HFBoards

Go Back   HFBoards > NHL Western Conference > Pacific Division > Edmonton Oilers
Notices

NHL Standings, Luck and Coincidence

Closed Thread
 
Thread Tools
Old
12-08-2003, 07:58 AM
  #1
igor*
 
Join Date: Mar 2002
Posts: 3,276
vCash: 500
NHL Standings, Luck and Coincidence

Luck and Coincidence really factor into the results of a team, especially over the short term. I think we can all agree on that.

Sometimes the one-timer finds the top corner of the net ... sometimes it doesn't. Sometimes the rebound lands right on an Oiler stick, sometimes it bounces the other way. Usually the shooter is just aiming for the net, or at best a quadrant of the net ... sometimes the puck hits the inside of the post and goes in, sometimes it doesn't. Good players have better success rates ... but luck still plays a big factor on a game by game basis. That's just common sense IMO.

Over a few games luck can really play in to a streak. And getting the bounces at the right times makes a heckuva difference. Get a lucky bounce or call when you're tied in the third ... it can make a 1 or 2 point difference in the standings. (SEE CBJ goal called back earlier, or Cross one-timer into the top corner vs PIT). Have the same good fortune when you're leading or trailing by a few goals ... doesn't affect the result.

The surprising thing is how much luck can play into things even over the course of a full season. Intuitively 82 games would seem to be enough to make it balance out ... but it really isn't. Baseball fans probably know that Bill James is the guy who really highlighted this with his 'Pythagorean Theory' for baseball ... it's crude but illustrates the point. And it applies to other sports as well.

Mudcrutch pointed out a couple weeks ago the the Oilers results (3 points over .500 at the time) were a bit flattering, considering they had only outscored opponents by one goal at that point. He had a really valid point IMO.

The numbers below are 'luck factor' for NHL teams so far. They don't mean anything specific, they are an indicator of how lucky a team has been, of how much better or worse their results have been than should be expected looking at schedule difficulty and team GF minus team GA.

This 'luck factor' is the Sagarin PREDICTOR number minus the ELO CHESS number and times 100. The bigger the number ... the luckier you've been. The totals of all the teams 'luck factors' add up to zero of course, which makes sense.

Anyhoo, the NHL teams so far this year:

Pittsburgh Penguins _____ 35
St. Louis Blues _________ 27
Nashville Predators _____ 23
Phoenix Coyotes _________ 22
New Jersey Devils _______ 20
Edmonton Oilers _________ 18
Philadelphia Flyers _____ 18
Toronto Maple Leafs _____ 18
Boston Bruins ___________ 15
Buffalo Sabres __________ 13
Dallas Stars ____________ 13
Tampa Bay Lightning _____ 8
Colorado Avalanche ______ 4
Calgary Flames __________ 4
Florida Panthers ________ 4
Columbus Blue Jackets ___ 3
Atlanta Thrashers _______ 2
Montreal Canadiens ______ 2
Carolina Hurricanes _____ 2
San Jose Sharks _________ 1
Anaheim Mighty Ducks ____ -2
Los Angeles Kings _______ -9
Vancouver Canucks _______ -14
Chicago Blackhawks ______ -15
Minnesota Wild __________ -20
New York Islanders ______ -21
New York Rangers ________ -28
Washington Capitals _____ -28
Detroit Red Wings _______ -42
Ottawa Senators _________ -73


.

.

And for last season: 2002/2003

Atlanta Thrashers _______ 26
Anaheim Mighty Ducks ____ 20
Toronto Maple Leafs _____ 13
Pittsburgh Penguins _____ 13
Vancouver Canucks _______ 12
Calgary Flames __________ 12
Tampa Bay Lightning _____ 11
New Jersey Devils _______ 8
Philadelphia Flyers _____ 6
New York Islanders ______ 5
Minnesota Wild __________ 5
Ottawa Senators _________ 4
Edmonton Oilers _________ 4
Washington Capitals _____ 4
Columbus Blue Jackets ___ 4
Florida Panthers ________ 3
Phoenix Coyotes _________ 0
New York Rangers ________ -4
Carolina Hurricanes _____ -6
Boston Bruins ___________ -7
Chicago Blackhawks ______ -7
Detroit Red Wings _______ -8
Los Angeles Kings _______ -8
Montreal Canadiens ______ -8
St. Louis Blues _________ -9
Nashville Predators _____ -17
Dallas Stars ____________ -18
Buffalo Sabres __________ -18
Colorado Avalanche ______ -19
San Jose Sharks _________ -19


.

