Capitals/Canadiens: Match‑up Overview
by David Getz on Apr 14, 2010 11:30 AM EDT
Email
PrintMatch-ups are at the core of any playoffs series, and the first round series between the Capitals and Canadiens is no exception. Here's how we see the two teams stacking up in seven key areas:
Offense - This one's pretty straightforward. The Capitals score more than anyone else in the League both in total and five-on-five, and have the League's best powerplay. The Canadiens score less than anyone else in the NHL five-on-five - the only thing keeping them from only getting one check mark is the fact that their powerplay, at 21.8%, is the League's second best.
Defense - Neither team is outstanding defensively, but both are solid, which makes sense when you look at each team's depth chart on defense - after the first couple of names, it's "solid-but-unspectacular" defined. The Capitals are better five-on-five, but the atrocious penalty kill hurts the team more than they'd like to admit.
Goaltending - The netminding situation in the Caps/Habs series is about as interesting as any you'll find in the first round of this year's playoffs. Each team has two guys who could be NHL starters, and both starters might be on short leashes. So who has the edge? Right now, we think it goes to Montreal. All four goalies serious question marks, but Jaroslav Halak clearly had the best regular season and is probably the player most likely to be a difference-maker.
Powerplay - The Capitals have the advantage here - anyone who finishes as far ahead of the pack as they did this season has a clear leg up - but the Canadiens have, literally, the most successful powerplay in the NHL after the Capitals, both in terms of conversion and in terms of scoring rate. All things being equal, you have to give this one to the Capitals, but whichever team gets to put their powerplay on the ice more often is the one who's going to have the advantage here.
Penalty Kill - The Canadiens have good, but not great, penalty killing numbers, and finished the season tied for 11th in penalty kill effectiveness. However, the reality is 'good-not-great' trumps the Capitals without a whole lot of discussion. Only one team (Nashville) that's going to be on the ice rather than the golf course next week had worse penalty killing numbers during the regular season than Washington. That gets you one check mark.
Coaching - Jacques Martin has had a pretty successful career as an NHL coach, as his 556 career wins, eight playoff appearances, and a trip to the Conference Finals in 2003 will attest. Still, he's made it that far just once and his marriage to his ultra-defensive system is arguably hurting the Canadiens. After all, a team with Scott Gomez, Mike Cammalleri, Andrei Markov, Brian Gionta, Tomas Plekanec, and Andrei Kostitsyn should do better than finishing last in the NHL in five-on-five scoring. As for the Caps, Bruce Boudreau obviously has a pretty nice track record but he hasn't proven his ability to take a team deep in the NHL playoffs, and his first-round coaching last year left more than a little to be desired.
Intangibles - It's hard not to like what the Capitals have: a nucleus that has been together for several years and has grown together, great chemistry on and off the ice, a coach that knows his players, and a significant home-ice advantage. You can't really give a team that hasn't even been to the Conference Finals five checks, but the Caps are about as close as you can get. As for the Canadiens? They don't seem to have a lot more, as whole, working for or against them than the average playoff team.
Overall, the advantage clearly goes to the Capitals, who are bigger, more skilled, and more well-rounded than the Habs. There are only two things that could make this a competitive series: a stellar performance from Halak or Carey Price and a huge powerplay disparity in Montreal's favor. Barring either of those occurrences, the Capitals should take this one and, hopefully, do it in five games or less.
162 comments
|
1 recs |
Do you like this story?
Comments
Regarding the PP, I really liked Peerless’s comment in the latest peice. Montreal’s PP is much better on the road at home, so maybe our respectable home PK unit can hold them in check.
Also understand that I am always an optimist (I actually believed that the Orioles could be good this year)
No, this is a very valid point. Overall, the Montreal PP is very good but on the road they’re the best in the league at 28.3%. Fortunately, the Caps PK is also pretty good at home killing off penalties at an 83% rate. The Caps road PK is terrible, but fortunately the Habs powerplay at home is very average at 16.9%. Must be all the Montreal fans yelling “shoot” in french
I agree, Montreal was 2 for 8 on the PP against us when they played at the Verizon Center this year. They were also 3 for 9 in Montreal. Most of the goals they scored against us were on the 5 on 5…
~~~ R0cK D@ R3D ~~~
by Chaz-Capapalooza on Apr 14, 2010 11:48 AM EDT up reply actions
I know they are still good, but in regards to the comment it looks like they were better on the PP in Montreal rather than in DC. Given that our PK has significantly improved at home since playing Montreal last, I’d say that those percentages are going to drop in the playoffs.
~~~ R0cK D@ R3D ~~~
by Chaz-Capapalooza on Apr 14, 2010 2:19 PM EDT up reply actions
I would imagine that record has alot to do with the pressure they feel in their building. That said, what concerns me is the what Colin Campbell stated regarding specific items that the NHL will be monitoring: pre-game confrontations, late-game “message sending”, but most relevant to the current Caps:
Finally, coaches and general managers were also told goaltender interference will be watched closely……
The NHL intends to watch Holmstrom and all "crease crashers" closely, and Colin Campbell suggests it’s best to stay out of the blue paint to avoid a penalty.
http://www.tsn.ca/columnists/darren_dreger/?id=317931
This doesn’t fare well for Knuble and Laich’s style of play, and could give the Habs an up in PP situations.
"It hit me on the pants. I had protection. It felt good. Why? I wanted to win."
I’ll believe it when I see it.
Is that the crinkling sound of the tinfoil hats with winged wheel logos being pulled out?
Kung-fu Rink Rabbit
Donate to the Rink Pledge Drive for SAVES FOR KIDS! Ain’t nothing [crappy] about giving $5 so a stranger’s premature baby can have the time on a respirator they need.~Gould Old Days
Wait til the folks in Buffalo get a hold of snippet.
"You ever use smelling salts, every time you type a bad blog?" Brooks Laich
by Carl Putnam on Apr 14, 2010 11:57 AM EDT up reply actions
In fairness to Dreger, he used an example for each of the three violations – Holmstrom was just the example he used for goalie interference.
"It hit me on the pants. I had protection. It felt good. Why? I wanted to win."
“fairness” doesn’t calculate into the extreme paranoia that is experienced collectively by Wings fans about, well, everything. Eleven Cups and they still have a persecution complex….
