On Power Play Rankings

by renstar on Oct 24, 2009 3:45 AM EDT Comment 17 comments

For quite a few years I have pondered the meaning of the percentages provided as official measure of power play and penalty killing success. To me, they never seemed to be a fair assessment of a team's PP or PK ability because

A PP/PK of any length counts toward the total. Team A, Player 1 takes a tripping call at 19:57 in the 3rd and it counts as a PK when it is inevitably killed off, helping Team B's PK ranking and hurting Team A's PP ranking.
If Team A scores two goals on a PP from a major penalty, are they 1 for 1 or 2 for 1. They aren't 200% on the PP for that game, are they? They were successful on 100% of their PP's but showing them as 1 for 1 is also not really indicative of the game situation.
In last year's painful first period against Calgary, where Tyler Sloan earned the Caps a 9 minute power play, Washington was able to do precisely nothing with it (well, they did manage to turn it into a 7 minute PP). On the official score sheet, this counted as 0 for 3 on the PP (2 minors and 1 major), however the team managed to score no goals on 7 full minutes of PP time.

This last one really got me thinking about this issue and I occasionally dabbled with writing a script to parse the RTSS data and figure out how good a PP team the Caps really are. I was even more motivated (yet still managed to do nothing) when the Caps finished the year ranked 2nd on the PP. One part laziness + one part not knowing how to quantify things + 3 parts grad school forced this to the back burner.

A couple of weeks ago I realized that PP effectiveness might be measured just like the Goals Against Average, i.e., Goals (scored or allowed) per 60 minutes. I've adapted my parser to handle the 2008-2009 and 2009-2010 seasons (2007-2008 will come, they changed the page formatting some), so we can look at the results. Thanks to F&B for the suggestion to look at Goals per 2 min of PP and the number of PPs per Goal. I've added these columns to the tables and rearranged things a little to make it more readable.

First, the 2008-2009 season, since that data is complete and we have a firm idea of where teams should be:

	Old Method				New Method						Comparison
Team	PPG	PP Att	PP %	PP / G	PPG	PP Time	Min / PP	PPG / 60	PPG / 2	PP / G	Old Rank	New Rank	Delta	PP / G Delta
WASHINGTON CAPITALS	85	337	27.27	3.96	85	08:22:27	00:01:29	10.15	0.34	2.96	2	1	1	-1.01
DETROIT RED WINGS	90	348	28.57	3.87	90	09:01:19	00:01:33	9.98	0.33	3.01	1	2	-1	-0.86
SAN JOSE SHARKS	87	360	28.57	4.14	87	09:24:09	00:01:34	9.25	0.31	3.24	3	3	0	-0.9
ANAHEIM DUCKS	73	308	27.50	4.22	73	08:11:30	00:01:36	8.91	0.30	3.37	4	4	0	-0.85
BOSTON BRUINS	74	313	27.27	4.23	74	08:20:18	00:01:36	8.88	0.30	3.38	5	5	0	-0.85
PHILADELPHIA FLYERS	71	316	28.57	4.45	71	08:09:30	00:01:33	8.7	0.29	3.45	6	6	0	-1
BUFFALO SABRES	75	358	27.78	4.77	75	09:23:54	00:01:35	7.98	0.27	3.76	7	7	0	-1.01
ST. LOUIS BLUES	72	351	25.71	4.88	72	09:24:09	00:01:36	7.66	0.26	3.92	8	8	0	-0.96
MINNESOTA WILD	66	328	25.00	4.97	66	08:46:29	00:01:36	7.52	0.25	3.99	9	9	0	-0.98
CANADIENS MONTREAL	72	374	21.62	5.19	72	09:49:25	00:01:35	7.33	0.24	4.09	12	10	2	-1.1
OTTAWA SENATORS	66	339	23.81	5.14	66	09:03:33	00:01:36	7.29	0.24	4.12	10	11	-1	-1.02
CHICAGO BLACKHAWKS	69	360	20.59	5.22	69	09:36:28	00:01:36	7.18	0.24	4.18	13	12	1	-1.04
ATLANTA THRASHERS	69	357	22.86	5.17	69	09:37:34	00:01:37	7.17	0.24	4.19	11	13	-2	-0.99
LOS ANGELES KINGS	69	360	20.00	5.22	69	09:40:46	00:01:37	7.13	0.24	4.21	14	14	0	-1.01
TORONTO MAPLE LEAFS	62	331	20.83	5.34	62	08:47:42	00:01:36	7.05	0.23	4.26	17	15	2	-1.08
CAROLINA HURRICANES	70	374	20.83	5.34	70	10:05:33	00:01:37	6.94	0.23	4.33	18	16	2	-1.02
VANCOUVER CANUCKS	67	357	18.18	5.33	67	09:43:37	00:01:38	6.89	0.23	4.36	16	17	-1	-0.97
NEW JERSEY DEVILS	58	307	19.35	5.29	58	08:25:26	00:01:39	6.89	0.23	4.36	15	18	-3	-0.94
TAMPA BAY LIGHTNING	61	343	18.75	5.62	61	09:12:14	00:01:37	6.63	0.22	4.53	19	19	0	-1.1
PITTSBURGH PENGUINS	62	360	16.67	5.81	62	09:39:59	00:01:37	6.41	0.21	4.68	20	20	0	-1.13
NEW YORK ISLANDERS	54	320	17.07	5.93	54	08:37:57	00:01:37	6.26	0.21	4.80	23	21	2	-1.13
EDMONTON OILERS	60	354	16.67	5.9	60	09:38:44	00:01:38	6.22	0.21	4.82	22	22	0	-1.08
CALGARY FLAMES	61	358	16.67	5.87	61	09:56:38	00:01:40	6.13	0.20	4.89	21	23	-2	-0.98
FLORIDA PANTHERS	51	308	16.22	6.04	51	08:40:02	00:01:41	5.88	0.20	5.10	24	24	0	-0.94
COLORADO AVALANCHE	50	317	14.29	6.34	50	08:40:03	00:01:38	5.77	0.19	5.20	25	25	0	-1.14
NASHVILLE PREDATORS	50	318	14.29	6.36	50	08:44:38	00:01:39	5.72	0.19	5.25	26	26	0	-1.11
DALLAS STARS	54	351	14.29	6.5	54	09:44:37	00:01:40	5.54	0.18	5.41	27	27	0	-1.09
PHOENIX COYOTES	50	341	13.79	6.82	50	09:35:21	00:01:41	5.21	0.17	5.75	28	28	0	-1.07
NEW YORK RANGERS	48	347	12.50	7.23	48	09:21:40	00:01:37	5.13	0.17	5.85	29	29	0	-1.38
COLUMBUS BLUE JACKETS	41	322	9.09	7.85	41	08:47:25	00:01:38	4.66	0.16	6.43	30	30	0	-1.42

The important things to see in this chart are the 2nd and 3rd columns, which contain the PP effectiveness as measured by the league and my method. The 8th and 9th columns have each teams rankings under the two metrics and the last column contains the difference.

So, what do we learn from this? The answer is not much. It appears that those rare arbitrary situations don't affect the final rankings too drastically. But, as fans we like to see that the Caps moved into the #1 spot. (Don't get your hopes up for the PK data, though.) The Devil's drop of three spots is interesting though, I'm not sure what to draw from that.

One thing that we learn is that the Capital's have really short power plays, on average, when compared to the rest of the league. This is clearly due to good goal scoring but I'd be willing to wager some of this is due to taking bad penalties that end PPs early.

For completeness, here is this season's data (through Friday, October 23):