So, is that about what you would have guessed?

igor* is offline  
Old
12-08-2003, 08:49 AM
  #2
rabi_sultan
Registered User
 
Join Date: Sep 2002
Location: London, England
Country: United Kingdom
Posts: 3,782
vCash: 500
Send a message via ICQ to rabi_sultan Send a message via AIM to rabi_sultan Send a message via MSN to rabi_sultan Send a message via Yahoo to rabi_sultan
i'm not surprised by teams like Detroit who basically manufacture their goals, the Oilers as we all know are streaky which means that luck plays a big way but the Sens this year, OUCH! As for the habs i thought they would be 'luckier' as they too play streaky (jmo) and interesting that the Penguins have been extremely lucky two years in a row could it be because they are a poor team with poor GA and GF with all the mathematics included that its been stated they are luckier when they aren't?

I don't know just throwin something useless out there coz perhaps when the pens lose they lose big otherwise when they win they win by a small score. But if they win enough then they seem luckier (right? or have i got the wrong end of the stick here?)

rabi_sultan is offline  
Old
12-08-2003, 09:23 AM
  #3
igor*
 
Join Date: Mar 2002
Posts: 3,276
vCash: 500
Quote:
Originally Posted by rabi_sultan
...
perhaps when the pens lose they lose big otherwise when they win they win by a small score. But if they win enough then they seem luckier (right? or have i got the wrong end of the stick here?)
Yup, that's true. They've been bad and lucky the past couple of years. But lots of terrible teams are ALSO unlucky. I think it is simple coincidence that they have been lucky two years in a row.

S.J were just a bit unlucky to have not gotten better results last year, that's the way the cookie crumbles sometimes.

The year before last S.J had pretty good luck, and CHI (finishing 5th in the west that year) had remarkable luck by this measure.

I used to always think that 'everything happens for a reason' ... but sometimes maybe it doesn't. Some teams have lucky years (relative to the points totals they should really have gotten), and sometimes it goes the other way. Doesn't seem to be any rhyme or reason to it. It's just coincidence I think.

igor* is offline  
Old
12-08-2003, 09:41 AM
  #4
PineJockey
Registered User
 
Join Date: Oct 2002
Location: E-Town
Posts: 616
vCash: 500
Igor,

I generally like your posts and their statistical bent, but unless you explain what exactly you mean by the Sagarin PREDICTOR, the ELO CHESS number, plus how you rank schedule difficulty, this post is difficult to evaluate.

PineJockey is offline  
Old
12-08-2003, 09:50 AM
  #5
serum114
Registered User
 
Join Date: Mar 2002
Location: Calgary
Country: Canada
Posts: 1,766
vCash: 500
Quote:
Originally Posted by rabi_sultan
interesting that the Penguins have been extremely lucky two years in a row could it be because they are a poor team with poor GA and GF with all the mathematics included that its been stated they are luckier when they aren't?
With the Pens I think anytime a number appears in the W column this year, it is directly attributed to luck.

After watching what I dubbed an "embarassing win" by the Oilers Saturday, I saw the Pens play (sans Fleury, the jerks) and get burned three times by Shean Donovan en route to a 6-1 loss. Shean Donovan!

The Pens baffle me in that they've won six games this year, I think they may actually be one of the worst teams of all time, if not THE worst.

serum114 is offline  
Old
12-08-2003, 11:22 AM
  #6
oilswell
Registered User
 
Join Date: Mar 2002
Location: Wandering the globe
Posts: 1,760
vCash: 500
Quote:
Originally Posted by serum114
The Pens baffle me in that they've won six games this year, I think they may actually be one of the worst teams of all time, if not THE worst.
There could be a case for this, although I guess if Mario would be playing it might not be so bad.

They're on pace for 19 wins and about 53 points. According to Budd Bailey's Hockey Abstract, the worst team might just be the '53-54 Blackhawks who had a 12-51-7 mark good for 31 points...and they were 37 points back of the 2nd last place finisher. 37 points! Prorated to the current 82 game schedule, the Hawks woul dhave had a record of 14-60-14, which would make them worse than this years Penguins, if they keep the same "pace".

