I may think of the optimal adjustment - if we have multiple GMs indicating that a .25pt adjustment is a fundamental change to the calculation then I think it should be treated as one and you should limit your range of options, for at least this year, to the .5pt adjustment.
As for Cam's numbers comparison, here is what I have based on last year's stats:
LW - top 1st liner 112
- ave 1st liner 77
- low 1st liner 65
C - top 1st liner 106
- ave 1st liner 87
- low 1st liner 77
RW - top 1st liner 98
- ave 1st liner 79
- low 1st liner 66
D - top 1st liner 70
- ave 1st liner 48
- low 1st liner 36
G 4.0 - top 141
- ave 109
- low 89
G 3.75 - top 122
- ave 92
- low 75
G 3.5 - top 103
- ave 76
- low 62
Cam and I seem to be a little off on our numbers (I'm using my spreadsheet and not including any PIMs in my calculations but really they are not too far off).
The 4.0 mark is a strawman, I don't think anyone is advocating staying at 4.0.
The 3.5 mark gives you a top-end goalie comprable to to a top-end center, an average goalie comprable to an average winger and a bottom goalie (#12) below the bottom 1st line wingers some 15pts below the bottom 1st line center.
The 3.75 mark gives you a top-end goalie with more points than at any other single position (10 more than LW and 16 more than C). The average goalie is, however, is just ahead of the average 1st line center and the 12th goalie is below the 12th best center.
I tend to think the 3.75 mark is closer to the valuation we give to goalies not simply because of these absolute numbers but, as I went on about at great length in the last post, because of the relative value of top goalies vis-a-vis the relative value of top players at other positions.
The fact remains, and both Chris and Richard have made note of it, that there are very few goalies (less than 4) who are really valuable both in terms of absolute pts and relative value. Any change to address that, if indeed we feel it is even desirable, will require far more tinkering than a .5 or .25 tweek to the current formula.
The more i read this, the more i am agreeing with Cam on the 3.5 values. Goalies are playing so many games now and goal scoring is low and getting lower every year it seems. There is no way i would value Nabokov more than Ovechkin and especially not by 20pts.
If i had the choice i would like to change the way we handle goalies period as going to 3.5 still wouldnt solve the bigger issue of disparity.
Hi Chris,
I think you have noted the real issue with goalies is the disparity in value between them.
The reason we adopted the formula we did was that it tried (clearly imperfectly) to create an individual-based calculation (as opposed to the W/L/T + SO formulas of other pools) for goalie pts that would be comprable to other players.
The only two goalie stats that make up our current calculation are GAA and MIN. SO are included in GAA but are not specifically or extraordinarily rewarded (but then again we count goals and assists equally - each valued at one quatluu). We do not make use of SV% or Shots Against. Not sure of how we would or could - or even if doing so would be any improvement on what we have now.
Once we reset things to the 3.5 standard I suspect that we won't hear much talk about goalie stats for a year or two (unless goal scoring goes significantly up or down again next season).
One thought that I have, and this would be a radical departure - which brings with it its own complications - which I think I'd like to post as a new post (since it goes way off from where we are right now).
Doug,
You're misreading me if you think I view .25 as a fundamental change--it's not, it's more in the line of "silly change" which is why I oppose it.
You are not addressing, in your argument for 3.75, what an obscure number it is. It really seems designed to just obfuscate goalie value while I'm just looking at a sports page. Drafting a goalie is already difficult for me. Telling me that every 4 complete games without over time that a goalie plays is now worth 11 pts... is just plain difficult to wrestle with. (Or, for more fun: Overtime, now worth 3.75 x 1.33 pts!) I prefer Mike's system to that kind of stats calculation.
I have to wonder, is this number a soft sell to the GMs who have goalie FPs instead of a number that would be easy to play with? I say give them the hard number that we want (3.5) and let them make their own case back.
It's no good addressing the mean value in such a slim pool of goalies, and, when the rules let you get away with playing a single goalie, all year long--then everyone stuck with "the median goalie" is basically being told to suck it.
Cam would not have won the Pred if he didn't have Nabokov, and unless the baseline is 3.5 or lower, he'd be a fool to draft anything but a top end goalie with the first pick.