	Old Method				New Method						Comparison
Team	PPG	PP Att	PP %	PP / G	PPG	PP Time	Min / PP	PPG / 60	PPG / 2	PP / G	Old Rank	New Rank	Delta	PP / G Delta
NEW YORK RANGERS	12	44	27.27	3.67	12	01:01:39	00:01:24	11.68	0.39	2.57	7	1	6	-1.1
CALGARY FLAMES	8	28	28.57	3.5	8	00:41:33	00:01:29	11.55	0.39	2.60	2	2	0	-0.9
ATLANTA THRASHERS	8	28	28.57	3.5	8	00:42:46	00:01:32	11.22	0.37	2.67	1	3	-2	-0.83
VANCOUVER CANUCKS	11	40	27.50	3.64	11	01:00:10	00:01:30	10.97	0.37	2.73	5	4	1	-0.9
SAN JOSE SHARKS	12	44	27.27	3.67	12	01:07:46	00:01:32	10.62	0.35	2.82	6	5	1	-0.84
COLORADO AVALANCHE	12	42	28.57	3.5	12	01:10:00	00:01:40	10.29	0.34	2.92	3	6	-3	-0.58
DETROIT RED WINGS	10	36	27.78	3.6	10	00:59:54	00:01:40	10.02	0.33	2.99	4	7	-3	-0.61
PHILADELPHIA FLYERS	9	35	25.71	3.89	9	00:54:31	00:01:33	9.91	0.33	3.03	8	8	0	-0.86
COLUMBUS BLUE JACKETS	8	32	25.00	4	8	00:49:34	00:01:33	9.68	0.32	3.10	9	9	0	-0.9
PHOENIX COYOTES	8	37	21.62	4.63	8	00:55:22	00:01:30	8.67	0.29	3.46	12	10	2	-1.16
MINNESOTA WILD	10	42	23.81	4.2	10	01:09:19	00:01:39	8.66	0.29	3.47	10	11	-1	-0.73
CHICAGO BLACKHAWKS	7	34	20.59	4.86	7	00:49:54	00:01:28	8.42	0.28	3.56	15	12	3	-1.29
EDMONTON OILERS	8	35	22.86	4.38	8	00:58:06	00:01:40	8.26	0.28	3.63	11	13	-2	-0.74
PITTSBURGH PENGUINS	9	45	20.00	5	9	01:07:02	00:01:29	8.06	0.27	3.72	16	14	2	-1.28
TORONTO MAPLE LEAFS	5	24	20.83	4.8	5	00:37:26	00:01:34	8.01	0.27	3.74	13	15	-2	-1.06
LOS ANGELES KINGS	10	48	20.83	4.8	10	01:15:04	00:01:34	7.99	0.27	3.75	14	16	-2	-1.05
WASHINGTON CAPITALS	8	44	18.18	5.5	8	01:08:16	00:01:33	7.03	0.23	4.27	19	17	2	-1.23
NEW YORK ISLANDERS	6	31	19.35	5.17	6	00:53:29	00:01:44	6.73	0.22	4.46	17	18	-1	-0.71
ST. LOUIS BLUES	6	32	18.75	5.33	6	00:53:44	00:01:41	6.70	0.22	4.48	18	19	-1	-0.86
BUFFALO SABRES	6	36	16.67	6	6	00:55:53	00:01:33	6.44	0.21	4.66	21	20	1	-1.34
TAMPA BAY LIGHTNING	7	41	17.07	5.86	7	01:06:15	00:01:37	6.34	0.21	4.73	20	21	-1	-1.13
BOSTON BRUINS	6	36	16.67	6	6	00:57:42	00:01:36	6.24	0.21	4.81	23	22	1	-1.19
DALLAS STARS	7	42	16.67	6	7	01:10:34	00:01:41	5.95	0.20	5.04	22	23	-1	-0.96
CANADIENS MONTREAL	6	37	16.22	6.17	6	01:00:33	00:01:38	5.95	0.20	5.05	24	24	0	-1.12
CAROLINA HURRICANES	6	42	14.29	7	6	01:06:38	00:01:35	5.40	0.18	5.55	27	25	2	-1.45
NEW JERSEY DEVILS	5	35	14.29	7	5	00:55:43	00:01:36	5.38	0.18	5.57	26	26	0	-1.43
FLORIDA PANTHERS	5	35	14.29	7	5	00:56:07	00:01:36	5.35	0.18	5.61	25	27	-2	-1.39
ANAHEIM DUCKS	4	29	13.79	7.25	4	00:46:12	00:01:36	5.20	0.17	5.77	28	28	0	-1.48
OTTAWA SENATORS	4	32	12.50	8	4	00:51:52	00:01:37	4.63	0.15	6.48	29	29	0	-1.52
NASHVILLE PREDATORS	3	33	9.09	11	3	00:58:29	00:01:46	3.08	0.10	9.75	30	30	0	-1.25

Interestingly, the Caps still have a far shorter average PP time than teams that are near them in the rankings. I wish I could explain that somehow. We'll have to see as the season goes on, I'll update my numbers and post them again sometime a month or two from now.

An interesting note from looking at the PP/G data is that the current method for computing PP efficiency seems to overestimate the number of PPs required to produce a goal by more than 1 on average.

