HFBoards

Go Back   HFBoards > General Hockey Discussion > The History of Hockey
Mobile Hockey's Future Become a Sponsor Site Rules Support Forum vBookie
Notices

The History of Hockey Relive great moments in hockey history and discuss how the game has changed over time.

Tom Paton, Olde Tyme Game Accounts, and How To Evaluate Pre-Cup Goaltenders

Reply
 
Thread Tools
Old
09-04-2012, 07:08 PM
  #51
Iain Fyffe
Hockey fact-checker
 
Iain Fyffe's Avatar
 
Join Date: Feb 2009
Location: Fredericton, NB
Country: Canada
Posts: 3,509
vCash: 500
Quote:
Originally Posted by tarheelhockey View Post
... but page long narratives that specifically describe the gameplay on a rush-by-rush basis are irrelevant because the writer might have been on the crapper during an important play!? Are you kidding?
The game play is not described on a rush-by-rush basis in the summary you provided. Parts of it are, certainly, but there are sections of the game that are glossed over. You can assume that there were no important plays, not even any mundane shots taken in that time, but that's an assumption. It may or may not be true, and we don't know. Any suggestion that there is nothing worth noting in that time is just a suggestion, and can be nothing more unless we have some sort of corroborative evidence.

Quote:
Originally Posted by tarheelhockey View Post
A whole lot of people have done a whole lot more research than you on a whole lot of subjects, and been dead wrong. Perhaps you should just do the graceful thing and acknowledge that your analysis of Paton is heavily team-biased, and that his teammates CLEARLY had a larger role in MAAA's early success than he did.
Writing something in caps does not make it true. There's a possibility that Paton deserves less credit than I currently give him. There's always uncertainty, and it increases as we go further back in time. But perhaps you should do the gracious thing and acknowledge that while your assertion may be true, and does have some circumstantial evidence to it, saying it must be true, that it cannot be incorrect, is unsupportable.

And of course, my analysis of Allan Cameron and James Stewart is also "heavily team-biased", because I start with the team results. Someone has to get credit for their goal prevention, though.

Quote:
Originally Posted by tarheelhockey View Post
If you can't do that, what good were all those hours of research just to draw tainted conclusions?
I think you have this backward. I'm basing my conclusions on the research I've conducted. You seem to be arriving at a conclusion based on a few things you've read, and then each new thing you read after that is interpreted through the filter of that conclusion.

Quote:
Originally Posted by tarheelhockey View Post
Edit: out of curiosity, where exactly did you spend countless hours reading about Paton? These summaries are easily available online and it only takes a little while to read a whole season's worth of them.
They haven't been online for that long. We used to have to do it the hard way, in the library with microfilm. Still do, for the papers that aren't online, which are many. There was no newspaperarchive.com when I was compiling Dan Bain's career stats. That was done at a microfilm machine.

Iain Fyffe is offline   Reply With Quote
Old
09-04-2012, 07:14 PM
  #52
Iain Fyffe
Hockey fact-checker
 
Iain Fyffe's Avatar
 
Join Date: Feb 2009
Location: Fredericton, NB
Country: Canada
Posts: 3,509
vCash: 500
Quote:
Originally Posted by tarheelhockey View Post
3-line game "summaries" are strictly accurate for recording scores, and their data can be crunched and twisted backward to "prove" that one player deserves a certain amount of credit for his team results...
You've got this exactly backwards, of course. The numerical analysis I did suggests that Paton was this good. It's an analysis that's been applied to all senior-level hockey from 1884 to well into the 20th century. To suggest it was designed to big up Paton is completely baseless.

Here you're essentially accusing me of intellectual dishonesty. Please stop. You're hurting any case you might have by resorting to this sort of thing.

Iain Fyffe is offline   Reply With Quote
Old
09-04-2012, 07:55 PM
  #53
tarheelhockey
Global Moderator
 
tarheelhockey's Avatar
 
Join Date: Feb 2010
Location: The Triangle
Country: United States
Posts: 53,909
vCash: 500
Quote:
Originally Posted by Iain Fyffe View Post
The game play is not described on a rush-by-rush basis in the summary you provided. Parts of it are, certainly, but there are sections of the game that are glossed over. You can assume that there were no important plays, not even any mundane shots taken in that time, but that's an assumption. It may or may not be true, and we don't know. Any suggestion that there is nothing worth noting in that time is just a suggestion, and can be nothing more unless we have some sort of corroborative evidence.
Those parts aren't completely glossed over -- they're described, briefly, as periods of time when the gameplay didn't result in significant offensive results. "Scrimmages" and "temporary advantages" imply game action that doesn't result in the ultimate goal of a shot on net.

You're talking this lawyerly approach of "anything could have happened, and you can't prove it didn't!" but some simple use of context clues makes it abundantly clear that there were periods in the game where the offense petered out, and the writer generalized those periods to save column space. I'm not sure what more "corroborating evidence" you expect on this point.


Quote:
Writing something in caps does not make it true. There's a possibility that Paton deserves less credit than I currently give him. There's always uncertainty, and it increases as we go further back in time. But perhaps you should do the gracious thing and acknowledge that while your assertion may be true, and does have some circumstantial evidence to it, saying it must be true, that it cannot be incorrect, is unsupportable.
Ever since the beginning of the conversation, before you established this ludicrous standard of absolute certainty, I've done nothing but call for as much information as possible to get to the bottom of things. At no point have I said that my reading "must" be true to the exclusion of all other possibility. This isn't a death-penalty case. I'm willing to take all the information we can collectively bring to the table, and apply a simple common-sense standard to what we have to work with. If that information is in fact incomplete, then we take the partial picture and do the best we can.

But I reject the idea that we should ignore anything that falls short of total proof, and pretend that the partial picture doesn't exist. If a writer says a goalie didn't face a single shot for an entire half, and doesn't bother to write about the other half, and the goalie's team wins 8-0, it is positively stupid to pretend that the goalie deserves equal credit for the win as the guys who shredded the opponent at both ends of the ice. If a writer repeatedly singles out the point man for praise, consistently throughout the season, and rarely even mentions that the goaltender had a hand in the win, it's pretty darn misguided to say "well, maybe he was looking at a pretty lady when the goalie saved the game".

Quote:
And of course, my analysis of Allan Cameron and James Stewart is also "heavily team-biased", because I start with the team results. Someone has to get credit for their goal prevention, though.
All that I've seen suggests Stewart and especially Cameron deserve the lion's share. And there's nothing wrong with that.

Quote:
Originally Posted by Iain Fyffe View Post
You've got this exactly backwards, of course. The numerical analysis I did suggests that Paton was this good. It's an analysis that's been applied to all senior-level hockey from 1884 to well into the 20th century. To suggest it was designed to big up Paton is completely baseless.
I'm not saying anything about the motivation behind it. I'm just saying that it conflicts with primary-source evidence. Not the first time someone has made a system that doesn't produce perfect results, is it? There's no shame in that.