And while I realise there's a lot to be said for arguing for the theoretical purity of the stats, I'm in the pool to win, and opposed to obfuscation: Either make goalies clearly worth a lot, and leave it at 4, or get them off the map so I can focus on forwards with my first few picks.
If you want the average goalie score to mean more, why not add another goalie to the roster (do we really need 8 Ds?) and have everyone play 2 each week? Scores would go up, but that "average" would begin to bog everyone down a lot more, and make having two "B" level goalies not such a bad situation. (Especially if the baseline went down to that easy to convert number of "3")
Finally, the pool is meant to represent that ideal of drafting from the pool of NHL players--if you're arguing 3.75, with the evidence you are presenting, then you are making the case, that had a 31st team joined the NHL, and told to pick their team from all existing players--no contracts or salary caps to hold them back--that the best possible draft that team could have made last year, would have been Nabokov.
I disagree rather profoundly.
But, if you do feel that way about Evgeni, I think you should volunteer to trade Crosby for Nabokov--because, N. is almost certain to be drafted in the first round, and you could promote Crosby to FP if necessary. And, no matter how this debate is resolved, your money should be on the Sharks have the best player in the NHL, right?
Richard.
Doug,
You're misreading me if you think I view .25 as a fundamental change--it's not, it's more in the line of "silly change" which is why I oppose it.
You are not addressing, in your argument for 3.75, what an obscure number it is. It really seems designed to just obfuscate goalie value while I'm just looking at a sports page. Drafting a goalie is already difficult for me. Telling me that every 4 complete games without over time that a goalie plays is now worth 11 pts... is just plain difficult to wrestle with. (Or, for more fun: Overtime, now worth 3.75 x 1.33 pts!) I prefer Mike's system to that kind of stats calculation.
I have to wonder, is this number a soft sell to the GMs who have goalie FPs instead of a number that would be easy to play with? I say give them the hard number that we want (3.5) and let them make their own case back.
It's no good addressing the mean value in such a slim pool of goalies, and, when the rules let you get away with playing a single goalie, all year long--then everyone stuck with "the median goalie" is basically being told to suck it.
Cam would not have won the Pred if he didn't have Nabokov, and unless the baseline is 3.5 or lower, he'd be a fool to draft anything but a top end goalie with the first pick.
And while I realise there's a lot to be said for arguing for the theoretical purity of the stats, I'm in the pool to win, and opposed to obfuscation: Either make goalies clearly worth a lot, and leave it at 4, or get them off the map so I can focus on forwards with my first few picks.
If you want the average goalie score to mean more, why not add another goalie to the roster (do we really need 8 Ds?) and have everyone play 2 each week? Scores would go up, but that "average" would begin to bog everyone down a lot more, and make having two "B" level goalies not such a bad situation. (Especially if the baseline went down to that easy to convert number of "3")
Finally, the pool is meant to represent that ideal of drafting from the pool of NHL players--if you're arguing 3.75, with the evidence you are presenting, then you are making the case, that had a 31st team joined the NHL, and told to pick their team from all existing players--no contracts or salary caps to hold them back--that the best possible draft that team could have made last year, would have been Nabokov.
I disagree rather profoundly.
But, if you do feel that way about Evgeni, I think you should volunteer to trade Crosby for Nabokov--because, N. is almost certain to be drafted in the first round, and you could promote Crosby to FP if necessary. And, no matter how this debate is resolved, your money should be on the Sharks have the best player in the NHL, right?
Richard.
HighlandeR: The fact remains, and both Chris and Richard have made note of it, that there are very few goalies (less than 4) who are really valuable both in terms of absolute pts and relative value. Any change to address that, if indeed we feel it is even desirable, will require far more tinkering than a .5 or .25 tweek to the current formula.
Cameron: I know this will astonish Americans who might read this, but did you know that 50% of them are of below average intelligence?
As it stands we have 12 starting goaltenders to dress every week. Six of them will be below average.
Now, add on top of that that out of those twelve we already have Brodeur, Lundqvist, Turco and Luongo creamed off the top as FPs.