Stay tuned for data on penalty kills. That data is processed but I havn't had a chance to look at it in too much detail, that and I'm still figuring out what these data mea, if anything. Additionally, if anyone is curious, I actually have the breakdowns for one- and two-man advantages as well as both 5-5 and 4-4 even strength times, so I can spit those into a table at any time.

If this FanPost is written by someone other than one of the blog's authors, the opinions expressed in it do not necessarily reflect those of this blog or SB Nation.

13 recs | Comment 17 comments | Add comment

Story-email Email Printer Print

Comments

Really good stuff. I like this approach a lot.

If Team A scores two goals on a PP from a major penalty, are they 1 for 1 or 2 for 1. They aren’t 200% on the PP for that game, are they? They were successful on 100% of their PP’s but showing them as 1 for 1 is also not really indicative of the game situation.

I think the percentages are calculated as Goals Scored/Times On Power Play or Goals Allowed / Times Shorthanded. So yeah, that does come out as 200%. It’s a weird result, but I think that’s how they do it.

Atta dinnin stick a who!

by Gould Old Days on Oct 24, 2009 12:43 PM EDT reply actions 0 recs

Thanks! I’m intrigued that it doesn’t change the rankings much, but I’d like to think it provides a better measure of teams. If two teams have 10 power plays and each score 10 goals, the percentages are gonna be the same, but if one does it 30 seconds in every time and the other 1 minute every time, the team that scores faster will be rated higher under this metric.

by renstar on Oct 24, 2009 12:48 PM EDT up reply actions 0 recs

Actually, it counts as — wait for it — 2-for-3. My understanding is that every time you score on a power play that doesn’t expire as a result, a new “power play” starts for the purposes of calculating percentages. If you scored ten goals on the five-minute major, you’d be 10-for-11, because you scored ten times and then the penalty ended without you scoring again. Does that make sense? This link from the Sabres’ site (first thing I found on Google) corroborates this and adds a little more.

by sixsevenfiftysix on Oct 24, 2009 1:05 PM EDT up reply actions 0 recs

And here’s that link I was talking about…

Clicky

by sixsevenfiftysix on Oct 24, 2009 1:08 PM EDT up reply actions 0 recs

Great link, I wish I had that last night when writing this. That is so dumb.. if the opposing team commits a major penalty, you can never go 100% on the PP.

by renstar on Oct 24, 2009 1:09 PM EDT up reply actions 0 recs

At least part of the shorter Caps power plays this season has been taking penalties while on the PP, we’ve managed to cut quite a few short thus far.

That and we’ve had some very fast power-play conversions and some brutal attempts where our boys didn’t do much more than chase short-handed opportunity after short-handed opportunity.

by Knee high to a duck on Oct 24, 2009 1:05 PM EDT reply actions 0 recs

If we’re talking about improving how power play efficiency is measured (which we absolutely should be, and this is a good start), another thing to keep in mind is short-handed goals against. I believe there was something on Puck Daddy before the season started about this — if you figure a short-handed goal as a “minus one-for-one” or something to that effect, some teams look a lot better and some look a lot worse.

by sixsevenfiftysix on Oct 24, 2009 1:09 PM EDT reply actions 0 recs

I’ve been thinking about that, I haven’t figured out how to take shorties into effect. I imagine they are so rare statistically that subtracting a goal from it would only knock a tenth or two off of any team (unless they are particularly bad). Though a team like Philly (?) would probably get a huge boost on their PK.

by renstar on Oct 24, 2009 1:11 PM EDT up reply actions 0 recs

I am rec’ing the hell out of the work. Now for my thoughts:

Short PPs: I think a lot of it is that we keep the same unit out for 2 full minutes. Thus, even if we don’t take a PIM to kill the PP, we are far less effective as the PP moves on. We are freshest and deadliest right off the bat, and the goals reflect that. More PP2 please.

I would measure the PP efficiency either by “Goals per 2 minutes of PP time” or “Minutes of PP time per goal.” 2 minutes is the normal PP so measuring it that way tells you roughly how many goals you are going to give up every time you put a particular team on the PP. Minutes per goal just tells you how long it takes a team to score a PP goal. So if it was 3:46 minutes per PP goal you would have a rough idea that if you take 2 penalties (4 PIM) against a team you are going to get scored on.

