Special Teams Efficiency

[pokerNhockey]

prior to games on January 21st, Calgary's PK was sitting at 22nd -> not great. However, I think that is a misleading stat since I think the goal of the PK should be to end even. Therefore, every shorty should count as a "credit" when judging the PK's efficiency (similar to how +/- works) so I decided to run the numbers to determine the league rank for what I would call "PK Efficiency" -> (GA - GF) / Times Shorthanded.

When using this metric the Flames jump to second in the league @ 88%, behind only Arizona at 96%. Both teams score shorties on 9% of their PK's.

A similar adjustment to powerplay efficiency move the Flames from 8th to 6th (21% efficiency) in the league.

A few other interesting observations:

• The worst few powerplays in the league (Nashville, Philly, Montreal) are scoring around 13% of the time on the powerplay. The Flames and Coyotes score almost as often when they are shorthanded!
• One thing I have noticed this year is that other teams rarely ever dump it in on the powerplay in spite of Calgary's penchant for trying to hold the blue line - and going the other way to score shorthanded when teams fail to break in. I am guessing having two goalies who actively play the puck is a factor in that, and I wonder how many of our shorties are indirectly because teams are afraid to dump it in.
• Your ability to score while short handed seems to be unrelated to your ability to successfully kill the penalty. On the other hand teams with worse powerplays generally seem to give up more shorthanded goals. That makes sense to me since, outside of the flames and coyotes, teams score between 1-4% of the time while shorthanded vs a swing of 13-30% on the powerplay.

I tried pasting my spreadsheet into this post for stat geeks viewing pleasure, but it was unreadable.

[pokerNhockey]

On other interesting tidbit. The Flames are on pace for 24 shorties this year, which would be 9th all time. the current nineth place team they would be displacing... is the 87-88 Flames. (the mid 1983-89 era Oilers hold 5 of the top 6 spots with only the 87-88 team missing). 8 of the top 10 took place in the 80s.


They would need 26 to tie the 1970-71 bruins for 5th all time. 36 would tie the 83-84 Oilers for first.


https://records.nhl.com/records/team...als-one-season

[redforever]

It seems so odd that our PK is sitting so low because when you watch the Flames on the ice, they are aggressive and always seem to be breaking out for SH goals.

[Jaybo]

Originally Posted by pokerNhockey

I tried pasting my spreadsheet into this post for stat geeks viewing pleasure, but it was unreadable.

Could you upload it to Google Sheets then post a link to that on this thread?

[Huntingwhale]

Having a PK that is a threat to score is so damn sweet. It feels like there's a extra 20 goal scorer on the roster.

[psyang]

Great point about having a good puck-handling goalie causes less dump-ins which creates a greater chance of turnovers and short-handed scoring opportunities. That said, I wonder how many of our goals came from blue line turnovers vs. active sticks or forced errors once they've gained the zone.

[GioforPM]

Originally Posted by redforever

It seems so odd that our PK is sitting so low because when you watch the Flames on the ice, they are aggressive and always seem to be breaking out for SH goals.

I think you probably need to look at some trends - did they rack up a few games with really bad PK performances (say, in the early season). Also I would imagine that the aggressiveness can lead to some great chances for the opposition, if they can beat the aggressive checking. If Janko "goes for it" at the blue line but doesn't make it, it's an extra man open.

There's been at least one time I can recall where they got a shorty but allowed the PP goal as well.

[---Hatrick---]

It sure seems that or 22nd ranked PK is heavily supported by a weak PK in the start of the year. I could be off, but it seems that the PK has been quite effective as of late

[GioforPM]

Looking at the game log it seems like October definitely was the weakest PK month. 4 of the 6 games in which they allowed 2 goals were in October (the others being the odd Dec 29 Sharks game and the recent Bruins loss). In January they've been good.

It's also worth mentioning that the margins are slim. 1% improvement (to 80%) puts Calgary from 21st to 17th place and 2% improvement would put in 13th.

Also, if you consider the shorties as set off against goals allowed on the PK, the Flames then go to only 19 net PK goals against (in 162 opportunities), which is something like 89%. Now, you'd have to similarly adjust all other teams (AZ would come out way ahead) but I suspect Calgary's net total would be near the top of the league.

[mrkajz44]

Not exactly on topic, but I've always felt PP and PK numbers being described in percentages wasn't the best method. Since penalties can be differing lengths (major, minor) and can be shortened when another penalty is taken during a PP, using just a percentage seems misleading. Always wondering if was was expressed something like goals/minute or minutes/goal would yield different results.

