HFBoards Greatest Teams in NHL History
 Mobile Hockey's Future Become a Sponsor Site Rules Support Forum vBookie Page 2
 Notices Please do not post or solicit links to illegal game streams.

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

# Greatest Teams in NHL History

 04-21-2005, 10:06 AM #28 Ogopogo*     Join Date: Apr 2005 Location: Edmonton Country: Posts: 15,951 vCash: 500 money, post your calclations for these ratings. I would like to see how yours and mine differ. Thanks
04-21-2005, 10:32 AM
#30
Snap Wilson
Registered User

Join Date: Sep 2003
Posts: 5,838
vCash: 500
Quote:
 Originally Posted by Ogopogo money, post your calclations for these ratings. I would like to see how yours and mine differ. Thanks
The calculation is (A-B)/C where:

A = goal differential per game of the team
B = average goal differential per game of the season (i.e. zero)
C = standard deviation of goal differential per game for all the teams for the season in question.

For example, the 1976-77 Canadiens have a goal differential of +2.70 (wow). The standard deviation of the 1976-77 season is 1.05, which is historically high (the NHL average is 0.78). What this means is that the league was historically unbalanced league. The Red Wings that year lost by 1.58 goals/game. The Colorado Rockies and Washington Capitols lost by more than a goal/game as well. You have to take that into account when figuring out the dominance of the Habs. Not that it hurt that much... they still come out on top, but it's a lot closer now.

04-21-2005, 10:59 AM
#32
Ogopogo*

Join Date: Apr 2005
Location: Edmonton
Country:
Posts: 15,951
vCash: 500
Quote:
 Originally Posted by moneyp The calculation is (A-B)/C where: A = goal differential per game of the team B = average goal differential per game of the season (i.e. zero) C = standard deviation of goal differential per game for all the teams for the season in question. For example, the 1976-77 Canadiens have a goal differential of +2.70 (wow). The standard deviation of the 1976-77 season is 1.05, which is historically high (the NHL average is 0.78). What this means is that the league was historically unbalanced league. The Red Wings that year lost by 1.58 goals/game. The Colorado Rockies and Washington Capitols lost by more than a goal/game as well. You have to take that into account when figuring out the dominance of the Habs. Not that it hurt that much... they still come out on top, but it's a lot closer now.
Thank you.

04-21-2005, 12:49 PM
#34
Trottier
Very Random

Join Date: Feb 2002
Location: San Diego
Posts: 29,234
vCash: 500
Quote:
 Originally Posted by moneyp Incidentally, I have a different take on the 1979 Islanders than our man Trots. I think they were on a mission to dethrone the Habs, who had rolled over everyone else the past three seasons. They were trying to beat them during the regular season, to not show they were intimidated, and they were trying to get home ice for the Finals. I didn't think they "shot their wad." I think if they played 100 games against the Rangers that year, they would have won 70 or 80 of them, if not more. They just happened to play six games where nothing went in their favor. It happens.
Fair point. I do think there is an intangible element involved however. It is the differentiator between outstanding teams and outstanding playoff teams. And, for the most part, it comes down to experience. Maybe not as much in today's NHL, but regardless, one sees a pattern whereby teams/players have to experience (lose in) the playoffs, often several times, before they understand what it takes to succeed come the post-season. I know some disagree with me on this point, but there really is such a difference in play come April. And that accelerated pace is required nightly for a full two months. Add to that the fact that a vast majority of playoff games are decided by 1 or 2 goals and you need a certain makeup, a certain mental toughness. And the ability to handle adversity.

Many people thought that the Isles had "graduated" to that point in the spring of '79. Apparently, they needed one more harsh lesson at the hands of NYR.

I agree with you that they were on a mission to dethrone the Habs and perhaps were even looking ahead to them, prematurely. But that in itself is emblamatic of a team that is not Cup-ready, IMO. They set themselves up ripe for an upset, and Shero & Co. obliged. To be sure, nothing went NYI's way that series, but ultimately they were incapable of playing through it.

Interesting thing was, by the next season, some had already written NYI off as a bunch of underachievers (or worse, "choke artists"), a good team that was never going to be capable of winning it all.

Anyhow, looking forward to your ranking of Corado Micalef among hockey's all-time netminders.

