League Wide Scoring - Year to Year Comparison

[hatterson]

2012-2013 is now over and I've updated my yearly scoring spreadsheet.

This year I've decided to post it in here as opposed to on the main boards as it generally is completely ignored on the main boards.

For those who haven't seen it before here's an explanation of what the sheet is
[collapse="Full Standings"]
So a couple years ago I started looking at the question of how much, and why, is scoring changing since the lockout.

In my opinion it's not enough to simply look at raw goals per game. Is it really fair to say that the NHL is in a lower scoring era if teams have 20 fewer PP minutes/game, but are only scoring .02 goals per game less? Or is it more fair to say that we're in a lower penalty minute era?

To help look at that I extracted some numbers (from NHL.com and ESPN.com, simple little rips) and found the values for each season of goals per minute of ES time, PP time and SH time. I then can multiple that by the various amounts of time spent at each 'phase' for different seasons and compare them.

This is a simplistic analysis and does not attempt to adjust for anything aside from ice time. I understand that changes in how rules are called (as seen after the lockout) can have a dramatic affect on what happens at all phases of the play, however I do not attempt to adjust for that. I'm looking purely at ice time.
[/collapse]
Full spreadsheet can be found:
Here

If you wish to modify the compare year you can either make a local copy of the spreadsheet or you can send me a note and I'll bounce it over. At some point in the future I may toss this through a simple web form and allow you to pick the year yourself, but I'll look at that over the summer.


Random notes about 2012-2013:

- By a statistical fluke, teams spent a virtually identical percentage of the game on the PP and at ES as they did in 2011-2012

- Scoring decreased by ~.0128 goals per game from last year. Both ES and SH 'efficiency' decreased, but PP 'efficiency' increases partially off-set this.

- Overall goals per game were almost identical to 2002-2003, however in 2002-2003 teams had an average of roughly 2 minutes extra powerplay time each per game. If teams played identical to this year but had the PP time of 2002-2003 scoring would have been nearly .2 goals per game higher.

Click to read more...

 

[Jyrki]

I'm almost doing complying a similar set of stats, though I also aggregated ES TOI stats from Behind The Net. Average ES play per game is closer to 50 minutes rather than the 60.5 estimate you have up there on the spreadsheet.

 

[hatterson]

I'm almost doing complying a similar set of stats, though I also aggregated ES TOI stats from Behind The Net. Average ES play per game is closer to 50 minutes rather than the 60.5 estimate you have up there on the spreadsheet.

Sorry if I wasn't clear. I estimated 60.5 per game overall, not just ES. Which means a total of 121 team minutes.

This past year there were 10.87 team minutes spent both at PP and PK leaving ~99.26 team minutes at even strength (49.63 per team per game)

 

[Jyrki]

Sorry if I wasn't clear. I estimated 60.5 per game overall, not just ES. Which means a total of 121 team minutes.

This past year there were 10.87 team minutes spent both at PP and PK leaving ~99.26 team minutes at even strength (49.63 per team per game)

Ah, my apologies.

 

[Nalens Oga]

I recommend including the playoffs and also doing it to take out the 2-3 lowest and highest scoring teams to see how much of a difference it makes.

The 12/13 playoffs had 4.97 goals per game and when you remove the first two Pittsburgh series then it was only 4.69 goals per game which is pretty bad.

 

[Tough Guy]

So, the long and the short of it is, we are back to dead puck.

 

[hatterson]

So, the long and the short of it is, we are back to dead puck.

Not exactly.

From 03-04 to 05-06 scoring went up nearly .9 goals per game, since then it has dropped back down by roughly .75 goals per game but roughly half of that is purely due to changes in ice time.

If teams had the same powerplay opportunities they had in 2005-2006 we would expect scoring to be *at least* .36 goals/game higher than it is now.

Referee's are simply not calling near as many penalties as they used to. The 4 years before the lockout each team had an average of 7.32 minutes of power play ice time per game. The last 2 years, each team had an average of 5.43 minutes.

However, it is worth looking at even strength performance as well. The scoring rate of .03997 goals per even strength minute (basically 1 goal every 25 even strength minutes per team) is lower than any pre-lockout year except 2003-2004 (.03831 g/min ~= 26 minutes per goal)

To me, the data suggests one fundamental thing. If you want to increase scoring, the referees need to start calling penalties more. Adjusting goalie pads, making the nets bigger are fine, but you can easily address at least half of the post lockout scoring drop without adjusting rules and simply deciding to enforce them.