Quote:
Here you're essentially accusing me of intellectual dishonesty. Please stop. You're hurting any case you might have by resorting to this sort of thing.
I'm going to let the mods handle that.

tarheelhockey is offline   Reply With Quote
Old
09-04-2012, 08:03 PM
  #54
Bear of Bad News
Mod Supervisor
HFBoards Escape Goat
 
Bear of Bad News's Avatar
 
Join Date: Sep 2005
Location: Windsor
Posts: 5,542
vCash: 500
Quote:
Originally Posted by tarheelhockey View Post
I'm going to let the mods handle that.
Guys, please stop making it personal. I'm not going to bring it up again.

Bear of Bad News is offline   Reply With Quote
Old
09-04-2012, 09:09 PM
  #55
Iain Fyffe
Hockey fact-checker
 
Iain Fyffe's Avatar
 
Join Date: Feb 2009
Location: Fredericton, NB
Country: Canada
Posts: 3,509
vCash: 500
Quote:
Originally Posted by tarheelhockey View Post
Those parts aren't completely glossed over -- they're described, briefly, as periods of time when the gameplay didn't result in significant offensive results. "Scrimmages" and "temporary advantages" imply game action that doesn't result in the ultimate goal of a shot on net.
No they don't. Did you see my earlier post about how 5 of 31 described goals were scored from scrimmages in 1893? You haven't addressed any of the evidence I presented, really.

Quote:
Originally Posted by tarheelhockey View Post
You're talking this lawyerly approach of "anything could have happened, and you can't prove it didn't!" but some simple use of context clues makes it abundantly clear that there were periods in the game where the offense petered out, and the writer generalized those periods to save column space. I'm not sure what more "corroborating evidence" you expect on this point.
The whole point is that I don't believe that there can be any corroboration, leaving your interpretation as guesswork. Plausible guesswork, certainly, but guesswork nonetheless.

There are differing standards of evidence required here:

1. "I believe hockey in 1890 probably featured essentially a similar number of shots on goal as hockey a few years later, whatever number that may be."

2. "I believe hockey in 1890 featured significantly fewer shots on goal as hockey a few years later. In fact, I think in 1890 it was about 10 per game."

The second requires much more evidence than the first. You'd not only have to substantiate the specific claim of the 10 shots per game, but also that it was substantially different than a few years later.

Quote:
Originally Posted by tarheelhockey View Post
I've done nothing but call for as much information as possible to get to the bottom of things.
And yet you don't address the information I've presented. When I do point out that there were apparently many more shots taken in 1893 than you believe were taken in say 1888, instead of taking that as evidence that you may be low on your shot estimates (considering that you're relying on incomplete game reports), you take it as evidence that the shots per game increased significantly in a very short period of time. That interpretation of the information is based on the conclusion you've already made.

When new information is viewed through the filter of an existing conclusion, rather than reviewing the conclusion to see if it holds up to the new information, there's a problem.

Quote:
Originally Posted by tarheelhockey View Post
But I reject the idea that we should ignore anything that falls short of total proof, and pretend that the partial picture doesn't exist.
That's not the idea; the idea is that extraordinary claims require extraordinary evidence, and more mundane claims require less so.

"The way hockey was played changed dramatically from 1888 to 1893, despite the short time frame and without significant changes to the rules or to the makeup of teams" is an extraordinary claim.

"The way hockey was played did not change dramatically from 1888 to 1893, because it was a short time frame and there were no significant rule changes or changes to the makeup of teams" is not.

Quote:
Originally Posted by tarheelhockey View Post
All that I've seen suggests Stewart and especially Cameron deserve the lion's share. And there's nothing wrong with that.
Yes, that's certainly defensible. How much makes up the lion's share is open to discussion, of course, and this has nothing to do with goaltenders in general being less valuable at the time. Paton could be far and away the best goalie at the time and still be behind Cameron and Stewart, if they're further ahead in their positions and goaltenders are generally less valuable. Or Paton could be an average goalie benefitting from the team defence, while goaltenders in general are still very valuable.

You're also probably operating under the assumption that my numerical analysis evaluates goalies based solely on their recorded GAA. It's not that crude. The quality of the team is considered.

Quote:
Originally Posted by tarheelhockey View Post
I'm not saying anything about the motivation behind it. I'm just saying that it conflicts with primary-source evidence. Not the first time someone has made a system that doesn't produce perfect results, is it? There's no shame in that.
If you're not ascribing motivation, you should be careful about wording like "and their data can be crunched and twisted backward to "prove" that one player deserves a certain amount of credit for his team results..."


Last edited by Iain Fyffe: 09-04-2012 at 09:14 PM.
Iain Fyffe is offline   Reply With Quote
Old
09-04-2012, 10:10 PM
  #56
tarheelhockey
Global Moderator
 
tarheelhockey's Avatar
 
Join Date: Feb 2010
Location: The Triangle
Country: United States
Posts: 53,909
vCash: 500
Quote:
Originally Posted by Iain Fyffe View Post
No they don't. Did you see my earlier post about how 5 of 31 described goals were scored from scrimmages in 1893? You haven't addressed any of the evidence I presented, really.
I have, repeatedly, and you sweep it aside with demands for "corroborating evidence" that you know is nonexistent.

Again, this is not a death penalty case. We don't need to be absolutely 100% certain that every single individual movement of the puck is accounted for, in order to get to what we're looking for. Did Tom Paton play exceptionally good hockey compared to his peers? Not according to the guys who stood there, watched the whole game, and wrote down their observations for posterity. If you are really bothered by the shot-count controversy, we can move that to another thread for discussion. The larger, more relevant point is not being addressed and I'm sure I'm not the only one with a headache from this back-and-forth.


Quote:
The whole point is that I don't believe that there can be any corroboration, leaving your interpretation as guesswork. Plausible guesswork, certainly, but guesswork nonetheless.
And this differs how from the approach of arbitrarily assigning all 7 players equal credit for a win?

Quote:
2. "I believe hockey in 1890 featured significantly fewer shots on goal as hockey a few years later. In fact, I think in 1890 it was about 10 per game."

The second requires much more evidence than the first. You'd not only have to substantiate the specific claim of the 10 shots per game, but also that it was substantially different than a few years later.
If you would in fact like to bring this into another thread, please begin by sharing the earliest known "proof" of a league where shot counts above 15 were the norm. I am sincerely interested to know how far back we can trace it, and having read basically all game summaries I'm sure you can get to it a lot faster than I.


Quote:
When new information is viewed through the filter of an existing conclusion, rather than reviewing the conclusion to see if it holds up to the new information, there's a problem.
I took as many game reports as I could find, and recorded for the group EVERY mention of a shot, save or goal. And EVERY time the totals came out around 5 or 10 aside, game after game. That is not working backward from a conclusion, it's working forward from primary sources that reinforce each other.

Again, if you know of a game in the 1880s in which more than a dozen shots were observed on a consistent basis, I'm all ears. Otherwise you're just guessing too.


Quote:
"The way hockey was played changed dramatically from 1888 to 1893, despite the short time frame and without significant changes to the rules or to the makeup of teams" is an extraordinary claim.
If you're going to build a straw man, etc.


Quote:
You're also probably operating under the assumption that my numerical analysis evaluates goalies based solely on their recorded GAA. It's not that crude. The quality of the team is considered.
Oh, I know; I've read the articles. You were not shy about declaring the assumption that "players considered to be the best must have been the best" and that the system is basically designed to reinforce that assumption. You also openly admit that the numbers are fudged where necessary to make sure they meet the conclusion that the most recognized players are necessarily statistically superior. And that all positions receive a pre-determined portion of credit for team results.