The thing is, Washington has just one more point in the standings and the Penguins have a game in hand, and Chicago has played more games than the Pens and have the same number of wins (the Hawks are on pace for 58 points). Even if the Penguins continue to flounder like they do, if Washington doesn't pick up the pace, this could be a really pathetic year for the NHL.

oilswell is offline  
Old
12-08-2003, 12:32 PM
  #7
igor*
 
Join Date: Mar 2002
Posts: 3,276
vCash: 500
Quote:
Originally Posted by PineJockey
Igor,

I generally like your posts and their statistical bent, but unless you explain what exactly you mean by the Sagarin PREDICTOR, the ELO CHESS number, plus how you rank schedule difficulty, this post is difficult to evaluate.
Sagarin is a sports handicapper who publishes numbers at USAToday.com
http://www.usatoday.com/sports/sagarin/nhl0304.htm

Basically these are gambling tools. The theory is always that past performance is the best predictor of future performance.

ELO CHESS: Basically this is the winning percentage of your team, corrected for your schedule.

e.g.: If you are a .500 team, and you've had a brutal schedule where your opponents, on average, were .550 winning% teams --> then you would have a .550 rating.
Makes sense, no?

PREDICTOR (Or Ballantine Number): Age old and much, much more reliable for predicting the future outcomes of games. This relies entirely on the goal differential of your team. Your team goals-for minus your team goals-against. Again corrected for the difficulty of your schedule.

e.g.: If, after 30 games, you have outscored the opposition by 15 goals
overall (therefore outscored them by .5 goals-per-game on average) then you would have a +.5 rating.
If you had a soft schedule over this period, say the oppositon you played averaged a -.2 rating. Then your corrected PREDICTOR rating would be (+.5) + (-.2) = +.3

.

They'll multiply this stuff by constants ... so that you can use the difference between two team's PREDICTOR numbers to figure out the point spread in a game.

e.g.: If your team has +.3 rating, and the opponent has a -.2 rating ... the spread should be +.5 goals.

.

THE INTERESTING THING is that the record of your team, be it good or bad, is a pretty poor predictor of future results. And historically the goal differential of your team (i.e. GF - GA) is a very good predictor of future performance.

So if you have a better record than your goal differential suggests you should have ... you've had a spot of luck on your side.


Last edited by igor*: 12-08-2003 at 12:39 PM.
igor* is offline  
Old
12-08-2003, 01:06 PM
  #8
PineJockey
Registered User
 
Join Date: Oct 2002
Location: E-Town
Posts: 616
vCash: 500
Thanks Igor,

That is helpful. Just out of curiosity, do the Predictor and ELO Chess numbers take into account the "rank" of your opponent on the day that you actually played, or the "ranking" that they have on the day that Sargarin is calculating his tables? I assume it is the former, but I couldn't find any confirmation of that anywhere. In other words, if you catch a team on a hot streak, the streak should (in part at least) determine their "rank" rather than simply looking retroactively at their overall record, goals for and against, game location, etc.

EDIT: Looking at his tables again, I see that there is a RECENT column that weights recent games more than the pure RATING column does. How are recent games (i.e., streaks) accomodated in the ELO and Predictor?


Last edited by PineJockey: 12-08-2003 at 01:11 PM.
PineJockey is offline  
Old
12-08-2003, 01:27 PM
  #9
igor*
 
Join Date: Mar 2002
Posts: 3,276
vCash: 500
Quote:
Originally Posted by PineJockey
Thanks Igor,

That is helpful. Just out of curiosity, do the Predictor and ELO Chess numbers take into account the "rank" of your opponent on the day that you actually played, or the "ranking" that they have on the day that Sargarin is calculating his tables? I assume it is the former, but I couldn't find any confirmation of that anywhere. In other words, if you catch a team on a hot streak, the streak should (in part at least) determine their "rank" rather than simply looking retroactively at their overall record, goals for and against, game location, etc.

EDIT: Looking at his tables again, I see that there is a RECENT column that weights recent games more than the pure RATING column does. How are recent games (i.e., streaks) accomodated in the ELO and Predictor?
I assume the schedule correction thing on Sagarin's site is based on overall results, but I don't know ... it would just be way too complicated to do it any other way though.

The 'recent games' thing is from a few years ago, no? I don't think he does that any more. Sagarin just uses pure math their, he doesn't take into account injuries or streaks.

For real gaming house odds: Things like recent games, head-to-head records, and such factor into it ... but goal differential, manipulated in different types of equations ... this is at the very heart of oddsmaking. And the oddsmakers are EXTREMELY GOOD at formulating hockey game odds. Spookily so. Better than any other sport by some distance, as far as I know.