Kung-fu Rink Rabbit
Donate to the Rink Pledge Drive for SAVES FOR KIDS! Ain’t nothing [crappy] about giving $5 so a stranger’s premature baby can have the time on a respirator they need.~Gould Old Days
Even if it doesn’t hold well for Knuble and Laich, we still have the carebears. I’m not too concerned.
Whoop Whoop Caps
by Whiter Mage on Apr 14, 2010 12:21 PM EDT up reply actions
I wouldn’t read into that too much, or at all. Home/Road splits in general, especially when a Road stat is better than a Home Stat, is basically useless information. If you ran MTL’s PP 1,000,000 times on the road and at home, you’re going to see them score a bit more at home than on the road, closely in line with Home/Road scoring ratios.
by Bushwood Bushwhacker on Apr 14, 2010 12:15 PM EDT up reply actions
Uh, on what do you base that assertion? Looks pretty clear they are better on the pp on the road (for whatever reason).
There are many things you can look at in sports statistics that just don’t carry weight in the long term. This is one of them. Qualitatively, there’s just no good reason why they would have a better PP on the road than at home.
by Bushwood Bushwhacker on Apr 14, 2010 12:35 PM EDT up reply actions
In other words, you have a small sample size distortion in the outcomes (probably).
What we do know is that the Habs converted at around 35% on the PP vs. the Caps this year. If that continues, and the Caps give up 5+ PP opportunities/game, they’ll have trouble.
"You want to start being part of the Rink? Fine, but more’s expected of you than John/Jane Cap Fan. Carry the cause of informed discussion to the unwashed masses and don’t crap in the yards of other SBN sites if you decide to go over there. They’re passionate about their teams too, no need to troll elsewhere and/or be a sore winner." --BP
It’s certainly going to be interesting to watch the Habs PP. Let’s just hope we see lots of Brooks/Backstrom/Semin and speed in general on the PK, and not as much Poti.
by Bushwood Bushwhacker on Apr 14, 2010 12:41 PM EDT up reply actions
The Habs PP against the Caps is a small sample size problem. I still think it’s a bad match up for the Caps, but if you are looking at a small sample problem that’s it. I’m not buying that a full season is too small a sample size to draw the conclusion that MON is better on the road than at home. Just because “qualitatively there’s just no good reason” why the PP is different doesn’t make the numbers less valuable. For one, many teams tend to be more cute at home, especially on the PP. That may or may not be the case with MON but it’s at least one plausible explanation. Second, the MON crowd is notoriously tough on their players.
A pretty girl is like a violent crime.
We’re talking about 25 or 30 road PP goals. If you throw 50 pennies up in the air, how many are going to land heads? If 30 do, are we going to think that pennies land more often on heads? I’m not trying give a snarky lecture on probability, it’s just not a huge difference here, and definitely not enough to mean anything.
by Bushwood Bushwhacker on Apr 14, 2010 1:23 PM EDT up reply actions
But a PP isn’t a coin flip. It’s a product of players doing something specific. When you have 41 home games and 41 away games and your PP% is much better on the road, that’s not usually a fluke. That means there is something else going on. Otherwise why would you ever talk about home/road splits? When would it be meaningful? Do you think our PK% being better at home is just another statistical fluke?
A pretty girl is like a violent crime.
by Rob Parker on Apr 14, 2010 1:25 PM EDT up reply actions 2 recs
We may be at an impasse. My point is all of their power plays are essentially the same event, replayed 200 times. Along the way they happen to flip a heads away more often then they flipped heads at home.
The outside influences you cited make for a nice story line, but in reality probably aren’t real. If they are, they may account for a minuscule amount in the long term.
by Bushwood Bushwhacker on Apr 14, 2010 1:46 PM EDT up reply actions
My point is all of their power plays are essentially the same event
Please expand. I think I see what you’re trying to say, but my feeble brain isn’t making the jump.
Do you think our PK% being better at home is just another statistical fluke?
Please expand. I think I see what you’re trying to say, but my feeble brain isn’t making the jump.
I hope I’m not being patronized…
http://en.wikipedia.org/wiki/Confirmation_bias
FnB has created a story to fit a small sample of data. The story sounds believable, and could even have some small degree of truth to it, but neither the numbers nor the story is enough to confirm one another, or leads me to believe that MTL will have a better PP% on the road than at home, if they played 100000000 games right now. The fact is, there is statistical proof that teams play better at home than on the road, and I’m sure that that would hold up in this case if you were to somehow simulate it, whereas there isn’t any existing proof that I know of where a team, player, group of players, or whatever, over a long period of time have ever performed better on the road than at home.
by Bushwood Bushwhacker on Apr 14, 2010 1:58 PM EDT up reply actions
That’s a question for the math majors, but it’s obviously a gradient, as you would imagine.
by Bushwood Bushwhacker on Apr 14, 2010 2:10 PM EDT up reply actions
The data set isn’t that small, and I’m sure someone from a Habs blog could explain why, more or less, the difference happens. I doubt it’s a statistical fluke.
Cидни Kросби: Александр Oвечкин, он твой папа теперь
Capitals Coming: for Capitals fans who can bear reading something less intelligent than a story at Japers' Rink
by red army line on Apr 14, 2010 2:03 PM EDT up reply actions
So we’re wandering into opinions, because I think It is pretty small, and it probably is a fluke.
by Bushwood Bushwhacker on Apr 14, 2010 2:11 PM EDT up reply actions
I don’t understand how 82 games is considered small, since that’s what actually constitutes a full season.
"Don't mind WM...he's an all-around jerk."
How can a goalie go .930 and follow it next season with a .900? Or a forward go 15% then 10%? 82 games seems like a lot, but in most cases the variance associated with season long stats is bigger than you’d imagine.
by Bushwood Bushwhacker on Apr 14, 2010 2:17 PM EDT up reply actions
Easy, because defenses change, tape emerges on goaltenders, and shooting percentage is fluky anyway. 82 games is plenty long enough to get a sense of trends and really should be long enough to understand how well a team can play at all facets of the game.Also players change.
Hockey, in spite of all of the stats that exist, is really an irrational game that we try to inject logic into with stats, but we can never truly predict or announce anything as a result. Stats are always changing from the moment the puck is dropped.
