Philadelphia Flyers Scoring Chance Summary: A Half Season Review

by ToddtheFox on Feb 7, 2012 12:00 PM EST in Philadelphia Flyers Scoring Chances

Many regular readers of BSH know that many here put a lot of stock into the so-called "advanced statistics" -- both as a way to understand what we're watching on the ice night in and night out, and as a way to predict future events.

For example, Fenwick tied -- a measure of unblocked shots directed at the net while the team is tied -- is known to be an excellent way of estimating future winning percentage. This has been shown several times over the last few years, like the Devils getting more points than the Dallas Stars over the last half of the season, and the Minnesota Wild being awful this year despite spending quite some time at the top of the Western Conference.

Peter Laviolette may be unaware of the statistical powers of Fenwick numbers, but we know that he puts a lot of stock into scoring chances -- and we also know that there is a correlation between Corsi/Fenwick and scoring chances. You can decide whether or not you put any stock in Fenwick numbers, but scoring chances are simple to understand and their importance is pretty obvious.

That's why we've been tracking them here at BSH over the course of the season, and it's time to get caught up on some trends over the first half of this year.

Those who want to read more about the importance of scoring chances should read this or this.

Scoring chance definition

A scoring chance is defined as a clear play directed toward the opposing net from a dangerous scoring area -- loosely defined as the top of the circle in and inside the faceoff dots (see below), though sometimes slightly more generous than that depending on the amount of immediately-preceding puck movement or screens in front of the net.

Blocked shots are generally not included but missed shots are. A player is awarded a scoring chance anytime he is on the ice and someone from either team has a chance to score. He is awarded a "chance for" if someone on his team has a chance to score and a "chance against" if the opposing team has a chance to score.

Scoring chance area

Simply put, I record every time an unblocked shot is taken from within the Scoring Chance Area -- using a script made by Vic Ferrari -- and then using that data we can see how the team and individuals are performing.

Before we get to the charts, I should note that none of this includes the Winter Classic due to script issues.

Total team chances: Games 1 through 41

Period	Total For	Total Against	Total %	EV For	EV Against	EV %	PP CF	PP CA	SH CF	SH CA
1	230	181	56%	146	137	52%	73	5	11	43
2	229	197	54%	149	153	49%	66	5	14	49
3	205	172	54%	162	136	54%	33	4	10	42
4	14	10	58%	7	5	58%	7	0	0	5
Totals	678	560	55%	464	431	52%	179	14	35	139

This table is shown from the Flyers perspective, so SH CF/CA indicates the totals while the opponent is on the powerplay.

The Flyers are getting 55 percent of the total chances and 52 percent of chances at even strength. Outchancing opponents = good.
The powerplay is getting approximately one chance every opportunity. A lot of the powerplay shots are from above the top of the scoring chance area and thus are not included.
The powerplay has allowed 14 total shorthanded chances. This is probably surprisingly low for people who think having four forwards on the powerplay is leading to many shorthanded rushes.
The penalty kill is still really good. The chances against are below one per kill and they get a lot of shorthanded chances

Individual Even-Strength Chances: among players with over 100 minutes

	ES TOI	ES CF	ES CA	Chance %	ES C/60	ES CA/60	SC +/- per 60
M. CARLE	725	204	161	56%	16.87	13.32	3.56
C. PRONGER	178	46	36	56%	15.48	12.11	3.37
W. SIMMONDS	486	123	98	56%	15.17	12.09	3.08
J. JAGR	428	123	102	55%	17.26	14.31	2.95
J. VAN RIEMSDYK	431	110	91	55%	15.32	12.67	2.65
J. VORACEK	484	148	127	54%	18.35	15.75	2.60
C. GIROUX	526	152	131	54%	17.33	14.93	2.39
K. TIMONEN	560	132	112	54%	14.15	12.01	2.14
S. HARTNELL	567	158	141	53%	16.71	14.92	1.80
B. SCHENN	136	37	33	53%	16.28	14.52	1.76
D. BRIERE	532	134	128	51%	15.12	14.44	0.68
M. READ	439	103	99	51%	14.08	13.53	0.55
A. MESZAROS	664	165	159	51%	14.91	14.37	0.54
S. COUTURIER	336	72	69	51%	12.86	12.33	0.54
M. TALBOT	474	121	118	51%	15.33	14.95	0.38
M. BOURDON	300	71	72	50%	14.21	14.41	-0.20
B. COBURN	707	170	176	49%	14.43	14.94	-0.51
A. LILJA	344	76	82	48%	13.27	14.32	-1.05
H. ZOLNIERCZYK	196	34	47	42%	10.40	14.38	-3.98
Z. RINALDO	211	22	48	31%	6.26	13.65	-7.39