I recommend including the playoffs and also doing it to take out the 2-3 lowest and highest scoring teams to see how much of a difference it makes.

The 12/13 playoffs had 4.97 goals per game and when you remove the first two Pittsburgh series then it was only 4.69 goals per game which is pretty bad.

The reason I didn't include the playoffs is because of the small sample size. Even in this shortened season there were 720 games played (1230 in a normal length season) each team plays in only 1/15th of those games, whereas the playoffs has a maximum of 105 games and that only occurs if every series went to 7 games. This year only had 86 games with Boston and Chicago both being involved in over 1/4 of them.

The variability due to which teams move on and which don't is gigantic in the playoffs so it doesn't seem fair to make judgement on league-wide scoring based on it.

 

[decma]

However, it is worth looking at even strength performance as well. The scoring rate of .03997 goals per even strength minute (basically 1 goal every 25 even strength minutes per team) is lower than any pre-lockout year except 2003-2004 (.03831 g/min ~= 26 minutes per goal)

To me, the data suggests one fundamental thing. If you want to increase scoring, the referees need to start calling penalties more. Adjusting goalie pads, making the nets bigger are fine, but you can easily address at least half of the post lockout scoring drop without adjusting rules and simply deciding to enforce them.

I am having trouble reconciling the two paragraphs I quoted. If ES scoring per minute was lower than any pre-lockout year except 03/04, doesn't that suggest that the problem is not just fewer PPs, but also low ES scoring?

How do ES goals/60 minutes compare with the average during the 1980s, for example?

If ES goals/60 minutes were at the 1980s average, what would total goals per game be, even given the drop in PP time compared with 03/04 to 05/06?

And re the PP time, is it that the recent numbers have been down, or is it that the 03/04 to 05/06 data were anomalous (temporary spike in PP opps during those seasons)?

Lastly, has there been more variation in PP scoring efficiency or ES scoring efficiency in the post-merger era?

Thanks.

 

[hatterson]

I am having trouble reconciling the two paragraphs I quoted. If ES scoring per minute was lower than any pre-lockout year except 03/04, doesn't that suggest that the problem is not just fewer PPs, but also low ES scoring?

Sorry, the second paragraph was largely my opinion.

Yes, lower ES scoring is a problem, however my opinion is that a large chunk of that difference is also due to how penalties are called, or at the very least, it's easier to fix. I may be right, I may not, but I believe it's the simplest approach to take. It's entirely within the league's control and they don't need to bother with the competition committee. Just call penalties when they occur and keep it consistent. If scoring still trends down, then address it with other rule changes.

How do ES goals/60 minutes compare with the average during the 1980s, for example?

If ES goals/60 minutes were at the 1980s average, what would total goals per game be, even given the drop in PP time compared with 03/04 to 05/06?

Goals data (ES, PP and SH) is fairly trivial to dig up, but ice time is less so. If you have a source for the data points (League wide PPTOI is basically the only one I'd need as I can get close enough for ESTOI) prior to 00-01 I'd gladly add it to the sheet.

I have PP opportunity back to 90-91 and I can throw in guestimates for 90-91 -> 99-00. I'll adjust the sheet now and see what happens.

And re the PP time, is it that the recent numbers have been down, or is it that the 03/04 to 05/06 data were anomalous (temporary spike in PP opps during those seasons)?

In terms of recent history, both. In terms of longer history (prior to the 90s, I'm not sure).

Number of PP opportunities was slightly lower in the late 90s/early 00s than it was in the early 90s, but it wasn't that significant a change. 91/92 through 93/94 were the highest three seasons I have records for and each team had an average of 413 Power Plays. That decreased to 356 per team from 00/01 through 03/04.

After the lockout we saw an immediate jump to 480 PPs/team which far eclipsed the previous record of 432 in 92/93. In 06/07 that dropped back down to 397 which was slightly higher than the decade average before the lockout, but not significantly so.

In 07/08 it dropped to 351, comfortably within the 'normal' zone. Then in 08/09 it went to 304 which was the lowest level since 90/91. 09/10 set another low record at 291, 11/12 lowered again to 271 and 12/13 had 272.

The last two years teams have received roughly only 3/4 of the PPs they did during the height of the dead puck era.

Lastly, has there been more variation in PP scoring efficiency or ES scoring efficiency in the post-merger era?

I'm adding estimated (albeit I believe good estimates) for data from 90/91 until 99/00 so I can give a more complete answer then.

 

[hatterson]

Ok, I have the updated data.