We try to incorporate so many stats, because they’re intangibles, but they never directly outcome a game. You could have a 100% power play on the road and a 0% power play at home, but still have winning percentages that speak to the contrary.
But since we’re arguing stats, you have to realize that since the game is itself irrational, stats will often reflect that. Case in point, everything points to having better PP% at home, but we see the opposite in this case. Washington clearly looks like a better team than Montreal, but both teams finished 2-1-1 against each other in the regular season.
"Don't mind WM...he's an all-around jerk."
Bottom line is: we are dealing with 57 PP goals.
32 came up on the road, 25 at home (These #s may be wrong.)
This is the definition of a small sample size.
by Bushwood Bushwhacker on Apr 14, 2010 2:29 PM EDT up reply actions
We’re not looking at powerplay goals, we’re looking at powerplay effectiveness, and the Habs had 261 powerplay chances.
See below. Simply not enough numbers. We are at an impasse.
by Bushwood Bushwhacker on Apr 14, 2010 2:40 PM EDT up reply actions
Because things aren’t random. Teammates change, opponents change, coaches change, systems change, people get older, they get hurt, and they get tired. There’s causality involved beyond statistical flukes.
What’s hilarious here is that I was on the other side of basically this exact convo with Hawerchuck yesterday or the day before on BtN.
by Bushwood Bushwhacker on Apr 14, 2010 2:30 PM EDT up reply actions
So you acknowledge that playing at home can have a beneficial impact on a team’s performance. So there is some difference between home/away. Why is it so outrageous that in some situations playing at home could have a negative impact?
I’m not the first guy advancing this theory. I’m ripping off much smarter hockey minds than my own. Skill teams are notorious for trying to be too cute at home, and especially on the PP. I’m just saying that’s one possible explanation. Ultimately, if one season isn’t a large enough sample size then there is basically never going to be a meaningful sample size.
A pretty girl is like a violent crime.
We may be at an impasse. My point is all of their power plays are essentially the same event, replayed 200 times.
I think that premise is flawed. Even if you take out intangibles like crowd support, ice familiarity, and so on, being the home team gives you last change, which is a significant difference from a strategy standpoint.
Sure, I’ll buy that. If you look at the PP% at home vs road for the entire NHL over the past 10 years you’d probably see a small advantage at home. At that point, when you have real statistical evidence of something, then you can begin to qualitatively explain it why.
by Bushwood Bushwhacker on Apr 14, 2010 2:09 PM EDT up reply actions
Sure, but that is related to projecting season to season. We are projecting within the same season after an 82 game sample. When a player has a great year, you can conclude that they’ll probably keep playing well because they are in the zone or whatever. Maybe it won’t happen next year, but if a guy plays well all year then you can’t just say “that’s a meaningless sample.” I think the same holds for MON’s PP. Maybe it won’t be better on the road next year, but I think it’ll hold this year.
A pretty girl is like a violent crime.
So then are home and away records just statistical flukes? Each game is essentially the same event. You shouldn’t have a different record on the road from your home record, according to your logic. But clearly some teams are better at home and some are better on the road.
And I will reiterate my vehement disagreement that a PP is analogous to a coin flip.
A pretty girl is like a violent crime.
Stratomatic hockey gives a significant faceoff advantage to home teams, as well as last line change. I mean, it is a game, but it tries to be as statistically reflective as possible. On a PP, it’s going to be easier to score if you win the face off, then also with the lines you want.
"Don't mind WM...he's an all-around jerk."
The whole point is that we’re trying to claim their PP is better on the road. If the convo was about them being better at home, which we have loads of stats to back up, then there wouldn’t be an issue. Ya know? They are in conflict with historical evidence, which makes it more likely to be an outlier… and that logic can be counter intuitive… but I’m just going to leave it.
by Bushwood Bushwhacker on Apr 14, 2010 2:21 PM EDT up reply actions
Counter-intuitive doesn’t mean wrong. The fact that teams are usually better at home suggests, to me, that it’s even more clear that the Habs have something else going on that does in fact make them better on the road.
A pretty girl is like a violent crime.
That’s exactly the reasoning that I was alluding to… but the fact remains that we have a small sample size (~30occurrences in ~60), and no proof statistical or otherwise that this team is better on the road. Just a few more goals this season. It doesn’t mean you’re wrong for thinking it, it just doesn’t mean you can be sure with any degree of certainty.
by Bushwood Bushwhacker on Apr 14, 2010 2:27 PM EDT up reply actions
So what numbers aren’t susceptible to this critique? What stats are valuable? The ones that conform to what you already know (home teams do better)?
A pretty girl is like a violent crime.
The more numbers, the more they mean.
by Bushwood Bushwhacker on Apr 14, 2010 2:30 PM EDT up reply actions
That’s nice and pithy, but 261 isn’t, apparently, enough for you. Are road/home splits valuable? You can’t really look at teams beyond one season as DMG explained, so, again, what numbers mean something to you?
A pretty girl is like a violent crime.
When you flip a coin 130 times, then get 30 heads. Then flip it 130 times again, then get 25…or whatever with the numbers….does this feel meaningful to you? Do you feel there is meaning to the fact that the first time you got 5 more than in the second batch? Just qualitatively think about that. I can’t do the math right now, but could bang out a quick graph later to show what I’m trying to say here.
by Bushwood Bushwhacker on Apr 14, 2010 2:39 PM EDT up reply actions
Not heads/tails, 1in 4 dice or whatever.
by Bushwood Bushwhacker on Apr 14, 2010 2:39 PM EDT up reply actions
Power plays aren’t a die roll or a random chance. Power plays are completely dependent on situation. Who’s in the box, is it the team’s best penalty killer? How good is the PP you’re up against? Are you up on a 5 on 4, 5 on 3, 4 on 3? I mean, the only impasse we’re at is that you seem to believe PP’s are random chance, and I think most of the rest of us don’t, and most of us seem to think it’s enough data and you don’t.
"Don't mind WM...he's an all-around jerk."
by Whiter Mage on Apr 14, 2010 2:42 PM EDT up reply actions 3 recs
First of all, you’re talking about half as many occurrences. But more importantly, why is a PP like a coin flip? How is this point not getting through? Are you just the guy that walks around popping balloons? Advance something of substance of your own. <em>What single season stats do you trust as meaningful? You keep saying “we’re at an impasse” but you aren’t really trying to resolve it. You don’t want to help us understand what would resolve the impasse.