I always find it interesting to see what happens to the game odds for a team after an injury or trade. I noticed that the Oiler game odds dropped quite a bit below expected when Reasoner went down with injury. Apparently they think he impacts the results of the team quite a bit.

BTW: I am NOT a gambler. Never have been or will be. I just think the bookmakers are the real experts, that's all.

igor* is offline  
Old
12-08-2003, 04:21 PM
  #10
PineJockey
Registered User
 
Join Date: Oct 2002
Location: E-Town
Posts: 616
vCash: 500
[QUOTE=igor]I assume the schedule correction thing on Sagarin's site is based on overall results, but I don't know ... it would just be way too complicated to do it any other way though.

Hmm, interesting. If this is true then I'm not sold on the approach. Every team's "ranking" must be different every game they play because of injuries, back to back, streaks, etc. A game against Detroit at the beginning of the season isn't the same as a game against Detroit at the end of the season. I bet that Sagarin actually does factor in the "daily" ranking of the teams when they actually play - complicated indeed! Otherwise it would seem to skew the effect of when you actually played a team. On the other hand if he doesn't then your 'luck' factor would include having the luck to hit teams when they are on skids, injuries, etc.

[QUOTE=igor]For real gaming house odds: Things like recent games, head-to-head records, and such factor into it ... but goal differential, manipulated in different types of equations ... this is at the very heart of oddsmaking. And the oddsmakers are EXTREMELY GOOD at formulating hockey game odds. Spookily so. Better than any other sport by some distance, as far as I know.

Again, interesting. I've actually heard/read the opposite. That is, that the bookies in Vegas, etc., are not nearly as good at hockey as they are at other sports. This seems to be reflected in the variety of odds that you can find for hockey between houses, as opposed to other sports which tend to be fairly conservative. I think the fact that goalies are very often not announced until right before game time hurts them since goalies (like pitchers in baseball) are so important in their odds-making. I don't gamble either, but do occasionally follow the numbers just for fun. It is especially a sucker's game here in Canada since the Proline makes you parlay everything.



[QUOTE=igor]I always find it interesting to see what happens to the game odds for a team after an injury or trade. I noticed that the Oiler game odds dropped quite a bit below expected when Reasoner went down with injury. Apparently they think he impacts the results of the team quite a bit.

Looks like they hit the nail on the head on that one. Sigh.

PineJockey is offline  
Old
12-08-2003, 06:58 PM
  #11
igor*
 
Join Date: Mar 2002
Posts: 3,276
vCash: 500
Quote:
Originally Posted by PineJockey
... I've actually heard/read the opposite. That is, that the bookies in Vegas, etc., are not nearly as good at hockey as they are at other sports. This seems to be reflected in the variety of odds that you can find for hockey between houses, as opposed to other sports which tend to be fairly conservative. I think the fact that goalies are very often not announced until right before game time hurts them since goalies (like pitchers in baseball) are so important in their odds-making. I don't gamble either, but do occasionally follow the numbers just for fun. It is especially a sucker's game here in Canada since the Proline makes you parlay everything.
Nope. If you check the online odds for individual hockey games. And do the math. You'll see that the NHL odds are run at the skinniest margins of all the major sports.

i.e. If you were unsensible , and bet on both teams in a hockey bet, in an effort to be guaranteed a return of $100 on a hockey bet ... you would have to bet a total of $ 103.50 at most betting houses. (3.5% margin)

That's really skinny. To do the same thing with and NFL game would typically cost you around $ 110. The oddsmakers are simply able to assess the probability of results in a hockey game much better than in other sports.

BTW: I agree on ProLine ... though I never gamble anyways ... they basically take the published odds and add 13% extra profit margin ... and then only allow parlay bets. I'm happy that other people play it though ... less taxes for me to pay

igor* is offline  
Old
12-08-2003, 07:05 PM
  #12
oilswell
Registered User
 
Join Date: Mar 2002
Location: Wandering the globe
Posts: 1,760
vCash: 500
Quote:
Originally Posted by igor
I assume the schedule correction thing on Sagarin's site is based on overall results, but I don't know ... it would just be way too complicated to do it any other way though.
Can you explain why tracking daily ranking would be way too complicated?

oilswell is offline  
Old
12-08-2003, 07:35 PM
  #13
Oiler_Fan
Registered User
 
Join Date: Oct 2002
Posts: 220
vCash: 500
Quote:
Originally Posted by igor
THE INTERESTING THING is that the record of your team, be it good or bad, is a pretty poor predictor of future results. And historically the goal differential of your team (i.e. GF - GA) is a very good predictor of future performance.
Being a bit of a math nut, this is something I have noticed over quite a few years. The Oiler's goal differential has been a bit depressing. That's why I hope Salo returns to form and that Lowe can pull a trade to improve the defense corps.

