HFBoards

Go Back   HFBoards > General Hockey Discussion > National Hockey League Talk
Mobile Hockey's Future Become a Sponsor Site Rules Support Forum vBookie Page 2
National Hockey League Talk Discuss NHL players, teams, games, and the Stanley Cup Playoffs.

+ / - suggestion to make this stat meaningful

Closed Thread
 
Thread Tools
Old
10-20-2005, 09:17 AM
  #1
eye
Registered User
 
Join Date: Feb 2003
Location: around the 49th para
Posts: 1,607
vCash: 500
+ / - suggestion to make this stat meaningful

I read on another board quite a few times in the past few years a terrific suggestion that for some unknown reason has never been applied in hockey. As it stands now the + / - stat is insignificant and really tells us very little about the player either offensively or defensively. It's time to make this stat meaningful by expanding the stat to show every players goals for and goals against and you can take it a step further to include PP and PK situations. It's time to bring this stat into the modern era of common sense. e.g. Peter Forsberg is on the ice for 14 goals for at ES and 6 against his + / - would be 14 / 6. I think most of us can figure out the difference between the 2. On the PP he has been on the ice for 21 goals for and 2 shorthanded goals against (PP 21 / 2) and on the PK 4 against and 1 for (PK 4 / 1). All hypothetical of course but it's time to use more detail or baseball type analogy to make this stat actually mean something, create more fan interest and to acurately describe the players contributions of lack there of.

Now let's find out if the NHL brass actually do consider the input of their most loyal fans on HF.

eye is offline  
Old
10-20-2005, 09:19 AM
  #2
ACC1224
Registered User
 
Join Date: Aug 2002
Posts: 30,990
vCash: 500
so what is the suggestion?

ACC1224 is online now  
Old
10-20-2005, 09:22 AM
  #3
eye
Registered User
 
Join Date: Feb 2003
Location: around the 49th para
Posts: 1,607
vCash: 500
Quote:
Originally Posted by ACC1224
so what is the suggestion?
Must be a Toronto fan right?

eye is offline  
Old
10-20-2005, 09:28 AM
  #4
ACC1224
Registered User
 
Join Date: Aug 2002
Posts: 30,990
vCash: 500
Quote:
Originally Posted by eye
Must be a Toronto fan right?

ACC1224 is online now  
Old
10-20-2005, 09:33 AM
  #5
eye
Registered User
 
Join Date: Feb 2003
Location: around the 49th para
Posts: 1,607
vCash: 500
Quote:
Originally Posted by ACC1224
I should not have lowered myself but you asked for it. The suggestion is quite obvious. Feel free to comment. I can't think of one negative reason not to change the way this stat is kept and presented to the fans and management alike.

eye is offline  
Old
10-20-2005, 09:36 AM
  #6
ACC1224
Registered User
 
Join Date: Aug 2002
Posts: 30,990
vCash: 500
Quote:
Originally Posted by eye
I should not have lowered myself but you asked for it. The suggestion is quite obvious. Feel free to comment. I can't think of one negative reason not to change the way this stat is kept and presented to the fans and management alike.
what was it that I asked for? the suggestion as I read it isn't clear. I don't see how dividing all these numbers is going to come up with a meaningful stat.

ACC1224 is online now  
Old
10-20-2005, 09:45 AM
  #7
tmg
Registered User
 
Join Date: Jul 2003
Location: Ottawa
Posts: 1,518
vCash: 500
A stat I've suggested before, along the vein of +/- purely for the purpose of measuring a player's relative defensive reliability, is a "player GAA". (Nonpowerplay goals against player's team while said player is on the ice)*60/(nonpowerplay minutes played by player).

Basically, how many goals are scored against your team while you're on the ice, per 60 minutes of icetime?

It's not perfect, no... but I think it's be more revealing than plus-minus, at least in terms of defensive specialists. If this stat were back-calculated on the past few seasons, I wonder how much correlation there would be between the strongest numbers by this stat, and the Selke voting.

tmg is offline  
Old
10-20-2005, 11:44 AM
  #8
zeke
#freewilly
 
zeke's Avatar
 
Join Date: Mar 2005
Posts: 30,648
vCash: 500
basically, what you're saying is forget the "plus" part of the "plus/minus", and just look at the "minus".

