HFBoards

Go Back   HFBoards > General Hockey Discussion > The History of Hockey
Mobile Hockey's Future Become a Sponsor Site Rules Support Forum vBookie Page 2
The History of Hockey Relive great moments in hockey history and discuss how the game has changed over time.

Hard and easy icetimes during even strength?

Reply
 
Thread Tools
Old
12-23-2011, 07:40 AM
  #1
plusandminus
Registered User
 
Join Date: Mar 2011
Posts: 980
vCash: 500
Most people here may know that +/- numbers should be looked at in context. One may for example try to isolate ES (even strength) play, and then compare the player's +/- during ES play with how the team did when the player wasn't on the ice. One can also look into which players the player seem to have played most with, and try to make an adjustment - mostly in ones mind - based on that.

To complicate things further, there is also often a talk about "easy" and "hard" minutes. It seems as if hard minutes is basically related to a "shutdown role" and "facing the opponent's best player(s)". Easy minutes, thus, would indicate one was allowed to play against the opponent's "lesser good" players, and/or that one was allowed to focus a lot on offense rather than defence. (Or?)

I have a bit of problems understanding this "hard/easy minutes" thing fully, because when I try to I end up thinking it should pretty much eaven itself out. ?

First, all teams play (to use recent seasons) 41 games at home, and 41 games on the road. When playing at home, one has the advantage of the opponent having to change (or remain) their players before oneself does it. One thus can very much choose which of ones players that will face which opponent(s). On the road, it's the opposite, and the opponent can choose.

I'm not very sophisticated in my understanding of how teams are constructed, and how they match up against each other. My limited knowledge is that it's usually said to be like below:
ForwardLine1st is Off1
ForwardLine2nd is Off2
ForwardLine3rd is Def1
ForwardLine4th is Def2

Is that a fairly accurate desciption of the most common useage of the forward lines?
Is there some consistency how these forward lines match up against the opponents? Is it often 1st vs opponent's 3rd, and vice versa? Or is to common to have 1st play against opponent's 1st? Maybe 1st aganist 2nd? If so, is it so sure that ones best line have to face the best opponents, or at least the most difficult ones to outscore during ES? Who are the most difficult opponents?

Playing on the road, wouldn't most players have to face about equally hard competition? (Some opponents use their checking line vs ones 1st line, others use their scoring line, etc.?)

Playing at home might be a different thing, though, as one can choose which opponents to play against. But is it so dead sure which opponent's player(s) one would want to play against? Making it easier for one of ones forward lines, might make it harder for ones other line(s). Or do some teams constantly let their players play the opponent's "lesser good" players? If so, let's suppose those "lesser good players" have an icetime of 6 minutes per game. Wouldn't one want ones best players to play more than that, and if so they thus would face other players than the opponent's "lesser good" ones.


Turning to the defencemen... Doesn't playing a certain defenceman say 18-20 minutes per game during ES mean he ends up facing all kinds of players, especially during the 41 games one plays on the road?


Would +/- on the road (where things may even themselves out more) be more telling than +/- at home (where one can "choose which opponents to face"?


Since this is mainly about +/-... Wouldn't focusing a lot to "shutdown" a certain player, mean there might be more opportunities for his unit mates? So while the player playing "hard minutes" may (if he is a offensive star) see his points production get affected negatively, his +/- stats might not get as affected as his unit mates makes up for it?


Has there been any reseach/study of some kind regarding "hard" and "easy" minutes? (How much does it affect things?)
My main thoughts are about how "easy" and "hard" minutes affects +/- over a whole season. Estimated numbers are welcome.
Is "easy minutes" on ES related to playing against "easy opponent player"? If so, what is an easy opponent player?

I know coaching would depend on many things, like if leading or trailing, etc.

(There will be another, more case specific, thread too. I hope moderators don't mind. It's not important to me that replies are totally on-topic in regard to which of the two threads to choose to post in.)


Last edited by plusandminus: 12-23-2011 at 07:56 AM.
plusandminus is offline   Reply With Quote
Old
12-23-2011, 08:50 AM
  #2
plusandminus
Registered User
 
Join Date: Mar 2011
Posts: 980
vCash: 500
Some more thoughts...