ES TOI, even-strength time on ice; ES CF/CA, total even-strength chances for/against; chance%, percentage of chances for when on the ice ES C/60, number of even-strength chances per 60 minutes of ice time; SC +/- per 60, chances for-chances against per 60 minutes of ice time.

If I had run a competition to see who the top four forwards as far as chances per 60, how many would have picked Simmonds, Jagr, JvR and Voracek? And in that order?
Matt Carle does not deserve your criticism.
Couturier, Talbot and Lilja have low offensive zone starts, so should probably be given a bit of leeway when looking at their chances.
Giroux is starting in the offensive zone just 46 percent of the time. I don't know what that number was at the start of the season but it is a potential reason for his reduction in chances for per 60 (which was 8.68 for the first 11 games).

Even Strength Chances: players with under 100 minutes

	EV TOI	EV For	EV Against	Chance %	ES C/60	ES CA/60	SC +/- per 60
E. WELLWOOD	13	5	3	63%	22.47	13.48	8.99
E. GUSTAFSSON	84	22	17	56%	15.70	12.13	3.57
A. NODL	99	18	13	58%	10.87	7.85	3.02
T. SESTITO	24	4	4	50%	10.06	10.06	0.00
M. WALKER	43	15	16	48%	20.81	22.20	-1.39
K. MARSHALL	87	27	30	47%	18.68	20.76	-2.08
J. SHELLEY	64	10	20	33%	9.34	18.68	-9.34
B. HOLMSTROM	11	0	4	0%	0.00	21.82	-21.82

Gustafsson was playing well before his injury, so it will be interesting to see further data on how he's played since his return.
Nodl barely missed the cutoff, but he was good in his limited time with the Flyers.

PP Individual Chances For (players with low TOI have been removed)

	PP TOI	PP CF	PP CF/2min
W. SIMMONDS	138	92	1.33
J. JAGR	139	92	1.32
C. GIROUX	146	96	1.31
C. PRONGER	51	32	1.27
J. VORACEK	141	81	1.15
J. VAN RIEMSDYK	104	58	1.12
K. TIMONEN	153	85	1.11
M. READ	107	59	1.11
D. BRIERE	142	74	1.04
S. HARTNELL	122	63	1.03
A. MESZAROS	61	30	0.99
M. CARLE	124	58	0.93

PP TOI, powerplay time on ice; PP CF, total chances for; PP CF/2min, powerplay chances for over 2 minutes

The most surprising thing here is that Carle is the worst performer on the powerplay.

PK Individual Chances Against

	SH TOI	SH CF	SH CF/2min	SH CA	SH CA/2min
A. LILJA	17.45	2	0.23	4	0.46
M. CARLE	63.02	5	0.16	17	0.54
S. COUTURIER	100.48	15	0.30	32	0.64
M. READ	99.40	12	0.24	34	0.68
J. VORACEK	26.25	1	0.08	10	0.76
A. MESZAROS	104.90	14	0.27	40	0.76
B. COBURN	128.73	19	0.30	51	0.79
M. TALBOT	148.63	21	0.28	60	0.81
C. PRONGER	58.80	7	0.24	26	0.88
C. GIROUX	93.77	14	0.30	43	0.92
K. TIMONEN	147.25	21	0.29	68	0.92

SH TOI, shorthanded time on ice; SH CF, shorthanded chances for; SH CF/2min, shorthanded chances for over 2 minutes; SH CA, shorthanded chances against, SH CA/2min, shorthanded chances against over 2 minutes.

