View Single Post
02-26-2013, 03:26 PM
Student Of The Game
seventieslord's Avatar
Join Date: Mar 2006
Location: Regina, SK
Country: Canada
Posts: 30,940
vCash: 500
Originally Posted by BraveCanadian View Post
I happen to think that the Tin Man thing is a little overblown but I certainly wouldn't be counting on him for anything outside of the "easy" points.

And by easy I don't mean in execution (he was very talented) I mean the ones where you don't have to pay a price to get them.

He picked his spots even if we've recently tried to revise that for some reason.
So the points that you “pay the price” for are somehow worth more on the scoreboard? Can you explain that one?

The softness thing is really overblown for Turgeon, and for a lot of players. Value for toughness is already included in the point totals in a lot of cases. A guy like Neely made more room for himself on the ice, and his ability to go right through a defender certainly led to more points. But we look at his point totals and then give him bonuses for being tough and physical. Turgeon, on the other hand, if it was true that he could be taken off his game by cracking a loud fart within 10 feet of him, then he’d be expected to score fewer points when “off his game”, therefore his softness is mostly accounted for in his point totals already. But we look at his point totals and then downplay them because he was soft. Doesn’t seem right.

Originally Posted by Sturminator View Post
...which are very good, most similar to Roenick, though a bit better. Nevertheless, a far cry from the numbers you posted.
Much better when you get beyond the best 5 seasons. Turgeon was a much better producer than Roenick.

Originally Posted by Sturminator View Post
What an awful play that was. Nevertheless, when deciding who was the more valuable player, it would be hard for me to choose between Turgeon and Hunter.
I don’t have a problem seeing Hunter get selected before Turgeon in an ATD setting because he is potentially a very good 3rd liner and Turgeon an OK 2nd liner. But he was in no way better or more valuable.

Originally Posted by Sturminator View Post
Bullsh!t. Mine are the Vs2 numbers with #2 outliers eliminated when there is more than a 10% gap between #2 and #3. Your definition of outlier is evidently as soft as your second line center.

I know which system you're using. You're using seventies' ridiculous system in which Gretzky, Lemieux, their grandmothers and everyone they ever skated with are automatically removed, and then another player is removed (as the theoretical non-G/L world #1 scorer), and then the percentages are compared. This is a terrible bad seventies has hardly bothered to defend it, because he's smart enough to understand what a trainwreck it really is.

Must I always check your numbers, dreak?
That’s kinda rude, don’t you think? Talking to dreakmur as though he was posting falsehoods. When (by your understanding at the time) he was using a system that is not “false” but simply a different interpretation of the same data – which, I should note again, was not used for the purpose of boosting any specific player or group of players.

Originally Posted by monster_bertuzzi View Post
Sedin holds his own. Playmaking finishes are off the charts (especially if we're factoring in this season in a month as well).
“Playmaking finishes” should not be a thing here. Points matter so much more, and let’s face it, they already give somewhat disproportionate credit for assists while forgiving abysmal goal scorers like Henrik.

Originally Posted by TheDevilMadeMe View Post
Given the wackiness of 1989:

1. Mario Lemieux*-PIT 199
2. Wayne Gretzky*-LAK 168
3. Steve Yzerman*-DET 155
4. XXX-LAK 150
5. XXX-PIT 115

I think it's reasonable to use 5th place as the standard. If so, you have (88/115 = 77) meaning Turgeon's 52 turns to a 77. That would give you:

89, 85, 82, 80, 78, 78, 77, 77, 75, 70, 69, 67, 61
Absolutely. A couple more of those seasons look off, too. We’re really going to call Lafontaine’s 148 points the standard for 1993?

What I will still never grasp (and this doesn’t apply to the seasons we’re looking at for the most part) is why we wouldn’t remove players whose point totals were greatly inflated by Gretzky and Lemieux. Kurri and Coffey immediately stand out as players who would not have occupied the #2-4 spots without Wayne, yet we end up using them as the standard some seasons under the system you and Sturm are championing. Doesn’t seem the least bit fair, particularly to forwards who were every bit as talented as Kurri but had no hope of posting his point totals.

What I’d like to see from you or Sturm, is a season-by-season breakdown of how many players reached each threshold under your systems (say, 90, 85, 80, 75, 70, 65) – it should not take too long. If what you guys are saying makes sense then the number of players at a certain level should increase slightly over time and follow the general paradigms we hold about the increase of the talent pool size and influx of European talent since 1980. Basically, it should make some visual sense. But I suspect what we’ll see is a system where 90s players are disproportionately credited (since all years are painted with pretty much the same brush regardless of whether there were 0, 1 or 2 generational talents) and what puzzles me about this is that Sturm lambasted me a couple of drafts ago for suggesting that hockey got gradually better from the 80s to the 90s (and therefore the number of players reaching certain percentages should be expected to gradually rise) – I believe what he ultimately concluded was that all the European and American stars just barely made up for the drastic decline in Canadian stars - if this percentage system flies in the face of what he seems to firmly believe, why is he pushing it so much, and why would he not be open to tweaks that provide more “realistic” results?

Originally Posted by BraveCanadian View Post
These vs 2, 5, outliers, percentages blah blah numbers are getting annoying when everyone does their own thing with them.
Complaining at every opportunity about the ways other people evaluate players, on the other hand… not annoying at all!

Originally Posted by TheDevilMadeMe View Post
I think it's something of a convention to divide the lower number by the higher number when telling the difference between them
Since when? If someone asked me what’s the percentage difference between 150 and 165, I’d say “165 is exactly 10% greater than 150”.

seventieslord is offline