Basically it's a way of taking what we already know about Paton, jumbling some numbers around, and pretending we have uncovered new information. Which is what I meant in that last line you quoted.

tarheelhockey is offline   Reply With Quote
Old
09-04-2012, 10:51 PM
  #57
Iain Fyffe
Hockey fact-checker
 
Iain Fyffe's Avatar
 
Join Date: Feb 2009
Location: Fredericton, NB
Country: Canada
Posts: 3,509
vCash: 500
Quote:
Originally Posted by tarheelhockey View Post
Did Tom Paton play exceptionally good hockey compared to his peers? Not according to the guys who stood there, watched the whole game, and wrote down their observations for posterity.
Are you saying there were goaltenders seen as better than Paton?

Quote:
Originally Posted by tarheelhockey View Post
And this differs how from the approach of arbitrarily assigning all 7 players equal credit for a win?
The approach is that, on average, each position will receive approximately equal credit, league-wide. That's based on the assumption that the players knew what they were doing, and that if one position were somehow contributing less than it could be, that position's role would be adjusted to something of an equilibrium.

That's not the same thing as giving all seven players on a team the same credit for a win. A team's goaltender is not guaranteed to receive one-seventh of the team's credit, just because he's a goaltender. All goaltenders, as a whole, will receive about one-seventh of the total credit, but even that can vary a bit.

Quote:
Originally Posted by tarheelhockey View Post
If you would in fact like to bring this into another thread, please begin by sharing the earliest known "proof" of a league where shot counts above 15 were the norm. I am sincerely interested to know how far back we can trace it, and having read basically all game summaries I'm sure you can get to it a lot faster than I.
The earliest I'd have any confidence in at all are those 1907 Stanley Cup summaries, as I mentioned early on, which provided about 20 shots per game. But even those are not solid, and are based on only four games involving only two teams.

The Manitoba senior league counted shots in the mid-1910s. In 1914/15 the average was 32.2 per game; in 1915/16 it was 24.3 per game; and in 1917/18 it was 23.0 per game. (Edit: this is per team per game, to be clear, not a total of both teams).

Quote:
Originally Posted by tarheelhockey View Post
I took as many game reports as I could find, and recorded for the group EVERY mention of a shot, save or goal.
You're capitalizing the wrong word. It should be "every MENTION of a shot, save or goal."

Quote:
Originally Posted by tarheelhockey View Post
Again, if you know of a game in the 1880s in which more than a dozen shots were observed on a consistent basis, I'm all ears. Otherwise you're just guessing too.
I'm not putting forward a specific number. I'm suggesting it was about the same as in the 1890s (not knowing exactly what that is), seeing no reason to believe otherwise, while you've suggested something changed significantly from the late 1880s to the early 1890s.

Quote:
Originally Posted by tarheelhockey View Post
If you're going to build a straw man, etc.
What part of that was inaccurate?

Quote:
Originally Posted by tarheelhockey View Post
Oh, I know; I've read the articles. You were not shy about declaring the assumption that "players considered to be the best must have been the best" and that the system is basically designed to reinforce that assumption.
So, you think a system that rated a journeyman defenceman as better than a Hall of Famer would be better?

Quote:
Originally Posted by tarheelhockey View Post
You also openly admit that the numbers are fudged where necessary to make sure they meet the conclusion that the most recognized players are necessarily statistically superior.
This is not accurate. Adjustments are made to players' defensive value, within certain bounds. Sometimes the adjustment that would be required to make a particular players superior to another would be too large to be realistic. There's only so much you can do without just making up numbers.

Quote:
Originally Posted by tarheelhockey View Post
And that all positions receive a pre-determined portion of credit for team results.
On average, across the league-season, approximately.

Quote:
Originally Posted by tarheelhockey View Post
Basically it's a way of taking what we already know about Paton, jumbling some numbers around, and pretending we have uncovered new information.
Nonsense. It's taking what we already know about Paton, and attaching a number to it in a systematic way, in order to better allow comparisons to other players.

It's not designed to bring surprises. "It turns out Baldy Spittal is the best player ever!" It's designed to allow comparisons of different players.


Last edited by Iain Fyffe: 09-05-2012 at 08:59 AM.
Iain Fyffe is offline   Reply With Quote
Old
09-05-2012, 12:33 AM
  #58
seventieslord
Student Of The Game
 
seventieslord's Avatar
 
Join Date: Mar 2006
Location: Regina, SK
Country: Canada
Posts: 31,183
vCash: 500
this is the most compelling thread of the year. Thanks guys.

seventieslord is online now   Reply With Quote
Old
09-05-2012, 11:40 AM
  #59
tarheelhockey
Global Moderator
 
tarheelhockey's Avatar
 
Join Date: Feb 2010
Location: The Triangle
Country: United States
Posts: 53,909
vCash: 500
Quote:
Originally Posted by Iain Fyffe View Post
Are you saying there were goaltenders seen as better than Paton?
I'm saying there is no evidence that Paton was seen as better than other goaltenders.


Quote:
The approach is that, on average, each position will receive approximately equal credit, league-wide.
Then it's no surprise that you jumped all over the suggestion that goalies were less important in the early days. That would undermine your method, wouldn't it?

Quote:
That's based on the assumption that the players knew what they were doing, and that if one position were somehow contributing less than it could be, that position's role would be adjusted to something of an equilibrium.
Of course, that is pure speculation.


Quote:
The earliest I'd have any confidence in at all are those 1907 Stanley Cup summaries, as I mentioned early on, which provided about 20 shots per game.
Earlier you went out of your way to correct me that trustworthy summaries from the 1880s were "rare" rather than nonexistent. So, cards on the table. Which ones were you referring to?


Quote:
I'm not putting forward a specific number. I'm suggesting it was about the same as in the 1890s (not knowing exactly what that is), seeing no reason to believe otherwise, while you've suggested something changed significantly from the late 1880s to the early 1890s.
Something DID change significantly in those years: the experience levels of organized hockey players. Top-level hockey matured from a carnival novelty to an establishment. That alone is reason to expect a change in the level of play.


Quote:
So, you think a system that rated a journeyman defenceman as better than a Hall of Famer would be better?
If only Paton was a Hall of Famer.

Quote:
This is not accurate. Adjustments are made to players' defensive value, within certain bounds. Sometimes the adjustment that would be required to make a particular players superior to another would be too large to be realistic. There's only so much you can do without just making up numbers.

...

Nonsense. It's taking what we already know about Paton, and attaching a number to it in a systematic way, in order to better allow comparisons to other players.

Right.

So tell me which part of the following description of your system is inaccurate.

- You have assumed that Paton was the best goalie of his time, based on his team's level of success.
- You have assumed that all other goalies were inferior to Paton, based on their teams' lack of success.
- You have assumed that the goaltending position was considered of equal importance to other positions, based on the fact that goalies existed.
- You have designed a system that will arrange players of that era in order according to the assumptions above.
- This system purports to demonstrate that Paton was the not only the best goalie of his time, but one of the best players of his time, but not based on team success and not based on sweeping assumptions about goaltenders' role in game outcomes.