I've included the chances-for while shorthanded to show how good the penalty kill is at generating chances a man down.
Initially I thought there was a quality of competition effect with Timonen, Giroux, Talbot and Coburn taking on the opponent's top PP line and Meszaros, Read, Couturier the main second PK unit. The players Corsi Rel QoC doesn't seem to support this theory though.

Goalies

	ES TOI	ES CF	ES CA	Chance %	PP TOI	PP CF	PP CF/2min	SH TOI	SH CA	SH CA/2min
I. BRYZGALOV	1248	319	292	52%	200	112	1.12	182	68	0.75
S. BOBROVSKY	583	139	139	50%	101	59	1.17	83	36	0.87

This was included just to show that the Flyers have been better with Bryz in net at even strength and while shorthanded. Confidence or something, right?

As always, there is plenty here that I haven't discussed so point them out in the comments. Also, if anyone would like the complete set of data, we'll be happy to post it, and if anybody has any ideas on how to better examine it, let us know in the comments

Tweet Comment 71 comments | Add comment | 4 recs |

Do you like this story?

Comments

You reference it, and it’s easy enough to find, but in the future adding in zone starts and maybe QoC metrics as columns might be helpful for those that want more context but are lazy.

being obnoxious and self righteous while ignoring the point since 9/29/11

by DLJr on Feb 7, 2012 12:14 PM EST reply actions

All I can say is that Pronger was absolute killing it out there and Carle hasnt suffered at all from playing with Bourdon.

by Anders Jensen on Feb 7, 2012 12:40 PM EST reply actions

Great work, Todd. Especially on the introduction.

It’s definitely surprising that Carle is so good at even strength but so bad on the power play. I wish I had an explanation.

Voracek having the most ECF/60 and ECA/60 is sterotypical. And yet I’m not mad at all.

Man-crushin' on Boucher since 1999 and Matt Calvert since May 2010
Broad Street Hockey - Makin' it look mean since 1967.
SB Nation Philly - Associate Editor

by Geoff Detweiler on Feb 7, 2012 12:56 PM EST reply actions

Why don’t shots from the point count? Seems odd.

"Because wives and girlfriends aren’t on the road."

by BannedStreetBully on Feb 7, 2012 1:31 PM EST reply actions

Any goalie worth their weight in $51m should be stopping 99%+ of unscreened shots from the point.

Blueshirt Banter - Where Rangers' Fans Matter
Tracking the Rangers - Numbers don't lie. They just don't agree with you.
Twitter: RangerSmurf
"Oh, that sensible and sober* Rangers fan guy who is cool, actually" - Dominik, Lighthouse Hockey
*Statement has not been verified nor regressed

by George E. Ays on Feb 7, 2012 2:03 PM EST up reply actions

That’s it?

"Because wives and girlfriends aren’t on the road."

by BannedStreetBully on Feb 7, 2012 2:14 PM EST up reply actions

That's the reason they're not included, yes.

They’re not ‘dangerous’ shots in and of themselves. Most coaches would be quite pleased in the defensive performance if every shot was coming from 60-65 feet away.

by George E. Ays on Feb 7, 2012 2:47 PM EST up reply actions

and presumably if there is a screen it’s not fair to credit the shooter for the quality of the chance. Isn’t there a “quality chances” stat that is more subjective (not restricted by the objective circles/hashmarks limit)? Does the stat crowd give this stat any credence?

/s, more often than not

by flyersfaninchicago on Feb 7, 2012 4:46 PM EST up reply actions

it’s not fair to credit the shooter for the quality of the chance

This reminds me of something I wanted to ask earlier.