Ideally, a coach is able to give all his players "easy minutes". Some players may do best in a checking role, others in a more offensive role. It would be logical to coach according to that, and I suppose it is being done. In addition, everything is being decided based on "What's best for the team?". (That includes PP and PK too. One can not necessarily look at PK time and determine which are the best PK players on a team. A player might be his team's best PK player, but be used much more on PP, because the team have otherwise weak PP players while they have some average PK players.)

Watching games, I often think it's fairly even which are the best lines on a team. Sometimes a "third line" may dominate the play most times they're on the ice, sometimes even if facing the opponent's "first line". When Sweden won in the Olympics in 2006 (thinking of it since it's currently being discussed), I remember Alfredsson/Axelsson at times being the best Swedes during games. From Detroit's Stanley Cup teams, I remember guys like Draper and Maltby doing great (but perhaps they played against "lesser good players"..?). Best players is not necessarily equal to "those scoring most points". ...On the other hand, I just mentioned top teams. Average or below average teams might be different.

What about the original 6 era? Those here who were around often say that the original six era was very strong, with each team consisting of good to great players (in comparison to today, when there is 30 teams and the bottom half players in the league may - or may not - be of lesser skill than the "worst" regulars during O6).

If coaches would pay no attention at all regarding matchups during ES, and just put players on the ice totally independant of which players the opponent put on the ice, how do you think +/- during ES play would be affected? (Note: I mean that each player still would still have the same icetime as now, and play in the same units as during games.)

plusandminus is offline   Reply With Quote
Old
12-23-2011, 09:15 AM
  #3
overpass
Registered User
 
Join Date: Jun 2007
Posts: 3,639
vCash: 500
Quote:
Originally Posted by plusandminus View Post
Most people here may know that +/- numbers should be looked at in context. One may for example try to isolate ES (even strength) play, and then compare the player's +/- during ES play with how the team did when the player wasn't on the ice. One can also look into which players the player seem to have played most with, and try to make an adjustment - mostly in ones mind - based on that.

To complicate things further, there is also often a talk about "easy" and "hard" minutes. It seems as if hard minutes is basically related to a "shutdown role" and "facing the opponent's best player(s)". Easy minutes, thus, would indicate one was allowed to play against the opponent's "lesser good" players, and/or that one was allowed to focus a lot on offense rather than defence. (Or?)

I have a bit of problems understanding this "hard/easy minutes" thing fully, because when I try to I end up thinking it should pretty much eaven itself out. ?

First, all teams play (to use recent seasons) 41 games at home, and 41 games on the road. When playing at home, one has the advantage of the opponent having to change (or remain) their players before oneself does it. One thus can very much choose which of ones players that will face which opponent(s). On the road, it's the opposite, and the opponent can choose.

I'm not very sophisticated in my understanding of how teams are constructed, and how they match up against each other. My limited knowledge is that it's usually said to be like below:
ForwardLine1st is Off1
ForwardLine2nd is Off2
ForwardLine3rd is Def1
ForwardLine4th is Def2

Is that a fairly accurate desciption of the most common useage of the forward lines?
Is there some consistency how these forward lines match up against the opponents? Is it often 1st vs opponent's 3rd, and vice versa? Or is to common to have 1st play against opponent's 1st? Maybe 1st aganist 2nd? If so, is it so sure that ones best line have to face the best opponents, or at least the most difficult ones to outscore during ES? Who are the most difficult opponents?

Playing on the road, wouldn't most players have to face about equally hard competition? (Some opponents use their checking line vs ones 1st line, others use their scoring line, etc.?)

Playing at home might be a different thing, though, as one can choose which opponents to play against. But is it so dead sure which opponent's player(s) one would want to play against? Making it easier for one of ones forward lines, might make it harder for ones other line(s). Or do some teams constantly let their players play the opponent's "lesser good" players? If so, let's suppose those "lesser good players" have an icetime of 6 minutes per game. Wouldn't one want ones best players to play more than that, and if so they thus would face other players than the opponent's "lesser good" ones.


Turning to the defencemen... Doesn't playing a certain defenceman say 18-20 minutes per game during ES mean he ends up facing all kinds of players, especially during the 41 games one plays on the road?