I'm not attempting to strawman you here. Is any of that inaccurate? Because it certainly sounds like you have created a system that fudges like crazy to reinforce conclusions about individuals drawn from nothing but team outcomes.

tarheelhockey is offline   Reply With Quote
Old
09-05-2012, 12:46 PM
  #60
Iain Fyffe
Hockey fact-checker
 
Iain Fyffe's Avatar
 
Join Date: Feb 2009
Location: Fredericton, NB
Country: Canada
Posts: 3,509
vCash: 500
Quote:
Originally Posted by tarheelhockey View Post
I'm saying there is no evidence that Paton was seen as better than other goaltenders.
Who was seen as the best goaltender then?

Quote:
Originally Posted by tarheelhockey View Post
Then it's no surprise that you jumped all over the suggestion that goalies were less important in the early days. That would undermine your method, wouldn't it?
I "jumped all over it" because it's based on an assumption that I see as being baseless, given the evidence proffered.

And you're questioning my motives again.

Quote:
Originally Posted by tarheelhockey View Post
Of course, that is pure speculation.
Yes it is, and is admitted as such. The analysis is only as good as the assumptions upon which it is built.

Now, can you point me to where I said Tom Paton is the best goaltender from this era because Point Allocation says so?

Quote:
Originally Posted by tarheelhockey View Post
Earlier you went out of your way to correct me that trustworthy summaries from the 1880s were "rare" rather than nonexistent. So, cards on the table. Which ones were you referring to?
Did I say 1880s? I meant to refer to the early days in general. Looking back I did once say "at this time", which was intended as a general "in early days". I clearly didn't mean 1880s only, because I proceeded to provide an example from the 1890s.

Quote:
Originally Posted by tarheelhockey View Post
Something DID change significantly in those years: the experience levels of organized hockey players. Top-level hockey matured from a carnival novelty to an establishment. That alone is reason to expect a change in the level of play.
Change in the level of play, sure. But a change in the nature of play? Why does more experience lead to more shots on goal? Wouldn't the defensive players be more experienced as well? In fact isn't defence usually seen as something that relies more heavily on experience than does offence?

You can provide a hypothesis about why something might happen, but (A) you haven't actually demonstrated that the thing happened at all, and (B) your hypothesis is questionable by itself. Why would greater experience lead to more shots on goal?

Carnival novelty...did you know that a fight broke out after the very first recorded indoor hockey game in 1875? Skaters were upset that hockeyists were taking up so much time and chewing up the ice, and a fight broke out. Does one often get into fights about carnival novelties?

Quote:
Originally Posted by tarheelhockey View Post
If only Paton was a Hall of Famer.
If only someone from the Hall of Fame selection committee had seen him play...

Quote:
Originally Posted by tarheelhockey View Post
So tell me which part of the following description of your system is inaccurate.

- You have assumed that Paton was the best goalie of his time, based on his team's level of success.
Not really accurate. I have assumed he was the best goalie of his time, because as far as I can tell he was the best goaltender of his time.

If Paton was not the best goaltender of his time, who was?

Quote:
Originally Posted by tarheelhockey View Post
- You have assumed that all other goalies were inferior to Paton, based on their teams' lack of success.
Again not accurate. It's not that simple. A goaltender on a team with less success can be rated as superior to one on a team with more success. Playing on a team with 10 wins does not automatically make your rating better than someone on a team with 5 wins.

Quote:
Originally Posted by tarheelhockey View Post
- You have assumed that the goaltending position was considered of equal importance to other positions, based on the fact that goalies existed.
Based on the fact that goalies existed throughout the history of hockey, and their role (stopping pucks to the exclusion of other things) had not changed, I do assume that, on average, a goaltender is as valuable to his team as another position.

And yet, the system rarely sees a goaltender as being the best player in a season. It's funny because most player rating systems have goaltenders as being the most volatile season-to-season, but in mine most goaltenders come out with quite consistent numbers. There's something of a regression to the mean built in, where goaltenders on great teams are pulled back a bit toward the avarage from what their GAA might imply, and goaltenders on bad teams are as well.

Quote:
Originally Posted by tarheelhockey View Post
- You have designed a system that will arrange players of that era in order according to the assumptions above.
No, because a couple of the assumptions above are incorrect. It is designed according to assumptions, of course, but you're off on some of them.

Quote:
Originally Posted by tarheelhockey View Post
- This system purports to demonstrate that Paton was the not only the best goalie of his time, but one of the best players of his time, but not based on team success and not based on sweeping assumptions about goaltenders' role in game outcomes.
What else would you suggest basing player success on, than team success? Isn't the point of hockey to win hockey games, and therefore the players that win the most are, in general, the best players?

I don't purport that the system demonstrates Paton was the best goalie of his time, I'd say it reflects it, and attaches a number to it.

Look, there's only so much credit you can give one player on a 7-man team. Given how good the team was defensively, it order to make Paton look like a humdrum player you're need to pile the defensive credit onto Cameron and Stewart so high as to be untenable.

You can't look at a player in a vaccuum, you must consider his team. Even with Paton being rated so highly, Cameron and Stewart are both still rated as dominant defensively. I believe I have Stewart rated as the best overall player one season, even though he did nothing offensively. These two are already rated as the two best defenders, by far, of their era, and you're suggesting that they should be rated even higher.

Based on this discussion I am inclined to up Cameron's ratings a bit at the expense of Paton. But unless this adjustment is unrealistically large, Paton will still be rated as the best goaltender of his time, and one of the best players (behind Cameron and Stewart).

Quote:
Originally Posted by tarheelhockey View Post
I'm not attempting to strawman you here. Is any of that inaccurate? Because it certainly sounds like you have created a system that fudges like crazy to reinforce conclusions about individuals drawn from nothing but team outcomes.
No, that's still wrong. It's a system designed to allow comparisons of players, and as an input includes how good a player was seen as being. It's not a matter of "Ha! I told you So-and-so was great!" it's "Well, we know So-and-so from the 1910s was really good, and also Such-and-such from the 1930s was really good. I wonder how they compare to each other?"

Do you know what the fudge factor is? It's an adjustment to the player's defensive value. Some players get their defensive credit upped by 10-30%. The players seen as dominant defensively, like Frank Nighbor, get a higher adjustment, say 60% (that is, if he was going to receive 2.0 defensive points he'd instead get 3.2, but actually it would be less than that because of the way the calculation goes, it would probably be 3.0. But he's an extreme case.)

Some players seen as particularly dreadful on defence have their defensive credit but by 10-20%. The great majority of players have no defensive adjustment.

If you want to call some players being adjusted by 10 to 30% with respect to half of the value equation "fudging like crazy", more power to you.

Iain Fyffe is offline   Reply With Quote
Old
09-05-2012, 01:43 PM
  #61
tarheelhockey
Global Moderator
 
tarheelhockey's Avatar
 
Join Date: Feb 2010
Location: The Triangle
Country: United States
Posts: 53,909
vCash: 500
The Great Tom Paton Debate

I'm creating this thread as a destination for the debate over Paton, which has currently taken over the HOH Top-40 Goaltenders thread. Seeing as he is generally not considered a candidate for that list, and conversation has just about stopped on a very important project, it would seem better to move the Paton discussion over here and let the list-making thread get back to business.