I agree.

It would improve the stat, although it would still be far from perfect. Ideally, you'd be able to adjust the stat based on minutes, quality of opponent, and quality of actual opposing players on the ice in your shift, as well as quality of your team, and your teammates.

Interestingly, most european leagues keep track of "PLUS" and "MINUS" seperately already.

zeke is offline  
Old
10-20-2005, 11:54 AM
  #9
lemieux32*
 
lemieux32*'s Avatar
 
Join Date: Aug 2005
Country: United States
Posts: 1,280
vCash: 500
It doesn't seem like any of these suggestions solve the real problem: giving someone a plus or a minus when he is uninvolved or responsible in any way. Such as the guy who just came on the ice and has nothing to do with the play. Or the winger sticking to his point man and the puck never comes there but the other team scores.

lemieux32* is offline  
Old
10-20-2005, 12:00 PM
  #10
Jester
Registered User
 
Jester's Avatar
 
Join Date: Jul 2004
Location: St. Andrews
Country: Scotland
Posts: 34,075
vCash: 500
+/- is flawed in the sense that it is difficult to have a bad one on a good team, and it is easy to have a bad one on a bad team. however, for outliers on the teams it says a great deal... very good defensive players, for example, tend to be close to even at the very least on bad teams... and horrible defensive players won't have a great plus even if they are on a good team.

the stat shows outliers more than anything else.

Jester is offline  
Old
10-20-2005, 12:20 PM
  #11
UnderratedBrooks44
Registered User
 
UnderratedBrooks44's Avatar
 
Join Date: Sep 2005
Location: Miranda's house
Posts: 13,409
vCash: 500
I'm not sure what the big deal is. If a player is listed as +7 or as 20/13 what's the difference? It's the same friggin stat except you only have to look at one number not two. Also, a team's power play and penalty kill percentages are directly correlated to the players out on the ice to some degree so you don't need a plus/minus for those. If Edmonton's penalty killing % is 85% you can make a value judgment on how good Peca does on the penalty kill to not warrant another stat for him.

UnderratedBrooks44 is online now  
Old
10-20-2005, 12:36 PM
  #12
MojoJojo
Registered User
 
MojoJojo's Avatar
 
Join Date: Jan 2003
Location: Philadelphia
Posts: 9,351
vCash: 500
Quote:
Originally Posted by tmg
A stat I've suggested before, along the vein of +/- purely for the purpose of measuring a player's relative defensive reliability, is a "player GAA". (Nonpowerplay goals against player's team while said player is on the ice)*60/(nonpowerplay minutes played by player).

Basically, how many goals are scored against your team while you're on the ice, per 60 minutes of icetime?

It's not perfect, no... but I think it's be more revealing than plus-minus, at least in terms of defensive specialists. If this stat were back-calculated on the past few seasons, I wonder how much correlation there would be between the strongest numbers by this stat, and the Selke voting.
Its good, though I would make one more adjustment: take into consideration the strength of the goalie. Take that number and multiply it by save % and divide by average goals against for the teams goaltending average.

If you give up lots of bad turnovers or cant clear the puck, etc. A good goalie can make your stats look better, and vice versa.

MojoJojo is offline  
Old
10-20-2005, 02:33 PM
  #13
eye
Registered User
 
Join Date: Feb 2003
Location: around the 49th para
Posts: 1,607
vCash: 500
Quote:
Originally Posted by UnderratedBrooks44
I'm not sure what the big deal is. If a player is listed as +7 or as 20/13 what's the difference? It's the same friggin stat except you only have to look at one number not two. Also, a team's power play and penalty kill percentages are directly correlated to the players out on the ice to some degree so you don't need a plus/minus for those. If Edmonton's penalty killing % is 85% you can make a value judgment on how good Peca does on the penalty kill to not warrant another stat for him.
I don't get what is so difficult to understand. Seperate stat categories would be much more relevent and useful than the +/- stat they use now. The +/- stat would become meaningful.

e.g you ask what's the difference between showing a player at +7 and ES 20/13. On it's own +7 means nothing to me. Was the player on the ice for 53 goals for and 46 against which might indicate some defensive flaws and good offensive abilities or was he on for 7 and 0 against which shows virtually no offense but great defensive play? Was his linemate or defense partner on for more goals for or less? There is no way of knowing by the current way of reporting +/-. By showing goals for and goals against in the ES category and the same for PP and PK then you can have a better idea if a player is a defensive liability or an offensive force, whether he is a good PP or PK player. Right now ES, PK, PP and goalie pulled stats are all combined into one which again in my mind is uesless.