We’re at an impasse, I think maybe we should compromise. We’re not unreasonable, I mean no one’s gonna eat your eyes.
A pretty girl is like a violent crime.
by Rob Parker on Apr 14, 2010 2:45 PM EDT up reply actions 1 recs
I’m sorry if I’m not communicating well here. I’m not trying to be snarky or not address what you’re saying. It just comes back to not having enough sample size, and having to let go of any preconceived qualitative reasons for why this is happening.
Read “Likelihood and the Way we think”
I don’t mean it in any negative way, it’s just a way to think about a single season of statistics, or a small sample size in the scope of a long term situation.
by Bushwood Bushwhacker on Apr 14, 2010 2:51 PM EDT up reply actions
I also have a lot of work and have to bolt, I’ll put up a quick fan shot later…
by Bushwood Bushwhacker on Apr 14, 2010 2:52 PM EDT up reply actions
What do you want, man?
You criticize lack of insight, analysis, whatever. Then when we start trying to look at the interplay between the Habs’ great road PP and our better home PK (and vice versa) you say the numbers aren’t meaningful.
Please, before I go crazy, name one stat that has meaning for you so that I can have some concrete idea of what you think is reliable. Otherwise just tell me you don’t trust stats. Please. Just pick one.
A pretty girl is like a violent crime.
Hey man, I really didn’t mean to turn this into any sort of shit flinging. I’m not sure why it seems to happen so often when i post, but I’m actively trying to avoid it. In this situation it screams fluke to me, mostly because the sample size is just too small. It doesn’t seem small, but it is small. I will try to show just how small it is later.
by Bushwood Bushwhacker on Apr 14, 2010 2:59 PM EDT up reply actions
I think part of the problem is that you’ve been quick to explain why everything is flawed but haven’t really offered solutions or alternatives.
A pretty girl is like a violent crime.
Yea, I am trying to get stuff done and can’t whip up any sort of real evidence of why the sample size is so small outside these little examples. And since that’s not compelling enough for anyone we’re going in circles. My fault. I will post a much clearer example of what I’m trying to say around 7.
by Bushwood Bushwhacker on Apr 14, 2010 3:09 PM EDT up reply actions
I don’t think the problem is people don’t understand that in repeated small samples random events can come out with different results. The problem is understanding why you think PP% is a random event analogous to a coin toss.
Of all our iniquities ignorance may be the worst
by Killer_Carlson on Apr 14, 2010 3:12 PM EDT up reply actions
Because a PP has a true rate, say in this case 22%. On the road it may be 21.5% and at home 22.5% (or vice versa as some believe). That true rate drives the outcome during the season. It may end up being 25% in a lucky year, and 20% in an unlucky year. But over 10000 years it will converge on teh true rate. There is a very real luck factor in sports.
Ovie taking a wrister from the point is just a roll of the die when you look at it over his career. Everything has a probability.
by Bushwood Bushwhacker on Apr 14, 2010 3:15 PM EDT up reply actions
And why a full season’s worth isn’t enough of a sample size. By his standard the Caps having the top PP in the league is essentially meaningless. It’s completely random and we just got lucky.
A pretty girl is like a violent crime.
When you try to discuss issues based in mathematics without using math, rifts between views are created. I will throw up numbers about our PP too later.
by Bushwood Bushwhacker on Apr 14, 2010 3:20 PM EDT up reply actions
The implication being that when issues based in math are discussed using math, no rifts between views are created?
Only YOU can prevent idiots from commenting!
by Knee high to a duck on Apr 14, 2010 3:30 PM EDT up reply actions
Math is math. It’s right or wrong. The simple stuff, at least. So yea. We are getting bogged down by something that is easily explainable with some actual math.
by Bushwood Bushwhacker on Apr 14, 2010 3:31 PM EDT up reply actions
Fuck math. Play hockey. Trends only get you so fucking far. I don’t give a damn at this point about Montreal’s PP.
"Don't mind WM...he's an all-around jerk."
by Whiter Mage on Apr 14, 2010 3:32 PM EDT up reply actions 7 recs
So rec’d.
Of all our iniquities ignorance may be the worst
by Killer_Carlson on Apr 14, 2010 3:33 PM EDT up reply actions
At the end of it, one stat really matters: wins. Well, points for the regular season too. But wins. Everything else is to help observe trends, I suppose.
Cидни Kросби: Александр Oвечкин, он твой папа теперь
Capitals Coming: for Capitals fans who can bear reading something less intelligent than a story at Japers' Rink
by red army line on Apr 14, 2010 4:21 PM EDT up reply actions
The application of math to real-world, complex events is rarely so cut-and-dried as the arithmetic or the simple algebras we use to describe it.
The calculations themselves have an objective correctness to them – they either are or they aren’t.
To me, it appears you’re trying to say that MTL’s PP differential between road and home is the result of noise. You cite the small sample size, but I haven’t seen anything yet that lets you rule out a causal factor, again due to sample size.
As far as I’m concerned, the best we can do in this case is to say that we can’t determine what, if any, causal factors there are in this PP.
Only YOU can prevent idiots from commenting!
by Knee high to a duck on Apr 14, 2010 3:40 PM EDT up reply actions
Exactly – the two situations are markedly different, both in tangible and intangible ways. To say it’s a fluke is like saying “Most baseball players are better hitters at home than on the road. Adrian Gonzalez hit 28 home runs on the road last year and only 12 at home, but because there’s evidence hitters hit better on the road, this must be a fluke” when the reality is that Gonzalez plays in a very pitcher-friendly park.
That’s not really apples-to-apples, because a coin flip is a random event and we’re talking about specific events with one very specific dichotomous variable that can affect performance.
I think a better analogy would be to think of yourself as a runner. Say, as a whole, runners run sub six minute miles 19% of the time on sunny days and 17% of the time on rainy days, presumably because the former presents a better environment to run in.
If you’d gone running on on 148 sunny days and 113 rainy days and found that you were notching sub six minute miles 17% of the time on sunny days and 28% of the time on rainy days, would it be strange to assume that you run better on rainy days?
by David Getz on Apr 14, 2010 2:59 PM EDT up reply actions 3 recs
Are there any sites displaying total 5-on-4 time split home/away, as opposed to just the number of PP opportunities?
