More Discussion Analysis of Team Defense and Goaltending
by MarioD on Oct 8, 2009 7:21 PM EDT
Email
PrintFrom Puck Prospectus
First, let's talk about how underated Marty Biron is. Gabe Desjardins has created a VORP statistic for goaltenders. VORP (value over replacement player) was set so that a score of 0 translates into the 61st rated goaltender in the league, aka a replacement player.
VORP is reported in regards to wins over the course of 82 games. So if Goaltender Steve has a VORP of 1, that means he should give his team one win more than a replacement level goaltender.
Here are the conclusions for the last two seasons combined:
Goaltender Wins/82
Thomas 8.0
Vokoun 7.6
Backstrom 7.2
Fleury 5.9
Luongo 5.3
Lundqvist 5.0
Huet 4.9
Biron 4.3
Miller 4.2
Bryzgalov 4.1
Brodeur 4.1
Giguere 4.1
Conklin 4.0
Ellis 3.8
Ward 3.3
Price 3.3
Nabokov 3.1
Mason 2.7
Roloson 2.7
Auld 2.3
Lehtonen 2.3
Khabibulin 1.8
Smith 1.7
Kiprusoff 1.2
Theodore 1.1
Turco 0.8
Gerber 0.5
Toskala -0.9
Budaj -0.9
Osgood -1.4
Hedberg -6.3
Marty Biron's performance the last two years was the 8th most valuable to his team amongst all netminders.
Then, Andrew Rothstein broke down the previously discussed VUKOTA projections focusing exclusively on defense.
How is it that the Red Wings have such success despite the horrible output from Osgood the last two years? Incredible team defense and elite defensemen. The Defensive GVT projections have Lidstrom and the top defensive defenseman and Rafalski as fifth in the league. Datsyuk and Zetterburg are first and third among forwards.
So, with the Flyers this offseason electing to follow the Detroit Model of building a team, here's Andrew's evaluation of the Flyers:
After acquiring the services of defenseman Chris Pronger this offseason, the popular belief is that the Philadelphia Flyers had established themselves as one of the elite defensive teams in the league. Those who have bought into that thinking might want to reconsider.
The Flyers' defensive depth is remarkably poor when calculated by Puck Prospectus' defensive GVT (DGVT) metric. After Pronger (4.5 DGVT), Kimmo Timonen (4.6 DGVT) and Braydon Coburn (4.0 DGVT), the Flyers have no one else with a defensive GVT over 4.0. Two of Philadelphia's defenders and 10 of Philadelphia's forwards have defensive GVTs below 2.0. To put that in context, 2.0 is the average mark for NHL forwards. The average for defenseman is 3.5.
...
Don't get us wrong: Pronger is a great addition. It's just that one player alone won't mask an entire team's deficiencies on defense.
Paul Holmgren's got a lot of work left to remake this team.
This item was written by a member of this community and is not necessarily endorsed by Broad Street Hockey.
45 comments
|
1 recs |
Do you like this story?
Comments
coburn and timonen had a horrendous game tonight so yeah, maybe our defensive woes are still an issue…
it feels like the last two games that pairing has struggled. They didn’t look very good against Washington either. I understand it’s against the caps and pens, but they haven’t played well at all. Coburn, in particular, has disappointed.
Calm down...
It’s only the 4th game of the season. The on-ice chemistry will develop in time. The team’s overall play will be sharper in time. If we’re still having this conversation in February then it’s time to worry, but not now.
As far as this “Biron is the 8th best goalie in the league” business, I don’t buy it. Sabermetrics are still new-ish to hockey and some of the kinks are still being worked out. While some measurements, like Corsi, seem to hold water, others do not (at least not yet).
Statistical analysis can be applied to hockey to some extent, but hockey does not lend itself to statistical analysis nearly as much as baseball does. That is because hockey is not as much of a game of probability as baseball is and it never will be.
How can something like GVT be applied as an absolute barometer to the Flyers? Even with their added veteran presence, the average age on the team is just over 26. How will GVT change if Carle has a breakout year, if Parent and Syvret become steady and reliable defenders, etc.? How will GVT change if Emery doesn’t make the same lousy decisions that Biron used to make with the puck which led to goals against? How will GVT change if Giroux and the other young players continue to progress?
If all of these things happen and the Flyers have a very good year, what will the VUKOTA folks say? “Sike!”?