I think — from looking at the totals — that when it says “individual chances” above, it means on-ice chances for each individual, not individual shooting chances, right? So Matt Carle hasn’t taken 204 ES shots that were counted as scoring chances; he’s been on the ice for 204 ES shots by the Flyers that were counted as scoring chances?

@BSH_EricT
Writer at Broad Street Hockey

by Eric T. on Feb 7, 2012 4:52 PM EST up reply actions

Thats how I read it, else how would you count a chance against a player?

by Anders Jensen on Feb 7, 2012 4:54 PM EST up reply actions

Haha, good point. “Dammit, Brayden, stop shooting at our goal!”

@BSH_EricT
Writer at Broad Street Hockey

by Eric T. on Feb 7, 2012 5:03 PM EST up reply actions

Maybe someone should tell Dan Boyle that?

by cova24 on Feb 7, 2012 5:56 PM EST up reply actions

Well, I dunno about fair, but if there’s a significant enough screen, I would count it as a scoring chance, and give credit to the player who took it.

Isn’t there a "quality chances" stat that is more subjective (not restricted by the objective circles/hashmarks limit)?

Not that I’m aware of.

by George E. Ays on Feb 7, 2012 6:28 PM EST up reply actions

I very rarely count screened chances at all. Unless you’ve got a goalies-eye-view, it’s really hard to even say the screen was successful.

Tracking the Flyers scoring chances at Broad Street Hockey

by ToddtheFox on Feb 7, 2012 6:42 PM EST up reply actions

I usually base it on the goalie reaction. If there is none, I feel pretty confident he didn’t see it. There’s definitely a heavy bias towards goals in that situation as well, as those usually give you 4-5 more replay angles to determine the screen.

by George E. Ays on Feb 8, 2012 1:06 PM EST up reply actions

I what to petition that all of Nodl’s shots shouldn’t count as a scoring chance because Nodl is the shooter.

Nice Work Todd

Nodl 12G 127GP, Talbot 10G 44GP

by chrislanci on Feb 7, 2012 1:36 PM EST reply actions

Still trolling about that, are we?

Lightning strikes once, Hextall strikes twice!
"I think there is virtue in pissing off idiots." - Fehr and Balanced

by hintzy64 on Feb 7, 2012 1:41 PM EST up reply actions

This is really well done. I’m wondering if it might be possible to impose a simple weighting system for scoring chances for/against that adjusts them based on %OZ starts and QoC. I’m not sure how tough that would be, as long as methodology was explained.

Another thought… Maybe it would be useful to split up the data set into score situations (i.e. tied, trailing by one, up by one) and see if there is any deviation in individual chances for/against. It might reveal help reveal clutch and dud players.

Awesome stuff!

by HeffToBeMad on Feb 7, 2012 1:40 PM EST reply actions

Maybe it would be useful to split up the data set into score situations (i.e. tied, trailing by one, up by one) and see if there is any deviation in individual chances for/against. It might reveal help reveal clutch and dud players.

I’m skeptical that the data would be meaningful if you divide it up like this — you’d end up with tens of chances in each bin, with each individual chance making a difference of a couple percent.

But I’m definitely up for looking at how the team rates change with score.

Awesome stuff!

Agree completely.

@BSH_EricT
Writer at Broad Street Hockey

by Eric T. on Feb 7, 2012 3:35 PM EST up reply actions

I’m hoping to update it now that we have a lot more data on the subject, but using two teams (NYR/WSH) worth of data last year, I came up with each offensive zone start being worth 0.425 chances.

It fits in line with Corsi (OZ = 0.8 Corsi) and Fenwick (OZ = 0.6), so it passed the first sniff test. I think it probably gets closer to 0.4 once you start looking at the bigger sample, but if you want a quick and dirty translation factor here, that’s what I’d go with.

by George E. Ays on Feb 7, 2012 3:50 PM EST up reply actions

Something I always caution with this analyses (and you’ve probably heard me say it before): this sort of correlation probably overemphasizes the zone start as the cause.