There are much better ways to keep statistical data on players than the current +/- stat and that's all I'm saying. Let's bring the game into this century and use all the technical equipment we now have at our disposal to rate and score players accurately and in a meaningful way. IMO, Only old school hockey people believe that +/- stat is useful as is.

eye is offline  
Old
10-20-2005, 02:42 PM
  #14
lemieux32*
 
lemieux32*'s Avatar
 
Join Date: Aug 2005
Country: United States
Posts: 1,280
vCash: 500
Quote:
Originally Posted by lemieux32
It doesn't seem like any of these suggestions solve the real problem: giving someone a plus or a minus when he is uninvolved or responsible in any way. Such as the guy who just came on the ice and has nothing to do with the play. Or the winger sticking to his point man and the puck never comes there but the other team scores.
Why has nobody addresed this? I don't mean my post but the issue I mention. That is the biggest problem with +/-.

lemieux32* is offline  
Old
10-20-2005, 02:43 PM
  #15
Lionel Hutz
Registered User
 
Lionel Hutz's Avatar
 
Join Date: Apr 2004
Location: Locking the Lounge??
Country: Canada
Posts: 13,054
vCash: 500
The problem with +/-, as I see it, is that the stat is skewed by the team the player plays for. Take Ray Bourque in his prime and put him on a line with 4 scrubs and a goaltender with swiss cheese for padding and his +/- will likely be bad regardless of the quality of his individual defence.

So, to remove this it would have to somehow be measured relative to the team. But, isolating this factor would remove objectivity and make the stat equally worthless in the opposite direction.

IMO, this is the greater weakness than PP/PK situations etc.

Lionel Hutz is offline  
Old
10-20-2005, 02:43 PM
  #16
sparr0w
Registered User
 
sparr0w's Avatar
 
Join Date: Apr 2005
Posts: 16,377
vCash: 500
I'd like to see the change in stat presentation as well.

Plus one other change. If you're in the box for a PP goal against you should get that -1.

sparr0w is offline  
Old
10-20-2005, 02:45 PM
  #17
Lionel Hutz
Registered User
 
Lionel Hutz's Avatar
 
Join Date: Apr 2004
Location: Locking the Lounge??
Country: Canada
Posts: 13,054
vCash: 500
Quote:
Originally Posted by lemieux32
Why has nobody addresed this? I don't mean my post but the issue I mention. That is the biggest problem with +/-.
It is. But I don't think there is a way to objectively measure that stat. This is the problem.

Lionel Hutz is offline  
Old
10-20-2005, 03:07 PM
  #18
DMVille
Registered User
 
Join Date: Sep 2005
Posts: 73
vCash: 500
I wonder what something like (Even Strength Points/ (Even Strength Goals against/5)) would look like. Only get positive credit for goals you materially participated in and get dinged only 1/5 for every goal against, figuring all 5 guys are not always responsible for a goal being scored against.

DMVille is offline  
Old
10-20-2005, 03:39 PM
  #19
Vote Quimby
Registered User
 
Join Date: Apr 2005
Location: Langley BC
Country: Portugal
Posts: 180
vCash: 500
Quote:
Originally Posted by DMVille
I wonder what something like (Even Strength Points/ (Even Strength Goals against/5)) would look like. Only get positive credit for goals you materially participated in and get dinged only 1/5 for every goal against, figuring all 5 guys are not always responsible for a goal being scored against.
During the course of the season, it all evens out. A player will get a minus even though he had his player in check but will also get a plus even though he just stepped on the ice. I agree the stat is far from perfect but the only way to get an accurate representation of a players offensive / defensive contributions is to come up with a formula along the lines of NFL QB ratings. Even then, it is too difficult to come up with a black and white number that shows a players contribution. I can guarantee that if someone did come up with a formula, there will be a number of players excluded or included that shouldn't be and I think that is why no one has bothered to change the current plus/minus stat. There are just too many variables and situations that would affect this stat.