Yep…that’s another source of noise on the PP, which means that numbers take even longer have meaning.
by Bushwood Bushwhacker on Apr 14, 2010 3:22 PM EDT up reply actions
OK I’m moonlighting in stats so by all means correct me if you can, but this whole thread is getting annoying. IF IF IF you made the assumption that powerplay success is equivalent to a coin toss, so the outcomes are a binomial distribution, the 80% confidence interval for road PP success rate (which was 32 out of 113, 28.3%), would be 22.7% to 34.5%. At home (which was 25 for 148, 16.9%), it would be 13.0% to 21.5% success rate.
Therefore, IF IF IF you assume the PP is just a coin flip, b/c the confidence intervals don’t overlap, you are >80% confident that there is a STATISTICALLY SIGNIFICANT DIFFERENCE between the home & road power play success rates.
By all means somebody correct me with a statistically based argument.
by vzotsky on Apr 14, 2010 3:42 PM EDT up reply actions 1 recs
Just because some statistic is an outlier from the general norm doesn’t mean that it is meaningless or just a fluke. You should figure out why it’s an outlier, not just write it off as unimportant. Again, we are not dealing with random events – these kind of statistics are very dependent on a number of factors that vary across teams.
Pick any team statistic that has a solid historical trend, and then break it down by team over the course of the season and I’d be willing to bet you have some team in any given season buck that trend. It doesn’t mean that it is fluke. It is perfectly plausible that some team reacts to the myriad of dependent factors influencing performance differently than the norm.
Of all our iniquities ignorance may be the worst
by Killer_Carlson on Apr 14, 2010 3:08 PM EDT up reply actions 1 recs
I’ll put it another way. Let’s say we are trying to determine if PPs are better at home or on the road. Every season by a single team is an observation point in the analysis. The vast majority of the observation points show that teams are better at home than on the road. However, every once in a while you have an observation point like the 2009-10 Canadiens who are better on the road than are at home. In the grand scheme of things, you are right – the sample size is not enough to change the general finding that teams are better at home than on the road.
HOWEVER, that does not mean that the outlier observation is meaningless and should be thrown out. So many other variables influence PP success besides game site that is not in the analysis, and thus we are not getting the full explanation for PP success variance. This observation (i.e. this team this season), for whatever reason provides a different result from the general trend that is not a fluke – they repeatedly perform better on the road.
Let’s take another statistical example. Let’s say the world of evidence shows that foreign funding has very clear consequences on the economy of 3rd world nations. However, there is an outlier country where the effects are way off from the predicted norm. Does that mean that it is just a fluke and thus we should continue to treat that country just like every other country despite the evidence? Or should we look at the reasons why that country is an outlier (i.e. things like political institutions, education, economic diversity, etc.)? That country’s status as an outlier is not meaningless, but it is also not enough to reverse the general trend that foreign funding helps 3rd world country’s in predictable ways.
Of all our iniquities ignorance may be the worst
by Killer_Carlson on Apr 14, 2010 3:24 PM EDT up reply actions
It’s easy to make it sound right when you say it like that, but it’s ignoring the math of the situation. I’m sorry I can’t elaborate. Check back later for a better explanation in the fanposts.
by Bushwood Bushwhacker on Apr 14, 2010 3:29 PM EDT up reply actions
I don’t understand what the math of the situation has to do with it. K_C’s point, at least as I read it, is that outliers aren’t always the result of statistical anomalies; that sometimes they’re the result of something unique to that particular observation.
The point is because of the small sample size and historical data, the chance that the true value of their PP is actually what you observe is far less. There is a real chance that this Habs team is way better on the PP on the road, I’m not saying there isn’t. It’s just far less than it seems like when you look at just this years numbers.
by Bushwood Bushwhacker on Apr 14, 2010 3:37 PM EDT up reply actions
I don’t see how math will change my point that outliers aren’t inherently fluky.
Trust me, I understand the statistical issues involved with small samples, statistical significance and “true” rates of statistical events. But you seem to be arguing that there is a “true” PP% that is constant across all teams in all years. That is not the case. There may be “true” rates for each team for a given year, but more likely for any given team in any given year there is a “true” rate for home and a “true” rate on the road, and the task we are looking at is if those two “true” rates are significantly different. They may not be for most teams, but for this year’s Canadiens they are.
Of all our iniquities ignorance may be the worst
by Killer_Carlson on Apr 14, 2010 3:37 PM EDT up reply actions
It will make lots of sense with a graph. Trust me. Sorry I can’t make it make sense with words.
by Bushwood Bushwhacker on Apr 14, 2010 3:40 PM EDT up reply actions
I’ve been curious about the second round and Eastern Conference Finals since Sunday, but hang on! Can’t think that far ahead! Thanks very much for the overview.
If Halak puts up an unholy six-goal stinker in Game 1, do you think it’s possible Martin could yank him for Price with the same speed of Theo’s demotion last year?
Definitely possible.
A pretty girl is like a violent crime.
by Rob Parker on Apr 14, 2010 12:06 PM EDT up reply actions 1 recs
Agreed. Especially since Price played the Caps pretty well this season.
Japers' Rink: Hockey blogging from the most powerful city in the world
Halak looked pretty pedestrian down the stretch, so I have to think his leash is way shorter than Theo’s. At the first sign of trouble in the crease, Lemaire Martin will go to Price, imo.
"You want to start being part of the Rink? Fine, but more’s expected of you than John/Jane Cap Fan. Carry the cause of informed discussion to the unwashed masses and don’t crap in the yards of other SBN sites if you decide to go over there. They’re passionate about their teams too, no need to troll elsewhere and/or be a sore winner." --BP
If the Canadians have a ‘good’ but not ‘great’ PK how come they got 5 checks? Just wondering because the overall looks like it should tilt towards montreal especially if you throw out coaching and intangibles.
~~~ R0cK D@ R3D ~~~
by Chaz-Capapalooza on Apr 14, 2010 11:42 AM EDT reply actions
I think it’s in comparison to the Caps’ PK is why.
by DrinkingPartner on Apr 14, 2010 11:46 AM EDT up reply actions
One extra check for Hal Gill, courtesy of Alex Ovechkin.