If players who are good at offense tend to get more offensive zone starts and also produce more chances, it would make the regression spit out too high of a number for the correlation between zone starts and chance rate. Whether that’s a tiny effect or a significant one is tough to guess, but it’s worth noting that a team’s Corsi is about +40 (per 60 min) in the time between an OZ faceoff and a player going off the ice, so if that period of time is typically something like 40 seconds, then the OZ faceoff was worth about 0.4-0.5 shots — which is reasonably close to the number that came out of our zone entry work.

@BSH_EricT
Writer at Broad Street Hockey

by Eric T. on Feb 7, 2012 4:05 PM EST up reply actions

Fair enough.

I just redid the math that JLikens and I beileve Vic did before that in correlating Fenwick/Corsi to zone starts. If they had a flaw (and it seems you’re saying they did), then my methodology would as well.

by George E. Ays on Feb 7, 2012 4:30 PM EST up reply actions

Yeah, I think the best way to do it would be to compare at the individual level and report the average across players for the difference between their Corsi in an OZ shift and their Corsi in a DZ shift.

Until that’s done, I’m OK with a) using the 0.8 standard but remembering that it might be high, b) using the 0.45 number from DP’s work but remembering it’s not what most people do, or c) treating OZ% qualitatively as part of context, the way people tend to do with qualcomp metrics, rather than making a quantitative adjustment.

@BSH_EricT
Writer at Broad Street Hockey

by Eric T. on Feb 7, 2012 4:40 PM EST up reply actions

As Eric and George alluded to, I think this will be stuff I do at the end of the season when the sample size is large enough. Score effects is something that I’ll get Eric on to soon

Tracking the Flyers scoring chances at Broad Street Hockey

by ToddtheFox on Feb 7, 2012 6:43 PM EST up reply actions

The most surprising thing here is that Carle is the worst performer on the powerplay.

Carle’s Corsi for PP time is 73.16 / 60 or about 2.44 / 2 min. Timonen meanwhile has about 15 more shot attempts per 60 minutes of PP time for a 88.68 / 60 or about 2.93 / 2 min.

Timonen is creating chances at 1.11 / 2.93 attempts (37.8%), while Carle is doing the same at a rate of 0.93 / 2.44 attempts, which is actually a rate of success higher than Timonen’s at 38.1%.

Timonen has mostly been deployed with Giroux, Voracek, Simmonds and Jagr/Hartnell, while Carle has alternated with Meszaros on the second unit, who have seemed to me to be holding the puck longer and spending more time behind the net and on the side boards.

Maybe I misinterpreted one of the stats to mean something other than what I thought. Is the link here between Corsi and PP chances missing some other data? Could we expect Carle to surpass Timonen in PP CF if he (and the rest of PP Unit 2) would simply try to take more shots?

by hebrew hammer on Feb 7, 2012 1:46 PM EST reply actions

SHOOOOOOOOOOOOOOOOOOOOT!!!!

Man-crushin' on Boucher since 1999 and Matt Calvert since May 2010
Broad Street Hockey - Makin' it look mean since 1967.
SB Nation Philly - Associate Editor

by Geoff Detweiler on Feb 7, 2012 1:49 PM EST up reply actions

That is interesting. Of course, he is the “worst performer” because he shoots less, so it’s somewhat circular.

Tracking the Flyers scoring chances at Broad Street Hockey

by ToddtheFox on Feb 7, 2012 6:48 PM EST up reply actions

I actually would have picked Jagr and Voracek to be near the top in chances … but then I’m partial to European players. I also know how Jagr loves to shoot the puck.

The most interesting line I see is the one that really details how badly Briere is slumping … +0.68 chances per 60? Ewww. Time to trade him and his $6.5 million cap hit, so we can replace Pronger in the offseason.