Would +/- on the road (where things may even themselves out more) be more telling than +/- at home (where one can "choose which opponents to face"?


Since this is mainly about +/-... Wouldn't focusing a lot to "shutdown" a certain player, mean there might be more opportunities for his unit mates? So while the player playing "hard minutes" may (if he is a offensive star) see his points production get affected negatively, his +/- stats might not get as affected as his unit mates makes up for it?


Has there been any reseach/study of some kind regarding "hard" and "easy" minutes? (How much does it affect things?)
My main thoughts are about how "easy" and "hard" minutes affects +/- over a whole season. Estimated numbers are welcome.
Is "easy minutes" on ES related to playing against "easy opponent player"? If so, what is an easy opponent player?

I know coaching would depend on many things, like if leading or trailing, etc.

(There will be another, more case specific, thread too. I hope moderators don't mind. It's not important to me that replies are totally on-topic in regard to which of the two threads to choose to post in.)
The quality of opponent that a player faces is important. Yes, this is more difficult for coaches to control on the road, so road results should be a more neutral metric. See, for example, Ryan Nugent-Hopkins and line mates, who have performed much better at home this season because their coach gets them favourable matchups.

Going beyond player matchups, the part of the ice in which a player starts his shift is important. Some players go out for many more offensive zone faceoffs, and others are used for more defensive zone faceoffs. This also can have a large effect on plus-minus.

There is a variety of ways that coaches choose to match up their lines. I believe most coaches use their first line against the toughest competition, going first line vs first line Some use their "second line" like Boston and Patrice Bergeron's line, or Vancouver and Ryan Kesler's line.

Historically, some teams have used the third line to check their opponent's first line. Players like Bob Gainey, Joel Otto, Steve Kaspar, etc were famous for it. But that is less common than it used to be and I'm not sure if any team still does it. The Anaheim Ducks did back in 2007 when they won the Stanley Cup. I think fourth lines almost always face easier minutes, and coaches often play 4th line vs 4th line in a sort of "gentleman's agreement." Third lines often take more than their share of defensive zone faceoffs, even if they don't play the toughest competition on average. Some coaches who are fortunate enough to have defensively strong fourth liners use them in the same way. For example, John Madden on Chicago in recent years.

I know the Ottawa Senators best and can tell you which matchups they currently tend to use.

First line (Jason Spezza) vs opponent's first line.
Third line (Zack Smith) vs opponent's second line.
Second line (Nick Foligno and now Kyle Turris) vs opponent's third line.
Fourth line vs opponent's fourth line...but Jesse Winchester gets some extra shifts in key defensive situations, like defensive zone faceoffs when protecting a lead.

Earlier this season the Senators were playing rookies (Mika Zibanejad and Stephane da Costa) in the 2nd and 3rd line centre positions and were trying to give both of them easier minutes, so the first and fourth lines had to play tougher minutes. In fact, the fourth line ended up becoming the third line some nights. (This has happened before...they would try giving some younger players a chance at the third line at the beginning of the season, put Chris Kelly on the fourth line, and then the fourth line would turn into the third line, especially on the road, because they could trust them more. Kelly and co. would always end the season on the third line.) Since they shipped first Zibanejad and then da Costa out and moved more real NHL players into their top three lines, they have been able to use more traditional roles for their lines, and have had much more success.

Some links:

55% of variation in shot plus-minus is driven by quality of competition. This is a fully iterated look at qualcomp, which really helps. First-order qualcomp tends to underestimate the magnitude of the effect, because in general good players play against good players and bad players play against bad players.

The effects of (first-order) quality of competition, as well as the zones in which players tend to take faceoffs.

Ryan Nugent-Hopkins and linemates have been put in very favourable positions by their coaches, and as a result have been much more successful at home than on the road.


Last edited by overpass: 12-23-2011 at 09:32 AM.
overpass is offline   Reply With Quote
Old
12-23-2011, 09:25 AM
  #4
TheDevilMadeMe
Global Moderator
 
TheDevilMadeMe's Avatar
 
Join Date: Aug 2006
Location: Brooklyn
Country: United States
Posts: 42,697
vCash: 500
Pre-lockout, it was easier to get the matchups you wanted.