I understand that there are many ways to look at the quality of a team, but in hockey it seems – at least to me – that the intangibles bear a little more weight than the pure statistics do.
Do you see what happens when you find a stranger in the Alps?
yea, i see things the same way. Sabermetrics are great for baseball but hockey is a much different sport. Over time I think there will be some breakthroughs on how to evaluate players, but it’s going to take some time. This is still fairly new, and hockey isn’t as much of a static game as baseball, so I think it will take some time before you can take these stats as gospel.
We played the Caps and Pens, Ovechkin, Malkin and Semin scored 6 out of the 12 goals Emery has allowed this year. All of those players might put in a least 40 by the end of the year. Our defense is solid enough, making bad turnovers has nothing to do with team defense we are able to contain teams in our zone, kill penalties and keep shots to the outside, and pick up the loose pucks in front and on top of that not give up too many odd man rushes coming out of the offensive zone. What we have done that needs to stop is our forwards are getting too far ahead of the play our defense is turning the puck over at our blueline. If we stop giving the puck away once we get it, through good team defense, we need to make smart decisions and not get to fancy on the breakout. With the long break following the Ducks game and some time looking at these last two games on film this will get corrected. We still almost game back to get that game tied up in 3rd it is not like we lost 3-0 at home the Phoenix.
People are misinterpreting
this isn’t a knee-jerk to a gut-wrencher of a game (we scored more goals than Pitt and yet lost). This is simply factual statements.
Now, I think VUKOTA is still in its youth and probably (like PECOTA before it) is not as good as predicting prospects as it is established players — for example, I think Carle, Coburn, and Parent will all be better than their projections indicate this year, BUT, the damn thing is right. We just aren’t that great defensively. Not atrocious, but not the Detroit Red Wings. Maybe if some of the offensive guys develop, then yeah, but the nature of statistics such as these are that they reflect reality — I realize it’s not an easy concept, and 99% of people will choose to disregard stats, but really they quite literally are a reflection of reality. To ignore them is as silly as ignoring scouting.
Rather than do that, we should read and educate ourselves, and thus when our players are playing well we can appreciate that even more, or at least be more educated about the reasons for success or failure.
I realize that advanced metrics are new to hockey, but that doesn’t make them wrong. The baseball community has essentially accepted the value of stats such as VORP et al, and the football community is coming around more quickly than expected. Why lag behind those two sports, oh ye fans of hockey? Might as well be intelligent fans, no?
To me, what this article says is that the Flyers DO need work. We can definitely expect internal improvement, but status quo from last year is not good enough. Do I think JVR is god-awful defensively, as Mario does? I suspect he’s deficient, but I don’t think it means he’ll have a negative VORP or even a bad VORP. The thing to realize is that we’re trying to maximize our VORP, and doing things like not re-signing a great goalie (I understand that it was going to be “impossible”) doesn’t help that cause. I think this team as is can go very far, but I simultaneously would love for another defender to come in that could allow for 3 exceptional defensive pairings. As far as offense goes, as I said, I expect internal improvement from Giroux, Richards, Carter, etc., and I think Pyorala will be a solid 2 way forward, but that may not be enough. Regardless, these metrics are inarguably an important part of the mysterious picture of how to bring the Stanley Cup back to Philly.
I think we’ll see some line juggling to get some players sparked up and improve the overall team defense. Betts being out hurts some, sure. But I don’t think it’s just that. It’s still early. The team chemistry is still forming and it looks like they’re working on an Identity still. Or at least incorporating the new guys into their identity.
Unbelievable
I guess dinosaurs of sports will always exist. Just be aware that like in baseball and football before, there is no stopping the statistical revolution.
Which is NOT to say
that watching the games is unnecessary. It’s more necessary than looking at the stats. But come on, guys, if 1 + 1 = 2, why ignore half the equation? Currently, all we’re doing is looking at 1 = 2, and that clearly does not compute. There is no reason to not read and understand statistics as an important way to understand the Flyers and hockey in general.
(As I dismount my soap box, I mention: it’s become accepted that statistics is a legitimate way to rationalize the world. It’s not pseudoscience or farcical numbers, it’s the way things are done. It’s not even the way of the future, it’s the way of the present. So let’s not get stuck in the past!)
Fine, but...