Vote Quimby is offline  
Old
10-20-2005, 04:12 PM
  #20
Troy McClure
Registered User
 
Troy McClure's Avatar
 
Join Date: Mar 2002
Location: The KlINGberg
Country: Switzerland
Posts: 27,464
vCash: 500
Quote:
Originally Posted by lemieux32
It doesn't seem like any of these suggestions solve the real problem: giving someone a plus or a minus when he is uninvolved or responsible in any way.
And that would be such a subjective measure, that it would never be right. Even now, there is no set definition of giveaways or takeaways.

As for the RS 23/4 PP 12/3 PK 10/34 idea of stat tracking, I don't like it because I don't know what it really proves. I think +/- as it is now does a better job of penalizing short handed goals against.

You can extract powerplay contribution by looking at things like power play goals, assists, and ice time.

Troy McClure is offline  
Old
10-20-2005, 06:00 PM
  #21
Hockey Outsider
Registered User
 
Hockey Outsider's Avatar
 
Join Date: Jan 2005
Country: Canada
Posts: 3,454
vCash: 500
Quote:
Originally Posted by tmg
A stat I've suggested before, along the vein of +/- purely for the purpose of measuring a player's relative defensive reliability, is a "player GAA". (Nonpowerplay goals against player's team while said player is on the ice)*60/(nonpowerplay minutes played by player).

Basically, how many goals are scored against your team while you're on the ice, per 60 minutes of icetime?

It's not perfect, no... but I think it's be more revealing than plus-minus, at least in terms of defensive specialists. If this stat were back-calculated on the past few seasons, I wonder how much correlation there would be between the strongest numbers by this stat, and the Selke voting.
This is a project I've been working on over the last few months. I have the results for ES and SH seperately, but I'll just show ES here. Note that there is still a lot of statistical noise from a player's teammates and goalies. And, clearly, there are VERY big surprises in most of the results.

2004: top even-strength GAA, >500 minutes
Andreas Dackell MON F 60 750 11 0.88
Ben Clymer TB F 66 611 10 0.98
Marty Murray CAR F 66 608 10 0.99
Pierre Turgeon DAL F 76 876 16 1.10
Chris Neil OTT F 82 704 13 1.11
Jon Klemm DAL D 58 840 17 1.21
Petr Cajanek STL F 70 889 18 1.21
Don Sweeney DAL D 63 888 18 1.22
Jason Ward MON F 53 541 11 1.22
Pierre Dagenais MON F 50 583 12 1.23
Shaun Van Allen OTT F 73 627 13 1.24
David Hale NJD D 65 896 19 1.27
Tommy Albelin NJD D 45 601 13 1.30
Mark Mowers DET F 52 501 11 1.32
Kirk Maltby DET F 79 1,035 23 1.33
Curtis Leschyshyn OTT D 56 717 16 1.34
Nils Ekman SJ F 82 968 23 1.43
Mike Ricci SJ F 71 842 20 1.43
Hal Gill BOS D 82 1,250 30 1.44
Glen Wesley CAR D 74 1,248 30 1.44
Jim Dowd MIN F 55 618 15 1.46
Andrei Nikolishin COL F 49 612 15 1.47
Igor Ulanov EDM D 42 612 15 1.47
Justin Williams PHI F 47 606 15 1.48
Mattias Timander PHI D 34 525 13 1.49
Erik Rasmussen NJD F 69 758 19 1.50
Nolan Pratt TB D 58 837 21 1.51
Henrik Sedin VAN F 76 836 21 1.51
Teemu Selanne COL F 78 994 25 1.51
Stu Barnes DAL F 77 1,112 28 1.51