In the playoffs, Scotr Stevens regularly saw 90%+ of his shifts against the top line of opponents. Which of course meant the 2nd and 3rd pairs weren't seeing those opponents much at all.

This was easier before the lockout. Pat Burns would have the NJ Devils ice the puck on purpose if the other top line came out there and Stevens wasn't out there. This is obviously no longer a viable strategy as teams can't change after icings.

There are more creative ways to do this. In the 1995 playoffs, if Lindros stepped on the ice, Jacques Lemaire would have the nearest Devil to the bench change out for Scott Stevens, even if that player was a forward. The rest of the lines would just get changed when they could. Again, being able to ice the puck of it looked like the line change would be poor made this easier.

TheDevilMadeMe is online now   Reply With Quote
Old
12-23-2011, 11:31 AM
  #5
overpass
Registered User
 
Join Date: Jun 2007
Posts: 3,639
vCash: 500
Quote:
Originally Posted by plusandminus View Post
If coaches would pay no attention at all regarding matchups during ES, and just put players on the ice totally independant of which players the opponent put on the ice, how do you think +/- during ES play would be affected? (Note: I mean that each player still would still have the same icetime as now, and play in the same units as during games.)
In the modern NHL I think the spread of plus-minus would widen if lines were randomly used with no regard to matchups. Hockey coaches tend to play good players against good players, which compresses the spread of plus-minus.

I think this is also true for the majority of NHL history, going back to the origin of multiple lines in the 1920s. The best players tend to play against the other team's best players. Checking lines, with lesser players going against the stars, have often been used as well. More at some times than others. But overall, I think their use has been a minority.

(Good players also tend to play with good teammates, which increases the spread of plus-minus. But that's another topic.)

overpass is offline   Reply With Quote
Old
12-23-2011, 03:37 PM
  #6
steve141
Registered User
 
Join Date: Aug 2009
Posts: 736
vCash: 500
Several interesting questions in this thread.

Quote:
Originally Posted by plusandminus View Post
Turning to the defencemen... Doesn't playing a certain defenceman say 18-20 minutes per game during ES mean he ends up facing all kinds of players, especially during the 41 games one plays on the road?
Yes, but there is often a big difference in who gets the offensive/defensive zone faceoffs.

Mike Babcock did an interesting experiment last season. Instead of having clearly defined 1st, 2nd and 3rd defensive pairings he put his three best defencemen (Rafalski, Kronwall, Lidstrom) on three different lines. The idea was to just roll the three lines evenly, and not worry about matchups. Each line was supposed to play twenty minutes a night, so that the best players were rested come playoff time.

I'm not sure the experiment was a great success, but it's a fun case to analyze. One interesting aspect of Detroit's games last year was that they actually had a better record away than at home. It's hard to say whether this had anything to do with the unusual defensive matchups, but it's possible.

As the year went on Babcock gradually abandoned the idea of rolling them evenly, shifting the Lidstrom-Stuart pairing to a more defensive role and the Rafalski-Kindl pairing to a more offensive role.

It's also interesting to analyze this experiment with the fact that Lidstrom won the Norris in mind.

One idea of Babcock's was that each line should have one defensive and one offensive defenceman. The offensive defenceman in Babcocks system almost plays like a rover, jumping up into the play constantly, while the defensive defenceman hangs back to cover for him. This meant that Lidstrom, who usually plays a defensive role, had to play the offensive role in his pairing with defensive specialist Stuart. This might be reflected in the fact that Lidstrom finished second in the league in defenceman scoring while finishing a minus for the year for the first time in his career.

Also, after the season Lidstrom apparently had a talk with Babcock where he explained that he needed to play more minutes to stay warm through-out the game. Babcock has not tried to repeat the even-minutes experiment.

steve141 is offline   Reply With Quote
Old
12-23-2011, 03:46 PM
  #7
plusandminus
Registered User
 
Join Date: Mar 2011
Posts: 980
vCash: 500
Quote:
Originally Posted by overpass View Post
(Interesting replies of good quality.)
Thanks for great replies. I need to spend more time on them (including doing it when I'm less tired).