Maxime Talbot - in the Orange and Black ... better than chocolate and peanut butter!

by MaximumTalbot on Feb 7, 2012 1:47 PM EST reply actions

If the new CBA give the team option for 1 amnesty buyout (or what it is called), Briere should be one of the options consider his cap hit (largest cap hit on the team), if they can not trade him

by Anders Jensen on Feb 7, 2012 1:51 PM EST up reply actions

….Aaaand I just replied to you in the other thread saying something virtually the same.

Maxime Talbot - in the Orange and Black ... better than chocolate and peanut butter!

by MaximumTalbot on Feb 7, 2012 2:04 PM EST up reply actions

Yeahhhh, no. If Pronger doesn’t look like he’s coming back then you’re an idiot to not buy him out. Yeah let’s keep a 35+ Contract on the books for a billion years instead of an at least proven scorer.

by captj on Feb 7, 2012 3:27 PM EST up reply actions

If the LTIR rules stay the same, there’s an argument that Pronger’s cap hit stashed on LTIR is good enough to buy out someone else.

Man-crushin' on Boucher since 1999 and Matt Calvert since May 2010
Broad Street Hockey - Makin' it look mean since 1967.
SB Nation Philly - Associate Editor

by Geoff Detweiler on Feb 7, 2012 3:33 PM EST up reply actions

boom.

Lightning strikes once, Hextall strikes twice!
"I think there is virtue in pissing off idiots." - Fehr and Balanced

by hintzy64 on Feb 7, 2012 3:35 PM EST up reply actions

My thought aswell. I highly doubt the Flyers would do such thing, but I also doubted that they would trade Richard and Carter.

by Anders Jensen on Feb 7, 2012 3:45 PM EST up reply actions

This still causes a hassle every offseason, in my opinion, because the cap can only be exceeded by 10% and you would eat up a lot of that overage with Pronger’s cap hit. It’s still manageable, but not ideal. I think I’d rather buy out Bryz because it was such a wasteful contract to begin with, but I could be sold on the Briere idea too, especially if the cap contracts.

by hebrew hammer on Feb 7, 2012 3:47 PM EST up reply actions

Not ideal, but not idiotic, as was suggested (with heavy snark, I might add) above.

Lightning strikes once, Hextall strikes twice!
"I think there is virtue in pissing off idiots." - Fehr and Balanced

by hintzy64 on Feb 7, 2012 3:51 PM EST up reply actions

It’s so sad that we have more than one albatross contract that is currently biting us in the ass.

by hebrew hammer on Feb 7, 2012 4:02 PM EST up reply actions

Right?

Lightning strikes once, Hextall strikes twice!
"I think there is virtue in pissing off idiots." - Fehr and Balanced

by hintzy64 on Feb 7, 2012 5:28 PM EST up reply actions

And it creates bonus cushion problems as well.

I’m not advocating one over the other, simply refuting the claim that choosing one over the other is idiotic.

Man-crushin' on Boucher since 1999 and Matt Calvert since May 2010
Broad Street Hockey - Makin' it look mean since 1967.
SB Nation Philly - Associate Editor

by Geoff Detweiler on Feb 7, 2012 4:14 PM EST up reply actions

Right. There may not even be a buy-out clause, so who knows.

by hebrew hammer on Feb 7, 2012 4:21 PM EST up reply actions

I’m sad that there’s no bubble graph :(

I heart our rookies.

by LeepinLizardz on Feb 7, 2012 2:46 PM EST reply actions

LL likes her visual representation of numbers, damnit! :-)

"I wouldn’t run if there was a fire. I wouldn’t run anywhere. I hate running." - O. Munn

by doubleh on Feb 7, 2012 3:09 PM EST up reply actions

Who doesn’t?