No one is really saying that statistics are a bad thing. They are not. The only issue with stats such as GVT is that they are relatively new to hockey. Baseball and hockey both have a number of variables, but hockey has more I think. As such, it is too early to take these stats as gospel truth.
WIth that said, I think hockey is headed in the right direction with this kind of analysis. I myself didn’t pay much attention to it until I joined Matchsticks and Gasoline because someone usually busted out the CORSI numbers after every game.
Should hockey analysis move in a more statistically-based direction? Sure. Is hockey ready to be analyzed by stats alone at this point? No. Can we discount such things like the chemistry which must develop on a team? Absolutely not. Guys like Lappy and Pronger are solid veterans, but they have never played on the Flyers before. Therefore, they do not know the nuances of Stevens’ system, and they do not yet know their teammates’ tendencies on the ice.
So we’re getting there, but we’re not quite there yet. In the interim, it is good for all of us to educate ourselves. Perhaps we may think of ways to make these analyses more accurate.
Do you see what happens when you find a stranger in the Alps?
I think the crux of the argument is sound
Marty Biron is a better goaltender than we all think and the Flyers’ defense probably is as deep as we’d like to think it is. The statistics are strong and I don’t discount them at all. The Flyers have some question marks and some holes and the stats make this very clear.
However, to say that Homer has a lot of work to do to rebuild this team seems a little premature. We need to remember that the Flyers are 3-1 and the two teams that have played them tough are two of the top teams in the east, both with exceptional firepower on the offensive end. We’ll have to beat one or both of them to win the East most likely, but we at least showed that we can hang with them and there’s no reason to suspect that the team won’t continue to jell better and improve.
"It was almost like if Harry didn't call it, it wasn't real." - Jayson Stark
I’m saying, for Homer to follow the Detroit Model, he’s got a lot of work to do. You can’t just bring in a mediocre goaltender and add a stud defenseman and expect to win the Cup.
By comparison, Detroit has two stud Dmen and two stud forwards while the Flyers, despite his changes, still have a bad team defense in front of a mediocre at best goaltender.
Can't expect to win the Cup
But they remain Cup contenders (at the moment). Emery needs to be better than mediocre from here on out, but this Flyers team, while they’ve show inconsistencies, is very capable at playing at a high level. I agree that they are not Detroit and have work to do to get there, but given that they’ve played three of the top teams in the East and beat 2 (one soundly, one sloppily), I think you can call them (VERY EARLY) Cup contenders. Not favorites by any stretch, but in the conversation.
"It was almost like if Harry didn't call it, it wasn't real." - Jayson Stark
Kimmo Pronger 2 Stud Defensemen
Carter RIchards 2 Stud Fowards
He is not trying to follow the DET model, just building with we got. Detroit is still a defense 1st team playing a neutral zone (left wing lock) trap and rely on the PP to score goals (unfortunately they got no PP in the finals or they would have swept the Pens) we are playing with aggressive forecheck, dump and chase crash the net hardnosed Flyers hockey, win games 4-3, 5-4, not 3—2, 2-1. Just because we sign a goalie on the cheap doesn’t mean we are trying to build a team in the same model as the RedWings. If there was an amazing proven goalie available who could fit in our cap structure we would have added one, same goes for the Redwings, but there wasn’t one so why over pay for a guy like Biron who is a middle pack goalie at best.
You can’t just bring in a mediocre goaltender and add a stud defenseman and expect to win the Cup.
Were we watching in 2008?
I see your point, of course – our team defense needs to be better. That’s on the coaching staff I think. Homer has added some good pieces to the puzzle, and Stevens now needs to get them to play right.
Do you see what happens when you find a stranger in the Alps?
Hold on.
So the Flyers…
have a bad team defense in front of a mediocre at best goaltender
While Martin Biron is…
8th out of 60 [ranked goaltenders].
I agree that the Flyers’ team defense needs to get better. But why is Emery suddenly “a mediocre at best goaltender,” hmm? If memory serves, you previously defended his signing as “a classic Moneyball move” by the Flyers. Why are we changing our tune four games into the season?
Question #2: Are all of the NHL teams “behind the times” when it comes to statistical analysis? Everyone except the Islanders must be, because none of them seem to understand that Biron is the 8th best goalie in the league.