Taco MacArthur agreed to move the Paton-related discussion to this thread at his earliest convenience. For now, I'll simply use it as a place to put a bit more info that I dug up on the 1892 season.

tarheelhockey is offline   Reply With Quote
Old
09-05-2012, 01:48 PM
  #62
Bear of Bad News
Mod Supervisor
HFBoards Escape Goat
 
Bear of Bad News's Avatar
 
Join Date: Sep 2005
Location: Windsor
Posts: 5,542
vCash: 500
This is an interesting offshoot of the goaltender ranking project, and I've attempted to pull the offshoot into its own thread.

To the extent that I've missed something, please let me know (although I think that I deserve bonus points for using "olde tyme" in the thread subject).

Bear of Bad News is offline   Reply With Quote
Old
09-05-2012, 02:00 PM
  #63
tarheelhockey
Global Moderator
 
tarheelhockey's Avatar
 
Join Date: Feb 2010
Location: The Triangle
Country: United States
Posts: 53,909
vCash: 500
Paton in the 1892 season.

January 8, 1892
Montreal AAA 3, Ottawa 4
Tom Paton vs J. Morell


I assume the Ottawa goalie is actually Albert Morel, a college student who played senior hockey during his school years.

Summary: Allan Cameron and Archie McNaughton were missing from the AAA lineup for most of the 1892 season, depriving them of two top players. Cameron was replaced on the cover point by a J. Barry, and as you can see in the first reference to Paton, the Montreal defense was out of sorts without him, leaving Paton exposed in the OT loss. Unfortunately there is very little detail about the gameplay other than this interesting strategic observation. Paton is described as a cool and steady veteran.

Quote:
Originally Posted by References to Paton
The [offensive] play was weak, and the point and cover point thought themselves obliged to keep close to the forwards. This left Paton practically without any support, and it cost the Montrealers just two games. With the forwards and the defence men bunched at the wrong end of the rink, and the puck once forced behind it was almost an impossibility to stop it. Even the coolness and steadiness of the veteran Paton was not equal to the emergency.
...
The second game was slow and about even, strong attacks being made on both goals and very close shots being turned off by both goal-keepers. This game lasted 19 minutes, when Kirby scored for Ottawa. The Montrealers claimed that the puck did not go through, and the umpire was changed.
...
In the fourth game things were evened up again by Bradley securing another goal. Two all.
...
Ottawa began to rush things and after five minutes' play Kerr scored for Ottawa. Three to two.
...
The Ottawas seemed the more confident and went in with a rush. They had the best of it in this game and finally won the match by 4 to 3 from a neat score by Kerr.
Quote:
Originally Posted by References to Morell
The second game was slow and about even, strong attacks being made on both goals and very close shots being turned off by both goal-keepers.
...
The third game fell to the Montrealers two minutes after the beginning of the second half.
...
The sixth game began to show some real hard hockey, and just before the call of time Montreal succeeded in getting another goal, making the match a tie.

January 21, 1892
Montreal AAA 2, Ottawa 10

Summary: For reasons unmentioned, Montreal appears to have brought their "B" team to this challenge rematch. Paton, Cameron, Stewart and McNaughton were among the missing, and the substitutes were trampled. The goaltender, Shaw, was said to have "stopped magnificently at times" but was frequently left by himself with an Ottawa forward as the inexperienced cover points were not up to the task of containing a fast Ottawa attack.


January 29, 1892
Montreal AAA 1, Shamrocks 2
Tom Paton vs J Fyfe

Summary: Again, AAA plays without their famous point/coverpoint duo. This time they have the opposite problem from the previous match -- instead of playing too aggressively, the subs play too conservatively and give up possession for the majority of the first half. This time Paton is singled out for having seen a lot of rubber and implicitly having kept his team in the game until they could get their possession situation straightened out.

Quote:
Originally Posted by References to Paton
In the first half the Shamrocks had the best of what play there was, which consisted of a series of scrambles around the Montreal end. During their progress Paton missed two shots out of a hundred or so [ed: a shot count! ] and of course it meant two points for Shamrock.
...
In the last half Montreal commenced to pick up their forwards... but their back division was weak, hugged their goals too much and naturally gave Brown, McQuisten, or Murray, the latter of whom is a regular star, the chance to keep the puck away...
Quote:
Originally Posted by References to Fyfe
The Montreal, however, made a desperate rush towards the end and after some hard work Irving scored the third and last game for Montreal.
Quote:
Originally Posted by Amusing Anecdote
Both teams were uniformed almost exactly alike, which was decidedly confusing to both players and referee. There ought to be something to prevent this.
[/QUOTE]


February 2, 1892
Montreal AAA 4, Victorias 8


The Gazette only mentions the score of this game, with no description or rosters.

February 3, 1892
Montreal AAA 2, Brittania 2
Tom Paton vs W Cameron


Summary: Still no Cameron in the lineup. Game play is described in only the broadest terms, and the entirety is quoted below. The writer names 4 outstanding players from each team, and both goalies are included.

Quote:
Originally Posted by Game summary
In the first half the M.A.A.A. kept the "puck" in the vicinity of the Brits' goals and when the whistle sounded for half time the score stood M.A.A.A. 2; Brits 0. The second half woke up the Brits and the puck was kept in the territory of both goals when time was called.

February 5, 1892
Montreal AAA 4, Shamrocks 1

Does not appear to have been covered by the Gazette.



February 11, 1892
Montreal AAA 1, Ottawa 3
Tom Paton vs Albert Morel


Summary: Finally a more detailed game summary from a season in which the Gazette's interest in MAAA seemed to have flagged. Montreal had Stewart on board for this game, but we see a real difference in their play with the new coverpoint, Barry. Even though he rushes well, he is no Allan Cameron and the Ottawa attack eventually wears down the Montreal defense. Both goaltenders apparently played well, but Morel was able to hold off the AAA attack until the Montreal defense cracked.

Quote:
Originally Posted by References to Paton
... the way the players hustled around from one goal to the other was surprising. Both goals were threatened time and again and the goal keepers were kept busy stopping the well directed shots off the forwards.
...
It seemed only a matter of a moment for a goal to be scored by either teams [sic]. Still the goal keepers were too sharp and they made several beautiful stops to the great delight of the 2,000 spectators.
...
Then Kerr rushed up at a lively pace and passed out to Russell, who carried the puck past Paton and scored.
...
They [Ottawa] rushed the game and after seven minutes Russell scored for Ottawa. After the face the Ottawas played a still better game and were continually attacking their opponents' flags. It took eleven minutes for the champions to score, however, the shot being again made by Russell.
Quote:
Originally Posted by References to Moran
... the way the players hustled around from one goal to the other was surprising. Both goals were threatened time and again and the goal keepers were kept busy stopping the well directed shots off the forwards.
...
It seemed only a matter of a moment for a goal to be scored by either teams [sic]. Still the goal keepers were too sharp and they made several beautiful stops to the great delight of the 2,000 spectators.
...
The latter [AAA coverpoint Barry] made many excellent runs with the "puck", but luck was always against him. [possible implicit shots]
...
They [AAA] had Ottawa playing a defence game, and after six minutes' play Kingman scored.
...
They [AAA] were playing the best game, but could not score. [possible implicit shots]

March 7, 1892
Montreal AAA 1, Ottawa 0
Tom Paton vs Albert Morel

Summary: Montreal brings stars Allan Cameron, Archie McNaughton and Archie Hodgson back into the fold and -- no surprise -- completely shuts down the Ottawa attack that had shredded them in earlier games. Ottawa appears to have recovered enough in the second half to put some pressure on Paton, but by and large MAAA dominated the game and Morel must have played well in Ottawa's goal to prevent a wider margin of defeat.