Right now, 10% better on the PK could easily mean 3 or 4 points better in the standings, which is the difference between being in the top 8 and being on the outside looking in. It would be nice if the PP improves, but it's absolutely critical that this team fixes the PK if they have any hope of making the playoffs this year.

Oiler_Fan is offline  
Old
12-09-2003, 06:34 AM
  #14
PineJockey
Registered User
 
Join Date: Oct 2002
Location: E-Town
Posts: 616
vCash: 500
No way. Check today's lines for example:
http://www.usatoday.com/sports/gaming/odds/index.htm

If you look at all of the NFL lines they are basically identical for each of the houses, and they are ALL using a 10 cent line.

Now look at the hockey lines (I'll use the TB, NYI for one example):

The first house (BetWWWTS) has TB favoured by half a goal and 5 cents positive money.

The second house (Bet365) has NYI favoured by half a goal and 10 cents positive money. Opposite the first house!

The third house (Intertops) has TB favoured by half a goal with 15 cents positive money.

The fourth house (Skybook) has both teams even at a 20 negative money.

And the fifth house has TB favoured by half a goal and 5 cents positive money.

They are all over the place. I don't really understand your 103.50 calculation. Typically, the hockey lines are set at 20 cents until the favourite becomes prohibitive and then they change it.


Quote:
Originally Posted by igor
Nope. If you check the online odds for individual hockey games. And do the math. You'll see that the NHL odds are run at the skinniest margins of all the major sports.

i.e. If you were unsensible , and bet on both teams in a hockey bet, in an effort to be guaranteed a return of $100 on a hockey bet ... you would have to bet a total of $ 103.50 at most betting houses. (3.5% margin)

That's really skinny. To do the same thing with and NFL game would typically cost you around $ 110. The oddsmakers are simply able to assess the probability of results in a hockey game much better than in other sports.

BTW: I agree on ProLine ... though I never gamble anyways ... they basically take the published odds and add 13% extra profit margin ... and then only allow parlay bets. I'm happy that other people play it though ... less taxes for me to pay

PineJockey is offline  
Old
12-09-2003, 06:38 AM
  #15
rabi_sultan
Registered User
 
Join Date: Sep 2002
Location: London, England
Country: United Kingdom
Posts: 3,782
vCash: 500
Send a message via ICQ to rabi_sultan Send a message via AIM to rabi_sultan Send a message via MSN to rabi_sultan Send a message via Yahoo to rabi_sultan
Quote:
Originally Posted by Oiler_Fan
Being a bit of a math nut, this is something I have noticed over quite a few years. The Oiler's goal differential has been a bit depressing. That's why I hope Salo returns to form and that Lowe can pull a trade to improve the defense corps.

Right now, 10% better on the PK could easily mean 3 or 4 points better in the standings, which is the difference between being in the top 8 and being on the outside looking in. It would be nice if the PP improves, but it's absolutely critical that this team fixes the PK if they have any hope of making the playoffs this year.
my only explanation for this would be the truth in the statments of Kevin Lowe and Craig MacTavish from two years ago when they said they are going to open the Oilers style of play up a bit more .

rabi_sultan is offline  
Old
12-09-2003, 07:17 AM
  #16
igor*
 
Join Date: Mar 2002
Posts: 3,276
vCash: 500
Quote:
Originally Posted by Oiler_Fan
...
Right now, 10% better on the PK could easily mean 3 or 4 points better in the standings, which is the difference between being in the top 8 and being on the outside looking in. It would be nice if the PP improves, but it's absolutely critical that this team fixes the PK if they have any hope of making the playoffs this year.
Ya, I'm at a complete loss to explain the woeful PK. Just terrible ... I really wouldn't have believed that it was even possible for it to be so bad. I think it's gotten between the ears of the PKers now too. And you can just feel the whole team deflate with every PP goal they surrender.

On the bright side ... the Oilers have been terrific at even strength this year.

igor* 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 10:06 PM.

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

vBulletin Copyright ©2000 - 2014, Jelsoft Enterprises Ltd.
HFBoards.com is a property of CraveOnline Media, LLC, an Evolve Media, LLC company. 2014 All Rights Reserved.