2003: top even-strength GAA, >500 minutes
Shjon Podein STL F 68 541 9 1.00
Simon Gagne PHI F 46 580 11 1.14
Brian Sutherby WAS F 72 566 11 1.17
Rob DiMaio DAL F 69 745 15 1.21
Igor Ulanov FLA D 56 836 17 1.22
Vaclav Varada BUF F 44 535 12 1.34
Justin Williams PHI F 41 534 12 1.35
Alyn McCauley TOR F 64 668 15 1.35
Eric Desjardins PHI D 79 1,319 30 1.37
Niko Kapanen DAL F 82 966 22 1.37
Serge Aubin COL F 66 701 16 1.37
Brad Ference FLA D 60 830 19 1.37
Dan Hinote COL F 60 566 13 1.38
Jere Lehtinen DAL F 80 1,088 25 1.38
Darryl Sydor DAL D 81 1,120 26 1.39
Steve Thomas CHI F 69 815 19 1.40
Tom Fitzgerald TOR F 66 630 15 1.43
Marty Reasoner EDM F 70 754 18 1.43
Turner Stevenson NJD F 77 869 21 1.45
Stephane Robidas DAL D 76 944 23 1.46
Dainius Zubrus WAS F 63 779 19 1.46
Tyson Nash STL F 66 533 13 1.46
Boris Mironov NYR D 36 572 14 1.47
Jay Pandolfo NJD F 68 917 23 1.50
Pierre-Marc Bouchard MIN F 50 518 13 1.51
Jaroslav Svoboda CAR F 48 550 14 1.53
Scott Young DAL F 79 1,019 26 1.53
Chris Therien PHI D 67 977 25 1.53
Radovan Somik PHI F 60 777 20 1.55
Dan Cleary EDM F 57 619 16 1.55

2002: top even-strength GAA, >500 minutes
2002 Detroit Red Wings DET L Kirk Maltby 82 17 893.0 1.14
2002 New Jersey Devils NJD L Jay Pandolfo 65 15 729.3 1.23
2002 Toronto Maple Leafs TOR C Alyn McCauley 82 15 658.5 1.37
2002 Colorado Avalanche COL R Brian Willsie 56 13 569.0 1.37
2002 Philadelphia Flyers PHI R Paul Ranheim 79 13 563.3 1.38
2002 Colorado Avalanche COL C Dan Hinote 58 14 588.7 1.43
2002 Ottawa Senators OTT L Chris Herperger 72 16 659.5 1.46
2002 Edmonton Oilers EDM R Georges Laraque 80 18 740.0 1.46
2002 Edmonton Oilers EDM D Scott Ferguson 50 15 616.5 1.46
2002 Buffalo Sabres BUF L Denis Hamel 61 14 572.2 1.47
2002 New Jersey Devils NJD D Tommy Albelin 42 13 520.4 1.50
2002 Tampa Bay Lightning TB R Jimmie Olvestad 74 22 871.0 1.52
2002 San Jose Sharks SAN D Mike Rathje 52 21 828.9 1.52
2002 New Jersey Devils NJD L John Madden 82 26 1023.4 1.52
2002 Edmonton Oilers EDM L Josh Green 61 15 587.4 1.53
2002 San Jose Sharks SAN R Todd Harvey 69 17 655.5 1.56
2002 Anaheim Mighty Ducks ANA R Dan Bylsma 77 17 649.1 1.57
2002 New Jersey Devils NJD L Sergei Brylin 76 28 1066.3 1.58
2002 Detroit Red Wings DET R Darren McCarty 62 18 684.5 1.58
2002 St. Louis Blues STL L Sergei Varlamov 52 14 528.3 1.59
2002 Chicago Blackhawks CHI D Steve Poapst 56 20 740.9 1.62
2002 St. Louis Blues STL C Jamal Mayers 77 22 797.0 1.66
2002 Ottawa Senators OTT R Chris Neil 72 16 577.4 1.66
2002 Edmonton Oilers EDM D Jason Smith 74 36 1283.2 1.68
2002 Detroit Red Wings DET D Chris Chelios 79 41 1454.4 1.69
2002 St. Louis Blues STL D Jeff Finley 78 31 1095.9 1.70
2002 Edmonton Oilers EDM L Ethan Moreau 80 24 846.4 1.70
2002 Los Angeles Kings LA R Brad Chartrand 46 15 525.8 1.71
2002 Ottawa Senators OTT R Bill Muckalt 70 17 588.7 1.73
2002 Detroit Red Wings DET R Kris Draper 82 30 1038.1 1.73