Quote:
Originally Posted by References to Paton
The visitors' defence was a veritable stone wall, as the score of one goal to nothing clearly shows. Paton had many stops to make, nevertheless, but they were of the free and easy order and he cleverly drove the puck out of his territory. Stewart and Cameron swooped around the puck in admirable style...
...
The Ottawas made but few trials, as they were mostly on the defensive.
...
[In the second half] The Ottawas picked up wonderfully after a few minutes and they made many unsuccessful rushes up to Paton, who was always equal to the occasion.
...
There was then just 13 minutes left when the game was resumed, and during that time the Ottawas tried hard to score, but their efforts were in vain, and so the match stood 1 to 0. [possible implicit shots]
Quote:
Originally Posted by References to Morel
[Montreal forwards] almost continually had the puck... In the first half Hodgson, McNaughton and Lowe made nearly a hundred trials for goal, but generally missed by the merest chance.
...
After 17 minutes' play Hodgson got a straight tip from McNaughton. The puck went whizzing through a couple pools of water, and a well directed shot by Hodgson sent it through the flags. The remainder of the half was almost entirely in favor of the Montrealers, and when the referee whistled the puck was behind the Ottawa flags.

Conclusions: If we learn anything from the 1892 season, it's the value of Allan Cameron to those MAAA squads. Without him they were 0-3 against the up-and-coming Ottawa squad, with a weak cover point performance being a specific problem. At first Barry tried to play Cameron's aggressive style and got burnt; then, attempting to hang back and give Paton more help only meant longer defensive possessions for the opponent. The state of the AAA defense was even worse with Stewart out of the lineup, falling to the likes of the Shamrocks and losing 10-2 to Ottawa (to be fair, some poor sap took Paton's place in that game).

That said, Paton was clearly a more important part of the club's fortunes without a stellar defense in front of him. Based on the game summaries we can clearly identify three strong performances on his part, plus the 1-0 whitewashing of Ottawa once the MAAA lineup got back to full strength. Paton's opponents are also described in more charitable terms than before, particularly Morel of Ottawa. It would seem that Paton was considered a seasoned veteran by this time, and the younger Morel would probably have been viewed as a promising foil. It's a shame the Gazette didn't cover every game of this season.

tarheelhockey is offline   Reply With Quote
Old
09-05-2012, 03:01 PM
  #64
tarheelhockey
Global Moderator
 
tarheelhockey's Avatar
 
Join Date: Feb 2010
Location: The Triangle
Country: United States
Posts: 53,909
vCash: 500
Quote:
Originally Posted by Iain Fyffe View Post
Who was seen as the best goaltender then?
Sadly, we don't have any record of consensus opinion to guide us on that. If we did, this wouldn't be much of a debate at all.


Quote:
I "jumped all over it" because it's based on an assumption that I see as being baseless, given the evidence proffered.

And you're questioning my motives again.
You're the one putting yourself forth as an expert on the subject, and using proprietary ranking methods as evidence. If you want your points to stand on their own, don't hold up your personal history as verification of them.


Quote:
Now, can you point me to where I said Tom Paton is the best goaltender from this era because Point Allocation says so?
Point Allocation was mentioned in Paton's defense pages ago. Would you agree that it shouldn't be used as a method of "proving" Paton's standing compared to others?

Quote:
Did I say 1880s? I meant to refer to the early days in general. Looking back I did once say "at this time", which was intended as a general "in early days". I clearly didn't mean 1880s only, because I proceeded to provide an example from the 1890s.
Let's recap.

- First you said that "to get a reasonable estimate you'd need at least most game providing detail on at least most of the play. We don't have anything close to that.".
- Then you said that most game reporting of the early era lacked detail, with the exception of certain games.
- Then you said that even in the detailed games, mundane plays were not reported.
- Then I presented game reports which did in fact describe mundane, unremarkable saves. These reports most definitely give us a reasonable estimate based on most of the play -- your words.
- Then, all of a sudden, you changed your argument to state that we must be able to account for all saves or our estimates are invalid, and that games must be accounted for in full, significantly raising the bar from where you placed it earlier. This happened in post #36 in this thread.
- In subsequent posts, you insisted not only upon absolute completeness from game summaries, but also for corresponding confirmation in the form of shot totals from a different source.
- Separately, you claimed that the first game summaries which could provide reliable estimates of shot totals occurred in Stanley Cup matches.
- You then cited an 1893 article as evidence of shot counts, focusing largely on a single line from said article.
- Following that, in response to an article from 1888 that clearly gives a "reasonable estimate" of shot totals, you said that such summaries are rare but not nonexistent during THAT time period.
- Then you said that you were originally referring to articles from 1907 Stanley Cup matches.
- Having already claimed to have read essentially every existing game summary from the time period, you were asked to produce one which gives you a sense of shot totals from the era. You refused.
- Yet you still cited the 1893 article as evidence of shot totals during that era.


... enough moving the goalposts around to suit an argument. I'm asking you, as straightforwardly as possible, to get down to the bottom of what we are really talking about here.

Do you or do you not have reliable information regarding the number of shots in a typical game during the career of Tom Paton? If you don't, fine -- we can at least attempt to construct that information from the written record, because "reasonable estimates" (your words) from that time period do exist .


Quote:
Change in the level of play, sure. But a change in the nature of play?
Who said anything about changes in the nature of play?

Quote:
Why would greater experience lead to more shots on goal?
Off the top of my head -- more organized offensive play, particularly more organized passing between forwards, better recognition by the defensive players that they can have a role in the offense, more practice time, more experience shooting pucks, more innovation in the offensive game, flatter pucks, smoother ice, better refs. The list is probably a lot longer than that.

Quote:
If only someone from the Hall of Fame selection committee had seen him play...
Hey, you brought up the HOF as evidence of superiority. Can't have it both ways.

Quote:
Not really accurate. I have assumed he was the best goalie of his time, because as far as I can tell he was the best goaltender of his time.
Based on team results, correct? If there's some other factor, please share.

Quote:
Again not accurate. It's not that simple. A goaltender on a team with less success can be rated as superior to one on a team with more success. Playing on a team with 10 wins does not automatically make your rating better than someone on a team with 5 wins.
The key factor being goal differential, correct?

Quote:
Based on the fact that goalies existed throughout the history of hockey, and their role (stopping pucks to the exclusion of other things) had not changed, I do assume that, on average, a goaltender is as valuable to his team as another position.
Not speaking about the entire history of hockey here. In the isolated period 1885-1893, what is your evidence for the relative value of a good goaltender compared to, say, a good cover point?