I spent some days a couple of months ago studying faceoffs, downloading the play_by_play pages on nhl.com and writing a program to convert the html code to something I could store in a database. I noticed that taking an offensive faceoff resulted in an increased chance to score within 10(??) or so (I don't remember) seconds, followed by an increased chance to surrender a goal if one didn't score, until it evened itself out. Overall, I think (don't remember exactly) that it was a significant advantage to have offensive faceoffs compared to defensive.
It may even seem more important to create an offensive faceoff, than to actually win it (!), but I don't remember exactly. It's one of many things I hope to look more into eventually.

Until I have spent more time digesting your post...
How telling do you think +/- on the road is? (I mean, to what extent does it take away matchup biases?)
Would it be more "fair" to just look at away +/- (or even away statistics as a whole), than to look at all games?

plusandminus is offline   Reply With Quote
Old
12-23-2011, 05:42 PM
  #8
plusandminus
Registered User
 
Join Date: Mar 2011
Posts: 980
vCash: 500
Quote:
Originally Posted by steve141 View Post
... there is often a big difference in who gets the offensive/defensive zone faceoffs.


Mike Babcock did an interesting experiment last season. Instead of having clearly defined 1st, 2nd and 3rd defensive pairings he put his three best defencemen (Rafalski, Kronwall, Lidstrom) on three different lines. The idea was to just roll the three lines evenly, and not worry about matchups. Each line was supposed to play twenty minutes a night, so that the best players were rested come playoff time.

I'm not sure the experiment was a great success, but it's a fun case to analyze. One interesting aspect of Detroit's games last year was that they actually had a better record away than at home. It's hard to say whether this had anything to do with the unusual defensive matchups, but it's possible.

As the year went on Babcock gradually abandoned the idea of rolling them evenly, shifting the Lidstrom-Stuart pairing to a more defensive role and the Rafalski-Kindl pairing to a more offensive role.

It's also interesting to analyze this experiment with the fact that Lidstrom won the Norris in mind.

One idea of Babcock's was that each line should have one defensive and one offensive defenceman. The offensive defenceman in Babcocks system almost plays like a rover, jumping up into the play constantly, while the defensive defenceman hangs back to cover for him. This meant that Lidstrom, who usually plays a defensive role, had to play the offensive role in his pairing with defensive specialist Stuart. This might be reflected in the fact that Lidstrom finished second in the league in defenceman scoring while finishing a minus for the year for the first time in his career.

Also, after the season Lidstrom apparently had a talk with Babcock where he explained that he needed to play more minutes to stay warm through-out the game. Babcock has not tried to repeat the even-minutes experiment.
Thanks for your interesting reply, with its great timing. Some days ago, I was actually looking at the Detroit defencemen and their +/- during even strength. In 2009-10, Lidstrom seem to have played mostly with Rafalski, and they had great +/- (58-38, +20). Last season, he however - just like you confirm - seem to have played mostly with Brad Stuart, and they did pretty much worse (38-37, +1).
I started thinking about it, and that maybe Stuart is a below average player, while Rafalski is an above average player (of course I know about him, but he is also getting older). I was actually thinking about some things you mentioned, and that it might be more favourable for Detroit to use their defencemen differently. And now you suddenly writes about the background.

Sorry if someone don't like the "duplicates" below, but I intend to make the table sortable and if so one would want to sort on e.g. Lidstrom to see his partners.
Minimum of 11 goals (ESGF+ESGA) together.