Lightning strikes once, Hextall strikes twice!
"I think there is virtue in pissing off idiots." - Fehr and Balanced

by hintzy64 on Feb 7, 2012 3:11 PM EST up reply actions

Yeah! I feel guilty because I couldn’t understand the charts with just numbers. I was like “Stop glazing over, Eyes! Todd put a lot of work into this!” But I had to give up eventually and just skipped to Todd’s conclusions.

I heart our rookies.

by LeepinLizardz on Feb 7, 2012 3:12 PM EST up reply actions

I am same. I work in the creative side of business, so Excel is my enemy. I like more freeform programs.

"I wouldn’t run if there was a fire. I wouldn’t run anywhere. I hate running." - O. Munn

by doubleh on Feb 7, 2012 3:50 PM EST up reply actions

Silly non-maths people :)

Maxime Talbot - in the Orange and Black ... better than chocolate and peanut butter!

by MaximumTalbot on Feb 8, 2012 12:55 PM EST up reply actions

Don’t worry about it :)

I’m still not really sure what is the most important information in the charts, so I don’t want to miss presenting something by only including a graph. I’ll work on doing both in the future.

Tracking the Flyers scoring chances at Broad Street Hockey

by ToddtheFox on Feb 7, 2012 6:50 PM EST up reply actions

Technically a scatter plot, not a bubble graph, but hopefully adequate

It’s interesting that the spread in chances for is much larger than the spread in chances against — particularly for the forwards (meaning it’s particularly true for them, not that it’s particularly interesting for them).

I want to look at that and conclude that forwards’ offensive contributions matter much more than their defensive contributions. I can’t quite convince myself that confounding effects of matchups and zone starts aren’t problematic though. It seems like those effects would impact chances for and chances against equally, but I’m not sure that’s true.

@BSH_EricT
Writer at Broad Street Hockey

by Eric T. on Feb 7, 2012 3:58 PM EST up reply actions

Also really interesting that Coburn and Timonen are D partners and still have the widest spread amongst the D-men for chances against.

by hebrew hammer on Feb 7, 2012 4:05 PM EST up reply actions

I thought that a forward had no real impact on his goalies sv%, so isnt it kinda logic that he have no really impact on chance against(assuming there is correlation between sv% and scoring chances) ?

by Anders Jensen on Feb 7, 2012 4:36 PM EST up reply actions 1 recs

This is an excellent point, and is arguably true for defensemen as well.

Hmmmmm.

@BSH_EricT
Writer at Broad Street Hockey

by Eric T. on Feb 7, 2012 4:41 PM EST up reply actions

Possibility 1: skaters have no impact (in the long term, of course) on what kind of shots are taken, only on how many, and we should replace chances against with (shots against) * (league average fraction of shots that are chances) to reduce noise.

Possibility 2: skaters have impact on what fraction of shots are chances, but it is small enough that it is overwhelmed in a given year by random variance, and we should replace chances against with (shots against) * (career average fraction of shots that are chances) once we have career data on such things.

Possibility 3: skaters have impact on what fraction of shots are chances that recurs from year to year, and the only reason we don’t see it in goalie save percentage is because the added goalie variability makes it disappear.

I’m open to other suggestions.

@BSH_EricT
Writer at Broad Street Hockey

by Eric T. on Feb 7, 2012 4:46 PM EST up reply actions

I think it is a combination of 1 and 3. I think the only real impact a forward has is how many shots on goal he can prevent, because even the best forwards have no control over rebounds, redirected shots and funny bounces.

by Anders Jensen on Feb 7, 2012 5:00 PM EST up reply actions

Where is Rinaldo? Without him, Scrabble looks really bad on that graph.

Lightning strikes once, Hextall strikes twice!
"I think there is virtue in pissing off idiots." - Fehr and Balanced

by hintzy64 on Feb 8, 2012 9:32 AM EST up reply actions

___

Hahaha, Rinaldo sucks so bad that I accidentally chopped him off the graph without noticing.

Here’s the corrected graph:

@BSH_EricT
Writer at Broad Street Hockey

by Eric T. on Feb 8, 2012 12:50 PM EST up reply actions 1 recs

OUCH.