2001: top even-strength GAA, >500 minutes
2001 Detroit Red Wings DET C Kris Draper 75 13 825 0.95
2001 Vancouver Canucks VAN D Jason Strudwick 60 11 556 1.19
2001 Detroit Red Wings DET L Kirk Maltby 79 18 886 1.22
2001 Washington Capitals WAS R Joe Sacco 69 14 606 1.39
2001 Colorado Avalanche COL D Jon Klemm 78 29 1,238 1.41
2001 Dallas Stars DAL D Brad Lukowich 80 26 1,052 1.48
2001 New Jersey Devils NJD R Turner Stevenson 69 18 704 1.53
2001 Colorado Avalanche COL D Greg de Vries 79 31 1,211 1.54
2001 Dallas Stars DAL R Roman Lyashenko 60 14 547 1.54
2001 Ottawa Senators OTT L Magnus Arvedson 51 18 680 1.59
2001 Atlanta Thrashers ATL L Denny Lambert 67 16 598 1.61
2001 San Jose Sharks QUE C Vincent Damphousse 45 16 594 1.62
2001 Colorado Avalanche COL R Dan Hinote 76 19 691 1.65
2001 Detroit Red Wings DET D Todd Gill 68 28 1,010 1.66
2001 Columbus Blue Jackets CLB D Jamie Heward 69 18 639 1.69
2001 Toronto Maple Leafs TOR R Tie Domi 82 19 669 1.70
2001 San Jose Sharks QUE C Todd Harvey 69 20 700 1.71
2001 St. Louis Blues STL D Todd Reirden 38 16 559 1.72
2001 Nashville Predators NAS R Rob Valicevic 60 20 698 1.72
2001 Buffalo Sabres BUF D James Patrick 54 20 698 1.72
2001 Toronto Maple Leafs TOR D Danny Markov 59 25 873 1.72
2001 Buffalo Sabres BUF L Doug Gilmour 71 27 941 1.72
2001 Phoenix Coyotes PHO C Michal Handzus 46 15 520 1.73
2001 New Jersey Devils NJD L John Madden 80 27 933 1.74
2001 San Jose Sharks QUE D Bryan Marchment 75 33 1,137 1.74
2001 Philadelphia Flyers PHI R Jody Hull 71 22 757 1.74
2001 Dallas Stars DAL L Mike Keane 67 24 817 1.76
2001 Philadelphia Flyers PHI D Michal Sykora 49 21 697 1.81
2001 Detroit Red Wings DET D Steve Duchesne 54 24 782 1.84
2001 Philadelphia Flyers PHI L Paul Ranheim 80 26 844 1.85

2000: top even-strength GAA, >500 minutes
2000 Washington Capitals WAS C Jeff Halpern 79 16 798.7 1.20
2000 Colorado Avalanche COL R Shjon Podein 75 16 793.5 1.21
2000 Philadelphia Flyers PHI L Craig Berube 77 13 609.1 1.28
2000 Carolina Hurricanes CAR L Bates Battaglia 77 24 1074.9 1.34
2000 Philadelphia Flyers PHI R Jody Hull 67 14 612.4 1.37
2000 Dallas Stars DAL R Blake Sloan 67 18 775.2 1.39
2000 St. Louis Blues STL L Tyson Nash 66 13 555.1 1.41
2000 Buffalo Sabres BUF L Vladimir Tsyplakov 63 17 712.5 1.43
2000 Dallas Stars DAL C Guy Carbonneau 69 21 854.2 1.48
2000 New York Islanders NYI R Bill Muckalt 45 13 522.0 1.49
2000 Edmonton Oilers EDM R Georges Laraque 76 16 639.9 1.50
2000 Colorado Avalanche COL D Jon Klemm 73 27 1078.2 1.50
2000 St. Louis Blues STL L Jochen Hecht 63 20 797.6 1.50
2000 Buffalo Sabres BUF D Rhett Warrener 61 26 998.6 1.56
2000 St. Louis Blues STL D Marc Bergevin 81 36 1349.5 1.60
2000 Philadelphia Flyers PHI D Dan McGillis 68 30 1117.9 1.61
2000 Philadelphia Flyers PHI D Luke Richardson 74 27 1000.5 1.62
2000 St. Louis Blues STL C Pierre Turgeon 52 21 758.7 1.66
2000 Anaheim Mighty Ducks ANA R Jeff Nielsen 79 23 827.9 1.67
2000 St. Louis Blues STL D Jeff Finley 74 32 1135.9 1.69
2000 San Jose Sharks QUE C Mike Ricci 82 29 1017.6 1.71
2000 Colorado Avalanche COL C Dave Reid 65 22 769.0 1.72
2000 St. Louis Blues STL L Lubos Bartecko 67 23 802.7 1.72
2000 St. Louis Blues STL D Chris Pronger 79 44 1529.4 1.73
2000 Edmonton Oilers EDM C Jim Dowd 69 19 648.6 1.76
2000 Buffalo Sabres BUF R Vaclav Varada 76 30 1014.6 1.77
2000 St. Louis Blues STL D Al MacInnis 61 32 1068.7 1.80
2000 Montreal Canadiens MON L Benoit Brunet 50 20 665.5 1.80
2000 Dallas Stars DAL D Jamie Pushor 62 21 692.5 1.82
2000 Philadelphia Flyers PHI L Simon Gagne 80 26 856.8 1.82