I don’t think you can/should fold SHGs straight into PP efficiency; it’s good to know what the PP efficiency is as a stand alone stat. I would just make that a new stat and subtract total SHG from total PPP which gives you your basic PP +/- as a team. Then you can do the average I suggested above (Min/SHG or SHG/2 min) and then subtract them from your PP efficiency number and find the adjusted PP efficiency.

If consequences dictate the course of action, then it doesn't matter what's right, it's only wrong if you get caught. If consequences dictate the course of action, then I should play God...

by Fehr and Balanced on Oct 24, 2009 1:21 PM EDT reply actions 0 recs

Thanks!

I don’t see how using one tired unit would shorten the PP time. I’d think it would make it longer, in general. If goal’s aren’t scored early, they aren’t likely to be scored. Hmm… histogram of goals scored in which 5 second bin of a PP…. by team…

I considered doing it as a Goals / 2 min PP and then went against it as it is just a scaling factor. I kinda thought that large whole numbers would be easier to interpret. Looking at last years data, the caps would be roughly 0.33 goals per PP or 3 PPs per goal. I can add those columns if people feel it would be a better measure.

As for SHGs, the way I would handle those is to try and develop a metric for overall PP efficiency which would include SHGs as well as some measure of the PP time lost to bad penalties that take the game to 4v4. I only started thinking about doing that this morning, so we’ll see what direction that takes.

by renstar on Oct 24, 2009 1:28 PM EDT up reply actions 0 recs

Yeah you are right re: one PP unit. Brain far. I was thinking about why the PP is overwhelmingly more effective early on.

The math is relatively straightforwad so whatever scale you use is fine. I’m just partial to the 2 min scale because that’s what you see in games so it gives you a better idea of what you are looking at in any given PP.

Good look on the wholistic metric. Sounds interesting and is probably the most comprehensive way to view how good a PP really is.

If consequences dictate the course of action, then it doesn't matter what's right, it's only wrong if you get caught. If consequences dictate the course of action, then I should play God...

by Fehr and Balanced on Oct 24, 2009 1:38 PM EDT up reply actions 0 recs

OK, I updated the tables to include the two columns you were interested in. It actually illuminates the difference in the two metrics. The old one is essentially overestimating the number of power plays required to produce a goal by more than 1.

by renstar on Oct 24, 2009 3:16 PM EDT up reply actions 0 recs

Fehr and Balanced

Do you want the mustache on or off?
Too bad.

by Marshall Pirate on Oct 28, 2009 4:27 PM EDT up reply actions 0 recs

Exactly.

If consequences dictate the course of action, then it doesn't matter what's right, it's only wrong if you get caught. If consequences dictate the course of action, then I should play God...

by Fehr and Balanced on Oct 28, 2009 5:05 PM EDT up reply actions 0 recs

great work. highly rec’d.

by CP2Devil on Oct 25, 2009 12:04 PM EDT reply actions 0 recs

Thanks! I’ll try to get more actual analysis in the next post

by renstar on Oct 25, 2009 5:07 PM EDT up reply actions 0 recs

Stats definitely bring up a number of issues for analysis.

I’d like to see how Caps do on PP’s that last over a minute. It seems like when they score they do it quickly as you noted. Is it a feast or famine thing (in sync vs not in sync) or is it due to the fact that BB plays his 1st unit for a long period of time on PP therefore they wear down? Is it a combination of the two? Is there a another factor that isn’t as obvious? Just something I have pondered over the last year plus.

by CP2Devil on Oct 26, 2009 9:43 AM EDT up reply actions 0 recs

Join the community and dive in!

Search

FanPosts

Community blog posts and discussion.

Recommended FanPosts

Rejected Hockey Card - Matt Bradley

by Chris meet Alex 2 days ago

32 comments | 15 recs

Recent FanPosts

+ New FanPost All FanPosts >

FanShots

Quick hits of video, photos, quotes, chats, links and lists that you find around the web.

+ New FanShot All FanShots >

Southeast Standings

	GP	W	L	OTL	PT
Washington Japers' Rink	21	13	4	4	30
Tampa Bay Raw Charge	19	8	4	7	23
Atlanta Bird Watchers Anonymous	18	10	6	2	22
Florida Litter Box Cats	19	8	9	2	18
Carolina Canes Country	21	4	12	5	13

(updated 11.20.2009 at 9:16 AM EST)

Japers' Rink: An SB Nation Community

Network Bar

Japers' Rink -