[nfotiu]

Originally Posted by mrkajz44

Not exactly on topic, but I've always felt PP and PK numbers being described in percentages wasn't the best method. Since penalties can be differing lengths (major, minor) and can be shortened when another penalty is taken during a PP, using just a percentage seems misleading. Always wondering if was was expressed something like goals/minute or minutes/goal would yield different results.

Majors count as 2 PPs in the stats. Also if you are on the PP, and then take a penalty to shorten it, it should count against the effectiveness of the PP. Also if you score 10 seconds into a PP or give up one in the first 10 seconds, should that be counted as any better or worse?

[Enoch Root]

Originally Posted by nfotiu

Majors count as 2 PPs in the stats. Also if you are on the PP, and then take a penalty to shorten it, it should count against the effectiveness of the PP. Also if you score 10 seconds into a PP or give up one in the first 10 seconds, should that be counted as any better or worse?

Are you sure about that?

My understanding is that it starts as 1, then if you score, a 2nd one starts, and if you score again, a 3rd, etc.

So if you don't score in the 5 minutes, you go 0 for 1. If you score once, you go 1 for 2, if you score twice, 2 for 3, and so on.

[nfotiu]

Originally Posted by Enoch Root

Are you sure about that?

My understanding is that it starts as 1, then if you score, a 2nd one starts, and if you score again, a 3rd, etc.

So if you don't score in the 5 minutes, you go 0 for 1. If you score once, you go 1 for 2, if you score twice, 2 for 3, and so on.

It looks like you are right.

Power-Play Goal
A goal scored by a Club while it has a manpower advantage due to an opponent's penalty. Following are some examples of what is and is not considered a power-play goal:
if a Club has an advantage on a minor penalty starting at 2:02 of the period and it scores at 4:02, the goal is not a power-play goal.
if a Club scores on a delayed penalty, the goal is not a power-play goal.
if a Club has an advantage due to a five-minute major or match penalty, that Club is always credited with having one more advantage than the number of power-play goals it scores during that advantage, because the penalty does not expire. A new advantage begins after each power-play goal. For example, if Team A scores three goals during a major penalty, it is credited with four advantages.
if a Club is on a power-play for any length of time, it is considered to have had an advantage.
if a minor penalty is incurred by a Club on a power-play due to a major penalty, a new advantage is given to that Club when its minor penalty expires, provided the opponent's major penalty is still in effect.

http://www.nhl.com/ice/page.htm?id=26374

I wonder if it changed somewhere along the way. I felt pretty sure they were counted as 2 PPs at some point.

[Enoch Root]

I'm pretty old and this is how I have always understood it to be.

[nfotiu]

Originally Posted by Enoch Root

I'm pretty old and this is how I have always understood it to be.

I'll chalk it up to one of those things my dad told me when I was 7 and never had reason to question

[pokerNhockey]

Originally Posted by GioforPM

Also, if you consider the shorties as set off against goals allowed on the PK, the Flames then go to only 19 net PK goals against (in 162 opportunities), which is something like 89%. Now, you'd have to similarly adjust all other teams (AZ would come out way ahead) but I suspect Calgary's net total would be near the top of the league.

yup, that's what I did. most teams have a handful of shorties, so the only 2 big gainers are the flames and AZ.

Originally Posted by nfotiu

It looks like you are right.

http://www.nhl.com/ice/page.htm?id=26374

I wonder if it changed somewhere along the way. I felt pretty sure they were counted as 2 PPs at some point.

It's counted as n+1 powerplays where n is the number of powerplay goals that get scored on it, since the remaining time is always assumed to be a new powerplay.

[pokerNhockey]

Originally Posted by mrkajz44

Not exactly on topic, but I've always felt PP and PK numbers being described in percentages wasn't the best method. Since penalties can be differing lengths (major, minor) and can be shortened when another penalty is taken during a PP, using just a percentage seems misleading. Always wondering if was was expressed something like goals/minute or minutes/goal would yield different results.

I definitely agree that it is overly simplified, and generally that would probably tell you more, but even using your statistic isn't the whole story. Assuming they are going to get a goal

• If you are down, you are better off getting scored on earlier in the pk (more time to come back)
• If you are protecting a multi-goal lead you are better off being at the end

To me, the best way to do it would only be to count it when games are close. There have been a few pp goals the flames have given up during blowouts in garbage time. It is reasonable to assume that players are looking to the next game - especially when they are winning those games. heck, players probably try a lot harder in a 1-1 game in the third then they do a 6-6 game in the second since that goal probably matters more.