1999: top even-strength GAA, >500 minutes
1999 Ottawa Senators OTT R Bruce Gardiner 59 13 602.4 1.29
1999 Detroit Red Wings DET D Mathieu Dandenault 75 24 1043.3 1.38
1999 Buffalo Sabres BUF D Jay McKee 72 28 1175.8 1.43
1999 Buffalo Sabres BUF D James Patrick 45 13 543.6 1.43
1999 Toronto Maple Leafs TOR D Alexander Karpovtsev 58 22 888.6 1.49
1999 Phoenix Coyotes PHO L Juha Ylonen 59 17 682.6 1.49
1999 Phoenix Coyotes PHO D Gerald Diduck 44 17 673.6 1.51
1999 New Jersey Devils NJD L Sergei Brylin 47 14 525.5 1.60
1999 Washington Capitals WAS L Kelly Miller 62 15 549.9 1.64
1999 Los Angeles Kings LA C Ian Laperriere 72 18 658.8 1.64
1999 Boston Bruins BOS D Don Sweeney 81 39 1407.0 1.66
1999 Ottawa Senators OTT R Magnus Arvedson 80 32 1137.6 1.69
1999 Pittsburgh Penguins PIT L Aleksey Morozov 67 21 740.4 1.70
1999 Buffalo Sabres BUF C Curtis Brown 78 29 1020.2 1.71
1999 Ottawa Senators OTT D Janne Laukkanen 50 22 768.5 1.72
1999 Pittsburgh Penguins PIT D Darius Kasparaitis 48 19 659.0 1.73
1999 Anaheim Mighty Ducks ANA C Antti Aalto 73 18 622.7 1.73
1999 Ottawa Senators OTT L Marian Hossa 60 21 726.0 1.74
1999 Philadelphia Flyers PHI C Marc Bureau 71 18 619.1 1.74
1999 Ottawa Senators OTT D Patrick Traverse 46 17 583.7 1.75
1999 Buffalo Sabres BUF L Michal Grosek 76 30 1021.4 1.76
1999 Carolina Hurricanes CAR D Nolan Pratt 61 26 883.9 1.76
1999 Dallas Stars DAL R Pat Verbeek 78 27 914.2 1.77
1999 Buffalo Sabres BUF D Darryl Shannon 71 33 1115.4 1.78
1999 Ottawa Senators OTT D Sami Salo 61 27 911.3 1.78
1999 Dallas Stars DAL D Shawn Chambers 61 24 797.9 1.80
1999 Buffalo Sabres BUF L Miroslav Satan 81 35 1163.2 1.81
1999 Carolina Hurricanes CAR C Jeff O'Neill 75 30 989.3 1.82
1999 Montreal Canadiens MON R Jason Dawe 59 18 588.8 1.83
1999 Anaheim Mighty Ducks ANA R Ted Drury 75 16 521.3 1.84

* Data from 2003 and 2004 is from Alan Ryder's work.

I'm pretty sure there aren't any errors here but this is a ton of data, so I can't make any guarantees.

I've also done some historical numbers for this and Bobby Clarke and Bobby Orr are so much better than the rest of the league, it's ridiculous.

Hockey Outsider is offline  
Old
10-20-2005, 06:03 PM
  #22
dwkdnvr
Registered User
 
Join Date: Mar 2004
Posts: 398
vCash: 500
This is somewhat interesting, as it starts to get into the more general question of secondary or derived stats. FootballOutsiders is doing some very interesting stuff in this regard for the NFL, and I hope that we'll see more in hockey.