Quote:
What else would you suggest basing player success on, than team success? Isn't the point of hockey to win hockey games, and therefore the players that win the most are, in general, the best players?
Well, that's a hell of a question. Maybe we should have skipped the Paton stuff and just started here.

To say the least, no, the players that win the most are not necessarily the best players. There is a lot of room for disconnect between individual performance and team success, and goaltender is probably the most disconnected.

Quote:
Look, there's only so much credit you can give one player on a 7-man team. Given how good the team was defensively, it order to make Paton look like a humdrum player you're need to pile the defensive credit onto Cameron and Stewart so high as to be untenable.
Not necessarily. We know that AAA's forwards were very good, as evidenced by their offensive results. We know that Cameron and Stewart were among the very best players at their position and were almost certainly the best duo. Montreal AAA was quite easily the best hockey team in the circuit for all but one year of Paton's career. There's nothing in particular about his individual performances that seems much more than "pretty good", so I don't see any need to remove credit from him and give it to other players in order to balance out any equation.

Quote:
These two are already rated as the two best defenders, by far, of their era, and you're suggesting that they should be rated even higher.
Or that there's something wrong with the method which causes this kind of conflict.


Last edited by Bear of Bad News: 09-05-2012 at 03:05 PM. Reason: Getting nasty
tarheelhockey is offline   Reply With Quote
Old
09-05-2012, 03:09 PM
  #65
Bear of Bad News
Mod Supervisor
HFBoards Escape Goat
 
Bear of Bad News's Avatar
 
Join Date: Sep 2005
Location: Windsor
Posts: 5,542
vCash: 500
I'm not going to ask again (and quite frankly, I'm tired of - for lack of a better word - babysitting this thread).

The next comment that is directed at an individual, and not a subject, will be the last comment by the poster in the thread.

Bear of Bad News is offline   Reply With Quote
Old
09-05-2012, 05:29 PM
  #66
Iain Fyffe
Hockey fact-checker
 
Iain Fyffe's Avatar
 
Join Date: Feb 2009
Location: Fredericton, NB
Country: Canada
Posts: 3,509
vCash: 500
Still not sure when Paton began playing hockey, but there is a reference to him playing lacrosse on ice in 1877. That was briefly a popular game as hockey was getting going as well.

Iain Fyffe is offline   Reply With Quote
Old
09-05-2012, 05:36 PM
  #67
tarheelhockey
Global Moderator
 
tarheelhockey's Avatar
 
Join Date: Feb 2010
Location: The Triangle
Country: United States
Posts: 53,909
vCash: 500
Quote:
Originally Posted by Iain Fyffe View Post
Still not sure when Paton began playing hockey, but there is a reference to him playing lacrosse on ice in 1877. That was briefly a popular game as hockey was getting going as well.
Hard to imagine. On skates or on foot, a la broomball?

tarheelhockey is offline   Reply With Quote
Old
09-05-2012, 06:01 PM
  #68
Iain Fyffe
Hockey fact-checker
 
Iain Fyffe's Avatar
 
Join Date: Feb 2009
Location: Fredericton, NB
Country: Canada
Posts: 3,509
vCash: 500
Quote:
Originally Posted by tarheelhockey View Post
Point Allocation was mentioned in Paton's defense pages ago.
That's incorrect. I mentioned that goaltender Wins should not be used in evaluating them, and then you asked whether wins were the key part to point allocation, and the discussion went from there. At least that's what I can see by reviewing the thread. If there another post I'm not considering?

Again, can you point to where I said Paton should be considered a great goaltender because Point Allocation says so? Just link to the post.

Quote:
Originally Posted by tarheelhockey View Post
Would you agree that it shouldn't be used as a method of "proving" Paton's standing compared to others?
Obviously, since I've said that already. That's not the point of the system.

Quote:
Originally Posted by tarheelhockey View Post
Let's recap.
Let's.

I said that to get a reasonable estimate you'd need at least most game summaries to provide detail on at least most of the play, and that we don't have anything close to that.

This appears to be true, because thus far you have been able to provide exactly one game summary that provides detail on at least most of the play.

I also said that most game reporting in the early era lacked this detail, with the exception of certain games. This also appears to be true, and is borne out by the fact that you only have one game summary so far with this level of detail.

So, until you can provide a majority of game reports detailing the majority of the play, the initial standard of evidence I suggested will not be met.

Now, in reference to the single game summary, yes I do expect that if you are relying upon a single game summary, you'd better be sure it's complete. I mentioned the majority of game summaries, and you provide one. If that's all you have, the information contained in that single game summary had better at least be complete.

As you provide more and more such game summaries, the standard of information in each one shrinks, because we have more information to work with.

But right now it seems your case really rests on that one summary, and as such it needs to meet a high standard of evidence.

Quote:
Originally Posted by tarheelhockey View Post
- Having already claimed to have read essentially every existing game summary from the time period, you were asked to produce one which gives you a sense of shot totals from the era. You refused.
I didn't refuse. I simply don't recall any that do. If I don't think there are any, how can I provide them?

Quote:
Originally Posted by tarheelhockey View Post
- Yet you still cited the 1893 article as evidence of shot totals during that era.
No, I only cited it as evidence that a 10-shot average in 1893 was unlikely. I'm not putting forward a number from this era because I don't believe we have anything solid to base it on.

Quote:
Originally Posted by tarheelhockey View Post
Who said anything about changes in the nature of play?
You did. You took the 1893 reference as evidence that more shots were taken in a game in 1893 than in earlier years.

Quote:
Originally Posted by tarheelhockey View Post
Off the top of my head -- more organized offensive play, particularly more organized passing between forwards, better recognition by the defensive players that they can have a role in the offense, more practice time, more experience shooting pucks, more innovation in the offensive game, flatter pucks, smoother ice, better refs. The list is probably a lot longer than that.
Now, please provide evidence that experience aids offensive play significantly more than it aids defensive play.

Quote:
Originally Posted by tarheelhockey View Post
Hey, you brought up the HOF as evidence of superiority. Can't have it both ways.
That was an illustration with respect to Point Allocation, not a reference to Paton. I could have said All-Star instead and the point would remain.

Quote:
Originally Posted by tarheelhockey View Post
The key factor being goal differential, correct?
Sorry, no, I was shortcutting that statement. Replace "wins" with "goal differential" and again the point still stands.

Quote:
Originally Posted by tarheelhockey View Post
To say the least, no, the players that win the most are not necessarily the best players. There is a lot of room for disconnect between individual performance and team success, and goaltender is probably the most disconnected.
Don't take that too literally. If good hockey players don't lead to wins, then they're not actually good players, because winning is the point of competitive hockey. Speaking in a very general sense here, of course, not claiming a perfect 1:1 ratio and bearing in mind that hockey is a team game.

Quote:
Originally Posted by tarheelhockey View Post
Or that there's something wrong with the method which causes this kind of conflict.
The only wrongness that could cause that would be the underlying method of allocating a team's success to its players. A great defensive team is made up of players, and these players are given credit for the team's defence.

Iain Fyffe is offline   Reply With Quote
Old
09-05-2012, 06:02 PM
  #69
Iain Fyffe
Hockey fact-checker
 
Iain Fyffe's Avatar
 
Join Date: Feb 2009
Location: Fredericton, NB
Country: Canada
Posts: 3,509
vCash: 500
Quote:
Originally Posted by tarheelhockey View Post
Hard to imagine. On skates or on foot, a la broomball?
Skates. A number of similar sports sort of competed with hockey at this time, lacrosse on ice, ice polo, others.