Now helping to keep an eye on all things Gr8 at Alex Ovetjkin.
haha, look for that individual matchup to flare up at some point during the series.
~~~ R0cK D@ R3D ~~~
by Chaz-Capapalooza on Apr 14, 2010 12:14 PM EDT up reply actions
I remember the hit, but didn’t we decide that should actually have been a goal, as it wasn’t actually interference? It was so long ago.
by DrinkingPartner on Apr 14, 2010 12:14 PM EDT up reply actions
Yes, should have been a goal.
I don’t care what the checkmark count is, there’s zero way the Habs are as good as the Caps. The Habs are smaller, shallower in talent on every unit, and get outshot at ES. It will take a goalie conflagration and a rash of horribly undisciplined play to prevent the Caps from winning this one.
"You want to start being part of the Rink? Fine, but more’s expected of you than John/Jane Cap Fan. Carry the cause of informed discussion to the unwashed masses and don’t crap in the yards of other SBN sites if you decide to go over there. They’re passionate about their teams too, no need to troll elsewhere and/or be a sore winner." --BP
At least that discounted goal didn’t come back to haunt AO – he didn’t fall one goal shy of the Richard Trophy or one point shy of finishing ahead of Sidney Crosby in overall scoring, right?
Japers' Rink: Hockey blogging from the most powerful city in the world
by J.P. on Apr 14, 2010 12:55 PM EDT up reply actions 8 recs
Ugh. Hadn’t considered that…now I’m bitter and disillusioned.
"You want to start being part of the Rink? Fine, but more’s expected of you than John/Jane Cap Fan. Carry the cause of informed discussion to the unwashed masses and don’t crap in the yards of other SBN sites if you decide to go over there. They’re passionate about their teams too, no need to troll elsewhere and/or be a sore winner." --BP
Though you can say that about any of his shots that just went wide, or high, or pinged off the crossbarr, or caught the tiniest corner of Mike Knuble on its way into the net.
But still.
Now helping to keep an eye on all things Gr8 at Alex Ovetjkin.
It was an absolute legitimate goal. Gill had the puck, Ovechkin removed the problem and the puck went in.
However, it was an extremely embarrassing goal so my tinfoil says they couldn’t let it stand.
President and sole member of the Erskine lobby.
I hated it at the time, but by rulebook definition it had to not count, under the rule that pretty much states “If it didn’t go in using one of our defined ways, so it’s no goal”.
I still think it’s bs, but I can buy why it was discounted.
Whoop Whoop Caps
I can’t. The league explicitly came out and said the ref made the right call in the game 7 against PHI. If not for that I could take your position, but given an explicit league statement saying that goal should count, the MON no-goal is horseshit.
A pretty girl is like a violent crime.
Different circumstances. The Philly goal (That I agree was bullshit) was actually shot in.
Whoop Whoop Caps
Irrelevant. The Caps goal wasn’t called off because it was or wasn’t shot in. That was never even a factor.
A pretty girl is like a violent crime.
I think they disallowed it because of the rule where you can’t hit a guy into his own goalie.
My ability to post is only surpassed by my ability to pinch pennies.
Here’s the write-up from EOTP after the goal occurred.
And another on SBNation Hockey.
Pick your rule and apply it as you like to justify your point.
Watching more games on Slingbox than anyone else!
I would not assign each check mark or category equal weight. Obviously, one could assume that but I dont think it is a valid hypothesis.
For instance, you could weight Washington’s PP as being more effective not only in its proficiency, but also in that Washington drew 20% more PP opportunities because they are presumably better divers more highly skilled.
And I don’t think coaching means as much in the playoffs, although you don’t want to see your favorite team being Bykov’ed.
MTL PK
5 stars? Seems like 4 max.
Also, in relation to all of the other stars… Caps Offensive should have 6… or 20 stars.
by Bushwood Bushwhacker on Apr 14, 2010 12:17 PM EDT reply actions
Did you see the note under the graphic?
[Ed. note: We rated the Canadiens penalty kill unit as a ‘four’. The graphic is incorrect and will be changed as soon as we have the chance.]
Japers' Rink: Hockey blogging from the most powerful city in the world
Sorry, my bad! Delete if you want.
by Bushwood Bushwhacker on Apr 14, 2010 12:30 PM EDT up reply actions
Goaltending
I think we’re overselling the Habs goaltending a little and underselling our own. I mean, it may be a result of the Canadiens defense, but their goaltending looks average at best to me. Yeah, Halak is a solid goaltender, and Price is still I guess a good prospect, but they don’t strike me as solid as a guy who hasn’t lost in regulation in almost four months. I don’t even think Halak is better than Varlamov, and I don’t think Price is better than Neuvirth.
Whoop Whoop Caps
Halak’s at .924 with five shutouts in 43 starts. Obviously Montreal plays a defensive style, but those are some fantastic numbers. As for Theo, he has been very good but the having not lost has an awful lot to do with the team he’s on.
Yea, Halak could be the real deal. He played great in the Olympics. Let’s hope he doesn’t have a coming out party or anything…
by Bushwood Bushwhacker on Apr 14, 2010 12:32 PM EDT up reply actions
I think that’s an opinion thing, DMG. I place more into wins than save percentage when comparing two teams of contrasting styles.
Whoop Whoop Caps
by Whiter Mage on Apr 14, 2010 12:33 PM EDT up reply actions
Respectfully disagree with you here. W’s are a reflection of team superiority, not just goalie performance, much akin to W’s for pitchers in baseball. I like the peripheral stats much more as an indication of overall performance.
That said, Halak has some quesitons to answer, first and foremost among them whether his stretch run was an indication of an encounter with the “rookie wall” or just an aberration.
"You want to start being part of the Rink? Fine, but more’s expected of you than John/Jane Cap Fan. Carry the cause of informed discussion to the unwashed masses and don’t crap in the yards of other SBN sites if you decide to go over there. They’re passionate about their teams too, no need to troll elsewhere and/or be a sore winner." --BP
I’m not trying to take too much away from Halak. I think the kid’s good enough to be a starter in this league, albeit an average one. I’m just saying I don’t think Halak is better than Varlamov.
Whoop Whoop Caps
by Whiter Mage on Apr 14, 2010 12:38 PM EDT up reply actions
Pre-injury Varly, and I’m inclined to agree with you. After coming back off the long layoff, however, Varly has been underwhelming.