Obviously the main problem is that derived stats are merely 'indicative' rather than 'factual', and so there are always going to be problems. Still, they can prove to be interesting, if not really useful.

Off the top of my head, for +/- I think the most interesting first-order comparison would be to 'normalize' according to the strength of the team and possibly by ice-time. eg - give the player a % number indicating the fraction of the overall team +/- value (but arrange the sign so that folks above the team average are '+', even if their absolute ranking is -). ie (player - avg)/sum where sum is the total team +/- value and avg is the average for each player on that team.
The TOI normalization would be interesting as well, as someone who goes -10 in only 5 mins/game is probably more of a problem than someone that goes -10 in 25 mins/game. Unfortunately, TOI isn't broken down by even strength vs odd-man for either individuals or teams, so that would be a tough metric to compute.

The next level of this is probably not possible with the current stats that are published, but it would be to weight *every* + or - entry by the strength of the opposition players that were on the ice at the time of the goal. ie getting a + against a bad Pittsburgh defence is worth less than a + against Hasek/Ott. A start would be to simply weight it by a simple factor of the teams (GF-GA) value, or possibly something related to the point total of the goal scorer

This may be a dumb question, but is there an 'obvious' place to download stats in text/CSV format? There must be to support fantasy leagues somewhere.

dwkdnvr is offline  
Old
10-20-2005, 06:07 PM
  #23
eye
Registered User
 
Join Date: Feb 2003
Location: around the 49th para
Posts: 1,607
vCash: 500
Quote:
Originally Posted by Troy McClure
And that would be such a subjective measure, that it would never be right. Even now, there is no set definition of giveaways or takeaways.

As for the RS 23/4 PP 12/3 PK 10/34 idea of stat tracking, I don't like it because I don't know what it really proves. I think +/- as it is now does a better job of penalizing short handed goals against.

You can extract powerplay contribution by looking at things like power play goals, assists, and ice time.
There is no perfect statistical method to track a players performance but there are much better ways than what we currently use. If a player is on the ice for a short handed goal against how can you determine that just from looking at a player's plus minus? Similarly, how would you know if a player was on the ice in the last minute with the goalie pulled and his team gave up an empty net goal? It could happen a half dozen times a year to the last minute players in a close game. Open up your minds people.

eye is offline  
Old
10-20-2005, 06:24 PM
  #24
dwkdnvr
Registered User
 
Join Date: Mar 2004
Posts: 398
vCash: 500
Quote:
Originally Posted by Hockey Outsider
This is a project I've been working on over the last few months.
Our posts crossed in flight, which is bizarre - given what you're doing and your name, you must be aware of FootballOutsiders, right?

I like the GVA analysis on your site, although I'd like to know more about the details of the computations.

dwkdnvr is offline  
Old
10-20-2005, 06:36 PM
  #25
Hockey Outsider
Registered User
 
Hockey Outsider's Avatar
 
Join Date: Jan 2005
Country: Canada
Posts: 3,454
vCash: 500
Quote:
Originally Posted by dwkdnvr
Our posts crossed in flight, which is bizarre - given what you're doing and your name, you must be aware of FootballOutsiders, right?

I like the GVA analysis on your site, although I'd like to know more about the details of the computations.
Yes, I've heard of Football Outsiders. I've never been a big fan of football so I don't spend a lot of time looking at their site (and when I do a lot of it goes over my head, given my lack of knowledge about football). With that said, I like any sort of sports research/analysis.

Thanks, glad you like the GVA analysis. I explain GVA in nauseating detail here (http://www.geocities.com/thehockeyoutsider/gva2.pdf), using the 2004 season as a walk-through example. I know it's far from perfect and I'm definitely open to helpful suggestions, but I think it has some potential.

Hockey Outsider is offline  
Closed Thread

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 02:17 PM.

monitoring_string = "e4251c93e2ba248d29da988d93bf5144"
Contact Us - HFBoards - Archive - Privacy Statement - Terms of Use - Advertise - Top - AdChoices

vBulletin Copyright ©2000 - 2015, Jelsoft Enterprises Ltd.
HFBoards.com is a property of CraveOnline Media, LLC, an Evolve Media, LLC company. 2015 All Rights Reserved.