I’m not seeing #45 either ….

Maxime Talbot - in the Orange and Black ... better than chocolate and peanut butter!

by MaximumTalbot on Feb 8, 2012 12:57 PM EST up reply actions

Shelley didn’t make the 100-minute threshold. But he’d be at (9.3, 18.7) — a little lower in chances for than Zolnierczyk, and way higher than anyone else in chances against.

@BSH_EricT
Writer at Broad Street Hockey

by Eric T. on Feb 8, 2012 12:59 PM EST up reply actions

I’m not sure which disturbs me more – how bad those numbers are for Girlyname, or the fact that BSC has gotten more minutes this season.

But thanks for the answer nonetheless.

Maxime Talbot - in the Orange and Black ... better than chocolate and peanut butter!

by MaximumTalbot on Feb 8, 2012 1:02 PM EST up reply actions

It’s interesting that they are each uniquely bad.

Rinaldo has the same defense than Zolnierczyk but with much worse offense, and Shelley has the same offense as Zolnierczyk but with much worse defense.

We should slice them both in half and sew the good parts together to create a second Zolnierczyk, who can then feast on the flesh of the bad halves. (Note for those who are concerned about wishing harm on actual humans: I do not mean for this comment to be taken literally, and do not wish to see any living beings cut in half. Even in a magic show, to be honest.)

@BSH_EricT
Writer at Broad Street Hockey

by Eric T. on Feb 8, 2012 1:19 PM EST up reply actions

so then “JodyShelley’sFrankenstein” would be prescient, a hipster.

/s, more often than not

by flyersfaninchicago on Feb 8, 2012 1:21 PM EST up reply actions 1 recs

There shouldn’t be a problem, as long as the magician isn’t cheating on the head with the legs, or the legs and head aren’t talking, or the legs are actually the magician’s nephew.

I heart our rookies.

by LeepinLizardz on Feb 8, 2012 1:23 PM EST up reply actions

It’s interesting that they are each uniquely bad.

Truth.

Lightning strikes once, Hextall strikes twice!
"I think there is virtue in pissing off idiots." - Fehr and Balanced

by hintzy64 on Feb 8, 2012 2:04 PM EST up reply actions

Those who are concerned about wishing harm on other humans already think there’s no place for Girlyname on the team – that being his entire purpose. At least BSC can say he is an agent of entropy, which is more than Girlyname can ascribe to himself.

So warm up them chainsaws, y’all!

Maxime Talbot - in the Orange and Black ... better than chocolate and peanut butter!

by MaximumTalbot on Feb 8, 2012 2:25 PM EST up reply actions

Where’s Couturier? I’m probably just missing it.

I heart our rookies.

by LeepinLizardz on Feb 8, 2012 1:22 PM EST up reply actions

Mislabeled as 15. Oops.

@BSH_EricT
Writer at Broad Street Hockey

by Eric T. on Feb 8, 2012 1:50 PM EST up reply actions

That’s what I thought. Guess what? I figured it out by looking at Todd’s chart and finding where Scooter should be in your graph. I’m learning!

Also, rec’d the scatter plot, even though it’s not a bubble graph.

I heart our rookies.

by LeepinLizardz on Feb 8, 2012 2:21 PM EST up reply actions

Damn, now there’s an outlier.

Lightning strikes once, Hextall strikes twice!
"I think there is virtue in pissing off idiots." - Fehr and Balanced

by hintzy64 on Feb 8, 2012 2:02 PM EST up reply actions

Yeah, I remember Rinaldo being a giant red bubble on that rookie bubble graph the other day (that’s bad).
But at least he wasn’t too far away from the axis of the graph (that’s good).
But he was in the bottom left quadrant (that’s bad).
Also, he’s cursed.

I heart our rookies.

by LeepinLizardz on Feb 8, 2012 2:24 PM EST up reply actions