Iain Fyffe is offline   Reply With Quote
Old
09-06-2012, 02:53 AM
  #70
Theokritos
Moderator
 
Theokritos's Avatar
 
Join Date: Apr 2010
Posts: 8,422
vCash: 500
What's the evidence for Paton being the best goaltender of his time?

Theokritos is offline   Reply With Quote
Old
09-06-2012, 12:29 PM
  #71
Iain Fyffe
Hockey fact-checker
 
Iain Fyffe's Avatar
 
Join Date: Feb 2009
Location: Fredericton, NB
Country: Canada
Posts: 3,509
vCash: 500
Now, there is some evidence that shots may have been on the increase near the end of Paton's career. However, it should be said that this analysis is based entirely on game scores, and therefore may be completely invalid.

I'm talking about the average number of goals a team scores in a game. Now, if this number increases from one season to the next, we don't know if it's because there were more shots taken, or if the quality of shot taken increased. We can't tell that just from the fact that goals-per-game increases. Chances are it's a bit of both, but in what mix, we don't know.

Or heck, it could be because the overall quality of goaltending dropped. There are a number of reasons it could be, but an increase in shots is certainly one of them.

In 1887, AHAC challenge/championship matches featured 2.09 goals per team per game.

In 1888, it was 2.88.

In 1889, it was 2.79.

In 1890, it was 2.93.

In 1891, it was 2.50.

In 1892, it was 3.14.

In 1893, it was 3.88.

In 1894, it was 3.03.

In 1895, it was 3.47.

In 1896, it was 3.15.

There's a definite trend toward slightly higher goals-per-game in these years. 1893, Paton's last year, seems to be something of a fluke, although that level of goal-scoring became the norm from about 1897 to 1902, at which point it jumped up even higher, eventually going over 7.7 per game in 1907 before settling back down a bit.

The real dramatic shift seems to have occured around 1886/87.

In the 1883 Winter Carnival, there were 1.13 goals per team per game.

In 1884, it was 1.43.

In 1885, it was 1.13.

Not sure about the 1886 Montreal championship, since I'm at work and don't have access to all my files.

So if there was a dramatic shift, it seems much more likely to have occured near the beginning of Paton's career, rather than right at the end of it.

Then, of course, there's the assumption that if the game features less shots/goals, the goaltender position is necessarily less valuable. I don't know if that's true.

At times when hockey features 2.00 goals per game, each goal is much, much more valuable to a team that when it features 5.00 goals per game. Conversely, each goal saved is also that much more valuable. So you might have a goaltender saving less goals in the lower-scoring era, but each goal saved is more valuable. It's not as simple as saying "fewer shots = less valuable goaltender."

Iain Fyffe is offline   Reply With Quote
Old
09-06-2012, 12:56 PM
  #72
tarheelhockey
Global Moderator
 
tarheelhockey's Avatar
 
Join Date: Feb 2010
Location: The Triangle
Country: United States
Posts: 53,909
vCash: 500
Quote:
Originally Posted by Iain Fyffe View Post

At times when hockey features 2.00 goals per game, each goal is much, much more valuable to a team that when it features 5.00 goals per game. Conversely, each goal saved is also that much more valuable. So you might have a goaltender saving less goals in the lower-scoring era, but each goal saved is more valuable. It's not as simple as saying "fewer shots = less valuable goaltender."
I would think that the general rule would be "lower save% = less valuable goaltender". For example, goalies of the 1980s vs goalies of the 1990s. When save percentage is low, the best way to reduce your GAA is simply to keep the puck as far from your goalie as possible. The winner is largely decided according to shots allowed, since the rate of shots creating points is high. When save percentage goes up, shot quality becomes more important than quantity, because a good goalie will stop virtually any shot that he can see. A team might have 10 more shots than its opponent, and still lose due to low shot quality. We see this a lot in the NHL today.

Of course we are still stuck at Square One because without early era shot counts it's impossible to know the save percentages for certain. But let's say that the 1907 Cup games were accurately recorded at about 20 per game, which is also roughly consistent with the Manitoba senior league you referenced. The scores of those games were 4-2 and 8-6, implying a total save% of around .750 to .800. And while it is a bit of a stretch of the imagination, one would have to expect the goalies of the 1880s-90s were slightly less efficient at stopping the puck due to using only skaters' equipment, not to mention being 15-20 years less advanced in technique. I'd rather not try to put a number on it, but it's easy to see how a goalie of Paton's era might be doing his job decently well if he stopped three-quarters of the shots he saw.

That being the case, there were likely not a lot of games were the goalie singlehandedly made a difference in the outcome as we might expect from later goalies. It could happen with a well-timed save, but if a team was getting significantly outshot it would not have been likely to come out ahead on the scoreboard.

tarheelhockey is offline   Reply With Quote
Old
09-08-2012, 09:31 AM
  #73
Iain Fyffe
Hockey fact-checker
 
Iain Fyffe's Avatar
 
Join Date: Feb 2009
Location: Fredericton, NB
Country: Canada
Posts: 3,509
vCash: 500
Quote:
Originally Posted by tarheelhockey View Post
I would think that the general rule would be "lower save% = less valuable goaltender".
I don't know - another perspective would be that the lower the save percentage, the greater the effect that a particular goaltender has on his team's results, because the lower the average save percentage, the greater the variance in save percentage between goaltenders. If all goalies have numbers between .900 and .920 (average .910), then the best goaltender doesn't give you that much over the worst. But if goalies were between .700 and .800 instead (average .750), then the best gives his team a much bigger advantage over the worst.


Last edited by Iain Fyffe: 09-08-2012 at 04:40 PM. Reason: Clarification
Iain Fyffe is offline   Reply With Quote
Old
09-08-2012, 10:35 AM
  #74
overpass
Registered User
 
Join Date: Jun 2007
Posts: 3,995
vCash: 500
Quote:
Originally Posted by tarheelhockey View Post
I would think that the general rule would be "lower save% = less valuable goaltender". For example, goalies of the 1980s vs goalies of the 1990s. When save percentage is low, the best way to reduce your GAA is simply to keep the puck as far from your goalie as possible. The winner is largely decided according to shots allowed, since the rate of shots creating points is high. When save percentage goes up, shot quality becomes more important than quantity, because a good goalie will stop virtually any shot that he can see. A team might have 10 more shots than its opponent, and still lose due to low shot quality. We see this a lot in the NHL today.
Check out the Oilers' shooting numbers and percentage numbers in the 80s. They weren't winning by out shooting, they were winning with sky-high shooting percentages, or shot quality.

overpass is offline   Reply With Quote
Reply

Forum Jump


Bookmarks

Thread Tools

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off



All times are GMT -5. The time now is 04:41 PM.

monitoring_string = "e4251c93e2ba248d29da988d93bf5144"

vBulletin Copyright ©2000 - 2017, Jelsoft Enterprises Ltd.
HFBoards.com is a property of CraveOnline Media, LLC, an Evolve Media, LLC company. @2017 All Rights Reserved.