Caution: All my data prior to 2005-2006 has estimated power play time. This is done be computing an estimated average time per power play and multiplying by the number of powerplays. The formula was simply to assume all powerplays are 120 seconds long and to not count successful powerplays in the time. It sounds absurdly simple, but comparing it to the actual data (from 05/06 until 12/13) it yields surprisingly accurate results. The largest errors were roughly 1.5% away from expected. It's certainly not ideal, but it's the best I have for now.

How do ES goals/60 minutes compare with the average during the 1980s, for example?

As before, I don't have the 1980s, but I did add the early 90s.

ES scoring in 05/06 was markedly lower than ES scoring in the early 90s, and very slightly lower than in the mid 90s. .045 goals/min in 05/06 compared to .0462/min average from 94/95 through 96/97 and .054 in 90/91 through 92/93.

Currently we're at .041 goals/min in the last 6 years combined and it's stayed mostly stable, varying only .001 goals/min either way

If ES goals/60 minutes were at the 1980s average, what would total goals per game be, even given the drop in PP time compared with 03/04 to 05/06?

Interestingly enough, the rates of scoring in the early 90s were slightly less impacted by trading PP time for ES time.

The last 2 years average a minute of PP time has been worth ~.121 goals (.108 PP and .013 SH) and a minute of ES time has been worth .080 goals (.040 per teamminute). This means that trading 1 minute of PP time (and 1 min of SH time) for 2 minutes of ES time would result in goalscoring going down by roughly .041 goals per game.

However from 90/91 through 92/93 a minute of PP time was worth roughly .139 goals (.121+.018) and a minute of ES time has been worth .108 goals (.054 per teamminute). This means that trading 1 minute of PP time (and 1 min of SH time) for 2 minutes of ES time would result in only a loss of .031 goals.

That being said, if penalties were called at an 05-06 level and play was at a 90/91 - 92/93 level there would be roughly 7.1 goals per game. If penalties were called at a 2012/2013 level and scoring was at a 90/91-92/93 level there would be roughly 6.85 goals per game.

So clearly since the early 90s the largest impact on goal scoring has been play style. Although that should come as no surprise to anyone who has watched the game and seen the rise of the butterfly goaltender and the development of defensive systems.

The reason why I believe we should, for right now, look at calling penalties is because the record low level of power plays isn't helping the situation at all. Shooting for the 80s/early 90s doesn't seem reasonable, but shooting for 2006/2007 through 2008/2009 shouldn't be and based on the numbers, simply calling a historically normal level of penalties (league wide average from 90/91 until 03/04 was 11,175 The last two years have averaged 8,153) would get us at least half way there.

Lastly, has there been more variation in PP scoring efficiency or ES scoring efficiency in the post-merger era?

The largest variation (in terms of percentage change) is actually in SH time, although it makes up such a small percentage of scoring it hardly affects total numbers.

If we include PP and SH time in the same bin, it's clear that ES scoring has shrunk much faster than PP/SH scoring. Teams are scoring at roughly 75% the rate at even strength that they did in the early 90s, whereas they're scoring at 85-90% the rate on PP/SH

 

[Catamarca Livin]

Do not know the numbers but the West is a much lower scoring Conference overall. Especially last year with no interconference games. Is that just a style issue? As a Canuck fan seems everything changed in the 2011 finals. Last year it reached a new level as hooks that were automatic penalties since the lockout were ignored. The standard has changed. Another infraction is crosschecks along the boards. Seems to be considered good defense now except on the odd occasion where the players goes head first into the boards then that is considered a suspension. In the playoffs every series is reffed in a different manner. The Sharks/Canuck series had a strict interpretation (atleast one way) while the Kings/Blues series was a death match. Personally i like the let them play style with bodychecks. As the Kings/Blues series was a great series to watch. However if the league wants more scoring and skill the stick rules will have to be enforced. The Pittsburgh/Boston series was an example of hooking and slashing not being called early in the series, therefore giving Boston a great advantage. Something is wrong when the greatest players in the world can be so easily contained.

 

[BM67]

Goalie stats for 2012-13, so not including empty net time or goals.

Conf|GAA|SV%|SOG/60|H GAA|H SV%|H SOG/60|R GAA|R SV%|R SOG/60
East|2.60|.911|29.39|2.49|.913|28.75|2.72|.909|30.04
West|2.47|.912|28.18|2.33|.915|27.48|2.61|.910|28.88
NHL|2.54|.912|28.79|2.41|.914|28.11|2.66|.910|29.46

East teams were 198-129-33 at home, and West teams were 211-107-42.