Which leaves us with two above average goaltenders to whore our 1st rounders on.
Frankly so long as the math on goaltenders is similar to 1st line centers I don't have a problem with this scenario. I note that Darrell has had a number of close finishes by selecting late round goaltending that unexpectedly paid off. He chose to invest his early picks elsewhere hoping he could hit a home run by getting superior goaltending in the late rounds.
But its a scenario that makes the most sense where goaltenders aren't the runaway best position.
In that case, its get the best guys or suck it.
The more I think about it the more I like the notion that goaltenders could be pegged to either the number of goals scored in the league every year (using half a goal as the increment), or to the average performance of top line centers.
If we have a new golden era of offense the centers (Crosby, Malkin, Thornton, etc.) are going to be part of it, and who better for elite goaltenders to tag themselves to?
In any case all that is happening is a market correction, the kind for which we have a built in regulation for managing - we drop a half a point per game.
This will not solve the problem of half the goaltenders in the league being below average. Nothing can solve that problem.
But it will mean that having below average goaltending doesn't equal certain death.
As a member of the DC, I like getting an idea of what different GMs think who are not so active on the blog, at least on this subject.
Both Collin and Corey in recent conversations have given me permission to put their views on the blog.
Collin had the opinion that rules in the covenant implied only goal stat changes of 0.5 for the DC; but more importantly thought 3.75 was an odd number and was much more comfortable with 3.5.
Corey, who happens to have a FP goaltender in Turco, expressed a similar sentiment about 3.75 as being odd and favoured 3.5. It's an easier number to work with.
Cam, Chris, Richard, Collin, Corey is 5 which basically makes me 6.
However Rule proposals for next year can certainly be proposed on another post.
I seem to remember that cell phones have a calculator feature on them so you can figure out what ever number whereever you are if you really need to know how many points your goalie got at the bar or with the morning paper.
As I am probably the only one who does not have a cell phone, I am willing to live with this inconvienence.
For the rest of the time you are probably using spreadsheets of some form or another to calculate masses of stats so whether it is xW+yOTL+zS or (4-GAA)*min/60 it is still a formula.
With respect Rob, the calculation goes like this;
- Hey Bob how you doing tonight?
- Crappy. Backstrom just let in his fourth.
- Oh Noes! Sucks to be you.
- Yeah, but Nabokov has only let the one in so far I see.
- Yep, which means 2 and half for me!
- Check that, he just let one in from center ice.
- Oh Noes! Well, 1.5 is still better than losing points.
- Indeed it is.
You'll note that no calculartor or cell phone was ever whipped out during this conversation.
I still don't think that 12 intelligent GMs would even need a calculator during such a conversation with a 3.75 base:
l against = 2.75
2 goals against = 1.75
3 goals against = .75
4 goals against - -.25
If the goalie plays overtime or only 59 minutes or whatever, then a calculator is needed anyway (or you just rely on the statistician to take care of the differential).
That all said - I am not firmly opposed to 3.5 but I think that 3.5makes a mockery of goaltending after the top ones. I'd rather have several goalies over-valued at 4 then several plus goalies undervalued at 3.5.
Brian said: That all said - I am not firmly opposed to 3.5 but I think that 3.5makes a mockery of goaltending after the top ones.
Cameron: I hear several people complain that goaltending 'after the top ones' is mediocre.
Of course it is! Just as is the case with first line centres, once you take out the FPs (and prospects who will be FPs) you are left with a very short list of above average 1st line centres you can draft; Lecavalier -Datsyuk. End of list.
The reason everyone fixates on this perceived lack of goaltending depth is because we only play one goaltender.
But honestly I don't see the problem. Unlike other cup winning teams the Severed Heads never had a Patrick Roy or Dominik Hasek to brings us a victory. We used Ed Belfour, Olaf Kolzig and Evgeni Nabokov instead.
Having below average goaltending is something most of the league deal with every year, and it means that all things being equal, we will advantages at other positions that those goaltending heavy GMs don't.
Such is the nature of things.
I find it hard for you to say that Nabokov did not help you with your win for last season. He out performed Broduer in the last couple of weeks of the season to help your victory.
Post a Comment