"You want to start being part of the Rink? Fine, but more’s expected of you than John/Jane Cap Fan. Carry the cause of informed discussion to the unwashed masses and don’t crap in the yards of other SBN sites if you decide to go over there. They’re passionate about their teams too, no need to troll elsewhere and/or be a sore winner." --BP
He’s been hit miss, but Halak has those games too. I’m buying the system as the reason he’s had success more than how great a goalie he is. That’s a coaching check over a goaltending check.
Whoop Whoop Caps
by Whiter Mage on Apr 14, 2010 12:43 PM EDT up reply actions
Halak’s been a lot less hit/miss than Varlamov: in March and April, Halak had 14 starts, and 10 were “quality starts”; Varlamov had eight and only three were quality starts, and they were against Boston, Carolina, and Atlanta.
Obviously there’s no magic the quality start stat, but Varly’s performances post-injury have been underwhelming.
Part of the reason Varlamov isn’t getting the start numbers Halak is has to be the injury, but the bigger reason is Theodore. The Save% for Theo isn’t that far from Halak’s (I mean, we’re talking 1.5% really) and while over an entire season you could make a case for that being a big deal, going into a 7 game playoff it’s maybe three goals difference between the two before factoring in shots. When you factor in the shots the two guys face, you’re looking at Halak facing way way more pucks, which means he’s going to let in more goals and for the series likely have a higher GAA.
Whoop Whoop Caps
by Whiter Mage on Apr 14, 2010 12:53 PM EDT up reply actions
I don’t disagree, but I don’t see how Theo’s numbers impact whether or not Halak’s been better then Varlamov this year, especially later in the season, which is what I was getting at, and while Theo’s numbers have been very good, he’s also had hot streaks before.
Obviously the Caps have a huge advantage in offense and puck possession, which will probably lead to a higher GAA, but the post is intended to comparing goaltending in a vacuum, not rate who’s going to have the higher or lower GAA at the end of the series.
I’m not buying the systems argument. Martin has a defensive system, but it’s still not good at preventing chances from the other team. Montreal gives up a ton of shots and doesn’t get a lot – they ain’t the New Jersey Devils. The reason why Montreal is in the playoffs with a negative goal differential is because when they do win it’s usually because their goalies (usually Halak but at times Price) have been able to come up with huge games to keep the team in the game to get that extra 1 goal or win it in OT/SO.
Of all our iniquities ignorance may be the worst
by Killer_Carlson on Apr 14, 2010 1:06 PM EDT up reply actions
That’s fair. While I agree the Canadiens do give up a ton of shots, having watched a few of their games this year (I’m a little better at keeping how I feel for them in the closet than Becca is) I tend to notice they give up “better” shots for their goalies than the Caps do. The Caps tend to give up jailbreaks, tough shots, while the Canadiens have done a decent job of keeping them to the outside. It doesn’t really matter how much we argue this stuff once the puck drops, however.
Whoop Whoop Caps
I don’t watch the Canadiens a ton, but when I have watched them I almost never come away feeling that they made things easy on their goalies. They do keep the puck to the outside pretty well, but they are terrible about keeping traffic away from the front of the net, so they are still tough saves for the goalies.
Of all our iniquities ignorance may be the worst
by Killer_Carlson on Apr 14, 2010 1:42 PM EDT up reply actions
Sv% and GAA are just as much team functions as Ws are, IMO. Looking at that third period sv% for Theo I think you can make the case that a lot of the wins were in large part because of his ability to make big saves late in games.
A pretty girl is like a violent crime.
GAA definitely is, Huet and Niemi both have pretty good GAAs because CHI is an excellent defensive team, but they aren’t very good goalies.
Only YOU can prevent idiots from commenting!
by Knee high to a duck on Apr 14, 2010 3:11 PM EDT up reply actions
Yeah, take a look at the Blackhawks’ ridiculous SA/G and S differential.
On the Forecheck/Twitter/CLS
"What do you think this is? Major League Baseball?"- Shea Weber
by Chris Burton on Apr 14, 2010 3:26 PM EDT up reply actions
To a lesser extent Sv% is team influenced as well. Better defensive teams give up fewer quality chances making it easier to get a higher Sv%. The percentage of shots that are quality chances probably evens out across teams a bit over the course of a season.
Of all our iniquities ignorance may be the worst
by Killer_Carlson on Apr 14, 2010 3:30 PM EDT up reply actions
Different players also have different SA% against when they’re on the ice, it’s just less entangled than GAA or W is.
Only YOU can prevent idiots from commenting!
by Knee high to a duck on Apr 14, 2010 3:42 PM EDT up reply actions
Also, if you prorate Varlamov’s starts this year to 43 instead of his 26, he’ll have ~3-4 shutouts, and maintain a .909 save percentage. You don’t think Varlamov would be about the same as Halak in Montreal’s system?
Whoop Whoop Caps
by Whiter Mage on Apr 14, 2010 12:36 PM EDT up reply actions
It wouldn’t shock me either way, but Varlamov’s not starting for the Caps and hasn’t been great post-Olympics.
My point is that Halak has comparable numbers to our back up, season wide, yet we’re giving them the edge in goaltending. That doesn’t make sense to me.
Whoop Whoop Caps
by Whiter Mage on Apr 14, 2010 12:40 PM EDT up reply actions
They don’t have comparable numbers, in my opinion. Halak’s save percentage is a lot better and Varlamov’s season-long numbers don’t mean as much to me as his post-injury ones, which aren’t all that good.
I think both teams are in pretty similar situations with #1 and #1A goalies and guys who have questions, but I also think Halak’s had the best overall regular season, so I gave the Habs the nod.
Yea, there aren’t many ways to analyze the numbers to get any other conclusion than this. Halak has been really good. I know WM is leaning on systematic differences and wins, etc, but I still think it leads you to this.
by Bushwood Bushwhacker on Apr 14, 2010 12:46 PM EDT up reply actions
I’m not buying what you’re selling in that first paragraph. It’s completely comparable, given the systems at play.
I’m in agreement with the second paragraph, but I don’t think Halak’s been better enough to give the Hans the nod.