 04-24-2005, 09:32 PM #35 reckoning Registered User     Join Date: Jan 2005 Country: Posts: 5,933 vCash: 500 This thread has been dormant for a few days, but I just found it and wanted to make a few points. 1) Goal differential: There is a formula some stats guys use to incorporate how dominant a team is in their winning percentage: (Goals for squared) divided by ((Goals for squared) + (Goals Against squared)) That will usually give you a percentage that is very close to the teams actual winning percentage (although you`ll have to disregard the OT loss point), and it gives defensive teams like New Jersey a better chance than just goal differential. 2. The 1979-80 Philadelphia Flyers: Using the above formula, the Flyers end up about 16 points lower than their actual points total. It`s the biggest differential in NHL history, which has led some to suggest that the Flyers won a lot of close games that year, so I decided to check: 1979-80 Flyers- 48 wins: 12 by 1 goal 16 by 2 goals 20 by 3 or more goals It doesn`t look like they "got lucky" in a lot of close games; so why is their mark so low? Because when they lost, they really lost. Atlanta pounded them 9-2 in the 2nd game of the year, Minnesota crushed them 7-1 in the game that ended the record unbeaten streak, Boston beat them 7-2 and the Rangers won 8-3 in a meaningless game on the last day of the season 3) Quality of opposition: Since most of the past thirty years have had unbalanced schedules, that means some teams have it easier than others. For example, the 1995-96 Detroit Red Wings rank high on all these lists, but only 28 of their 82 games that year were against teams over .500. Should something like that be taken into consideration?
04-24-2005, 11:26 PM
#36
Hockey Outsider
Registered User

Join Date: Jan 2005
Country:
Posts: 3,973
vCash: 500
Quote:
 Originally Posted by reckoning This thread has been dormant for a few days, but I just found it and wanted to make a few points. 1) Goal differential: There is a formula some stats guys use to incorporate how dominant a team is in their winning percentage: (Goals for squared) divided by ((Goals for squared) + (Goals Against squared)) That will usually give you a percentage that is very close to the teams actual winning percentage (although you`ll have to disregard the OT loss point), and it gives defensive teams like New Jersey a better chance than just goal differential. 2. The 1979-80 Philadelphia Flyers: Using the above formula, the Flyers end up about 16 points lower than their actual points total. It`s the biggest differential in NHL history, which has led some to suggest that the Flyers won a lot of close games that year, so I decided to check: 1979-80 Flyers- 48 wins: 12 by 1 goal 16 by 2 goals 20 by 3 or more goals It doesn`t look like they "got lucky" in a lot of close games; so why is their mark so low? Because when they lost, they really lost. Atlanta pounded them 9-2 in the 2nd game of the year, Minnesota crushed them 7-1 in the game that ended the record unbeaten streak, Boston beat them 7-2 and the Rangers won 8-3 in a meaningless game on the last day of the season 3) Quality of opposition: Since most of the past thirty years have had unbalanced schedules, that means some teams have it easier than others. For example, the 1995-96 Detroit Red Wings rank high on all these lists, but only 28 of their 82 games that year were against teams over .500. Should something like that be taken into consideration?
Good post.

Point 1: The formula you're talking about is called "Pythaogrean Win Percentage". It was invented by baseball writer/analyst Bill James. It's an excellent predictor of winning: the correlation between actual and PWin% is around .95!

Alan Ryder wrote a very interesting (but also very complicated and technical) article about Pythagorean Win Percentage and other similar formulae. You may find it interesting. http://www.hockeyanalytics.com/Resea...babilities.pdf

Point 2: Interesting. The fact that they got blown out a lot makes sense, given the results.

Point 3: Yes! The quality of the schedule should definitely be taken into account. There's an interesting article about by Iain Fyffe, located here: http://www.puckerings.com/research/unbiased.html. The problem with looking at adjusted/"unbiased" standings is that it's extremely time-consuming to calculate.

 04-26-2005, 09:40 AM #37 Snap Wilson Registered User     Join Date: Sep 2003 Posts: 5,838 vCash: 500 Whoah, it's alive! Yes, unbalanced schedules (and odd flukes such as the '80 Flyers) should be taken into consideration, which is why I use multiple-year figures, so that all of the eggs aren't shoved into one year's proverbial "basket." In specific regard to the 1995-96 Red Wings, I'm not sure how much it matters any way. According to the estimable Mr. Fyffe, they would have only lost three points in the standings facing a "balanced" schedule.
 04-26-2005, 09:45 AM #38 Snap Wilson Registered User     Join Date: Sep 2003 Posts: 5,838 vCash: 500 Oh, and regarded expected winning percentage, I use an entirely different formula. I calculate the standard deviation scores (shown above) for point differential and simply adjust them to the standard deviation of points/game for that particular year. This is a bit harder to figure on the fly, but the correlation is slightly higher than the "squared" method. This method is actually more useful for the NBA, where the factor tends to fluctuate.
05-05-2005, 03:12 PM
#39
looooob
Registered User

Join Date: Feb 2002
Posts: 1,885
vCash: 500
Quote:
these are interesting. definitely proving the point from the other thread(s) that Gretzky benefited from a bad division (and that Calgary-Edmonton was nowhere near the battles of MTL-BOS in the 70s), as those late 80s calgary teams are nowhere to be seen on this list

Forum Jump

 Bookmarks