SeasTeamPosPos2NameName2ESGSESGDESGFESGA
2009DETDDBrian RafalskiNicklas Lidstrom96205838
2009DETDDNicklas LidstromBrian Rafalski96205838
2009DETDDNiklas KronwallBrad Stuart4352419
2009DETDDBrad StuartNiklas Kronwall4352419
2009DETDDAndreas LiljaJonathan Ericsson11165
2009DETDDJonathan EricssonAndreas Lilja11165
2009DETDDBrad StuartNicklas Lidstrom11165
2009DETDDNicklas LidstromBrad Stuart11165
2009DETDDJonathan EricssonBrett Lebda23-11112
2009DETDDBrett LebdaJonathan Ericsson23-11112
2009DETDDBrian RafalskiBrad Stuart11-347
2009DETDDBrad StuartBrian Rafalski11-347
2009DETDDBrad StuartDerek Meech18-4711
2009DETDDDerek MeechBrad Stuart18-4711
2009DETDDBrad StuartBrett Lebda13-549
2009DETDDBrett LebdaBrad Stuart13-549
2009DETDDBrad StuartJonathan Ericsson20-12416
2009DETDDJonathan EricssonBrad Stuart20-12416
SeasTeamPosPos2NameName2ESGSESGDESGFESGA
2010DETDDBrian RafalskiJonathan Ericsson68103929
2010DETDDJonathan EricssonBrian Rafalski68103929
2010DETDDNiklas KronwallRuslan Salei3241814
2010DETDDRuslan SaleiNiklas Kronwall3241814
2010DETDDNicklas LidstromBrian Rafalski2531411
2010DETDDBrian RafalskiNicklas Lidstrom2531411
2010DETDDNicklas LidstromBrad Stuart7513837
2010DETDDBrad StuartNicklas Lidstrom7513837
2010DETDDBrad StuartNiklas Kronwall15187
2010DETDDNiklas KronwallBrad Stuart15187
2010DETDDNiklas KronwallJonathan Ericsson18099
2010DETDDJonathan EricssonNiklas Kronwall18099
2010DETDDJakub KindlRuslan Salei34-21618
2010DETDDRuslan SaleiJakub Kindl34-21618
2010DETDDNicklas LidstromJonathan Ericsson12-257
2010DETDDJonathan EricssonNicklas Lidstrom12-257
2010DETDDNicklas LidstromNiklas Kronwall25-31114
2010DETDDNiklas KronwallNicklas Lidstrom25-31114
(Reload the page if the tables are not sortable.)

Above we can see how different Detroit defencemen did depending on which partner they were paired with.

It was basically when I thought further about it, that I came up with this thread, thinking that Lidstrom+Stuart might for example have had "harder ES minutes" than Rafalski+Ericsson.


Last edited by plusandminus: 12-23-2011 at 06:01 PM.
plusandminus is offline   Reply With Quote
Old
01-18-2012, 03:50 PM
  #9
plusandminus
Registered User
 
Join Date: Mar 2011
Posts: 980
vCash: 500
I have thought more about the graph. Does anyone know if it shows all situations, as opposed to only showing ES?

It shows (I think) that Sedins get 70 % offensive zone faceoffs, while Samuel Pahlsson get 30 % offensive zone faceoffs.
If that includes all situations, I think the results are very natural, and really may not tell much. I'll try to explain why...

There are about 60 faceoffs per game. I think there on average may be about 7 faceoffs when playing PP, 7 when playing SH, and 46 while playing SH. Those numbers may be slightly wrong.
Sedins play lots of PP, but little SH. Let's say the are on ice for 16 ES faceoffs, of which 8/16 (half of them) are offensive zone. They may be on ice for 4/4 PP faceoffs. If so, they'll end up with 12/20 = 60 % offensive zone faceoffs.
Pahlsson play lots of SH, but little PP. Maybe 8/16 faceoffs during ES play are in offensive zone, while 0/4 during SH are offensive zone. That's 8/20 = 40 % offensive zone.
It's still not 70 % and 30 % respectively, but still taking special team faceoffs affects the numbers.

So if the data shows all situations combined, one mat not necessarily be able to apply the numbers on ES play.
(Should it be ES only, and still show the percentages in the graph, it's however of course indicative.)

plusandminus is offline   Reply With Quote
Old
01-18-2012, 09:54 PM
  #10
Sens Rule
Registered User
 
Sens Rule's Avatar
 
Join Date: Sep 2005
Country: Canada
Posts: 15,629
vCash: 500
Now it is 45-55 second shifts for everyone. What about the more recent past... Early 90's when Gretzky or Mario or Yzerman or other too players play 1:30 shifts... Or the 70's when say Esposito plays over 2 minute shifts?

Do teams pla two lines against the top
Lines? Did any line spend 2+ minutes. He king Esposito... Did Otto or Kasper spend 2 minutes on ice vs Gretzky? Just wondering.

Sens Rule 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 05:41 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.