Whoop Whoop Caps
by Whiter Mage on Apr 14, 2010 12:47 PM EDT up reply actions
Post-Olympics, Halak’s been at .926 and 2.07; Varly’s been .896 and 2.76. I can’t buy a .69 difference in GAA or a .30 difference in save percentage being the product of systems.
My argument on Varly vs Halak at this point have been more to compare them as prospects long term than for playoff performance. I’d be fine with giving them the nod if Varly was starting (not by more than a check), but Theo’s starting.
Post-Olympics, Theo’s at .928 and 2.33. Halak’s had 6 of those 14 games allowing 3 or more goals, while Theo has had 6 of his 12 games allowing three or more goals. I’d argue the Caps are FAR better equipped to handle their goalie giving up three or more goals, and both of them are just as likely to do it, it seems.
Long and short, I think our goaltending situations are even. We’re not worse than the Canadiens in goal.
Whoop Whoop Caps
Post-Olympics, Theo’s at .928 and 2.33. Halak’s had 6 of those 14 games allowing 3 or more goals, while Theo has had 6 of his 12 games allowing three or more goals.
True, but Theo’s post-Olympic statistics represent a much smaller portion of his relevant NHL career’s than Halak’s. Theo has a pretty long track record of being a guy who’s average/slightly above average as a whole and who runs hot and cold. In his numbers I see more of a goalie playing well than a good goalie. With Halak I’m not sure which it is (or which it is more, if that makes sense).
It makes sense, but I don’t think it makes sense to justify an additional check for the Habs in goal.
Whoop Whoop Caps
WM is just sick and cranky, ignore him.
Follow me on Twitter!! I even say some relevant stuff sometimes
I wish I could quit you, Tyler Sloan! - Co-founder of the Tyler Sloan Fan Club
Okay, don’t mind WM he’s sick and an all around jerk, ignore him!
Follow me on Twitter!! I even say some relevant stuff sometimes
I wish I could quit you, Tyler Sloan! - Co-founder of the Tyler Sloan Fan Club
Theo has been inconsistent – if we get bad Theo, of the .898 Save Pct and regualr soft goals, we’re in trouble. Varly has not looked great since the Olympics, as DMG notes.
This is not to say the Habs don’t have some questions in net, but they seem to be smaller questions than the Caps have. In other words, the Caps have known potential shortcomings; the Habs have speculative potential shortcomings. If that makes sense, and with apologies for dipping into Rumsfeldian jargon.
"You want to start being part of the Rink? Fine, but more’s expected of you than John/Jane Cap Fan. Carry the cause of informed discussion to the unwashed masses and don’t crap in the yards of other SBN sites if you decide to go over there. They’re passionate about their teams too, no need to troll elsewhere and/or be a sore winner." --BP
by fat_daddyo on Apr 14, 2010 12:45 PM EDT up reply actions 1 recs
I just don’t see Montreal being able to wrestle the puck from us very much. The way to beat the Caps is to forecheck them to death and force them to turn it over. Being that MTL’s forwards are painfully undersized, they’re going to have a really difficult time pinning the Caps in their own zone. I think there are going to be some lopsided shot totals in this series.
Key to the Series/How the Caps Win
I think this series will be won and lost based on the Caps offensive depth versus the Habs (lack of) defensive depth.
As the BtN article points out, the Habs have at most 2 defensive pairings that can be effective: Spacek – Hamrlik and Markov – Gorges. That leaves Gil, Bergeron and O’Byrne to cobble together enough minutes to get them through, or it leaves Markov with a partner that does not suit him and a Gil – Gorges pairing that has only been effective in carefully-managed minutes.
If Martin rides the prior top pairs, eventually they will succumb to fatigue; if he goes with the latter pairings, eventually one of them is going to get exposed.
In the end, I’m looking for some production from the third and fourth line against the bottom pair, and also gritty forechecking and cycle work. Hopefully that can produce some icing calls that will allow the big guns to get on the ice against tired and outmanned Habs personnel.
(Better as a fanpost?)
"You want to start being part of the Rink? Fine, but more’s expected of you than John/Jane Cap Fan. Carry the cause of informed discussion to the unwashed masses and don’t crap in the yards of other SBN sites if you decide to go over there. They’re passionate about their teams too, no need to troll elsewhere and/or be a sore winner." --BP
Wasn't sure where to ask this...
Have they actually been using that crappy goal song that won that contest?
Haven’t been able to tell on tv.
Go USA, Braves, BU Terriers, Irish, Colts, Caps, Hoyas, NU Cats, Wizards, DC United, Washington Freedom
BU Hockey: National Champions 1971, 1972, 1978, 1995, 2009
apparently after a year and a half of using sbnation i still fail..
what’s caps clips?
Go USA, Braves, BU Terriers, Irish, Colts, Caps, Hoyas, NU Cats, Wizards, DC United, Washington Freedom
BU Hockey: National Champions 1971, 1972, 1978, 1995, 2009
This reminds me of the pre-game comparison graphic HTS used to do at the beginning of each broadcast.
by Pivonka, Michael Ridley on Apr 14, 2010 6:37 PM EDT reply actions
Comments For This Post Are Closed
Secondary Sidebar
User Tools
Related FanPosts on Japers' Rink
Related FanShots on Japers' Rink
FanPosts
Community blog posts and discussion.
Recent FanPosts
FanShots
Quick hits of video, photos, quotes, chats, links and lists that you find around the web.
Recent FanShots
The Vault
In Association With
Latest Puck Daddy - NHL from Yahoo! Sports
- Who are the Non-Smythe candidates for the Devils, Kings?
- End is near for Rick Nash, Blue Jackets as trade discussions set to resume
- Why the Devils and the Kings are evidence the Capitals shouldn’t trade Alex Ovechkin
- Eulogy: Remembering the 2011-12 New York Rangers
- Stanley Cup Final: 5 keys for New Jersey Devils and Los Angeles Kings
Latest stories and headlines from NHL.com
- Driven DeBoer delivers Devils back to NHL's elite
- Driven DeBoer, fired by Panthers, delivers Devils back to the land of elite in just one year
- Bruins acquire son of Hall of Famer Ray Bourque
- Bruins add a Bourque: Chris Bourque, son of Hall of Famer Ray Bourque, acquired from Capitals
- Shinnimin named CHL's Player of the Year