Honestly, I’m not trying to attack you or dismiss your research here. But Biron is a classic example (for me) why these stats are not yet spot on. I went to two games last year which we lost PRECISELY because Biron made lousy decisions with the puck. He made poor decisions in other games, too, but he got away with it. Dan Carcillo’s first goal last year came when he was on Phoenix, and Biron gave the puck to him while he (Biron) was standing behind the net and Carcillo was in front of it. That’s just one example. (Last Christmas, “Santa Coates” gave Biron a Flyers jersey to remind him which team he played on so that he wouldn’t give the puck to the other team.)
Do you see what happens when you find a stranger in the Alps?
1) Emery has always been a mediocre goalie at best, it’s not sudden by any means. Calling it a good signing doesn’t mean he’s a good goalie; it was a good cost-effective move because it’s cheaper to get defensive-minded forwards (eg: Blair Betts, Manny Malhotra, et al) than it is to get good goaltenders.
My criticism is that Holmgren didn’t continue down that road. He should’ve straight out signed both Betts and Malhotra in September when the best they could get were camp invites.
Also, my post was made three games into the season.
2) I have no idea what level of analysis hockey teams do, but I feel pretty comfortable saying “yes” adamantly. Given the fact that baseball teams are still behind the times in statistical analysis and hockey is an even more inbred sport than baseball which started statistical analysis about thirty years later.
The idea that Biron is a bad goalie because he mishandled the puck a couple times is exactly the type of misconception I’m setting out to disprove by using evidence from two complete seasons to demonstrate his strong play.
1) Emery has always been a mediocre goalie at best, it’s not sudden by any means. Calling it a good signing doesn’t mean he’s a good goalie; it was a good cost-effective move because it’s cheaper to get defensive-minded forwards (eg: Blair Betts, Manny Malhotra, et al) than it is to get good goaltenders.
I understand, but by referencing Moneyball you are implying some degree of success. The A’s took a bunch of players with under-appreciated abilities and turned them into a division champion. I would think the Flyers got Emery because they thought he’d be a part of their success – not that they’d be successful in spite of him.
My criticism is that Holmgren didn’t continue down that road. He should’ve straight out signed both Betts and Malhotra in September when the best they could get were camp invites.
There would not have been room for both of them, and we’d have $700K less Cap space to play with now. He did what he had to do by signing Betts.
The idea that Biron is a bad goalie because he mishandled the puck a couple times is exactly the type of misconception I’m setting out to disprove by using evidence from two complete seasons to demonstrate his strong play.
Biron is not a bad goalie. He’s a good goalie. I never said otherwise. I only said that he’s not the 8th best in the league, which I stand by. And as much as I agree with you that hockey should become increasingly analyzed via statistics, I think that the intangibles – such as Biron’s decision-making – need to be taken into consideration.
I think that Biron is solid, and I wouldn’t mind having him if I had to. But I’d rather have someone else who is equally as effective and who does not make as many poor decisions.
Do you see what happens when you find a stranger in the Alps?
There would not have been room for both of them, and we’d have $700K less Cap space to play with now. He did what he had to do by signing Betts.
This is just factually incorrect. First off, they still have the cap space for Malhotra. Second off, these are your bottom two lines when the team is healthy:
Carcillo-Betts-Laperriere
Powe-Pyrolla-Asham
They could have Malhotra as the second center and rid themselves of Powe and be a significantly improved team.
And that’s pretending the Flyers would never suffer an injury at the forward position. Whoops!
I think that Biron is solid, and I wouldn’t mind having him if I had to. But I’d rather have someone else who is equally as effective and who does not make as many poor decisions.
That’s fine, let me know when any of these guys who have played roughly equal over the last two years become available:
Biron 4.3
Miller 4.2
Bryzgalov 4.1
Brodeur 4.1
Giguere 4.1
PS: Biron is the only guy on the above list with a cap hit under $4.25mil.
Maybe theyve been pretty bad the first 4 games
But I think that Coburn, Carle and Parent will all take steps in the right direction. They are all fairly young (especially Parent) and are still becoming better players. Plus, with our offense, we dont necessarily need a Detroit defense to win the puck. We just have to be decent, and I think the offense will take it from there.
they have been pretty bad in the last two games at home and were very solid on the road be it against lesser offensive teams, but it is not a matter of coaching or the system just decisions and turnovers, not just any turnovers but really blatent giveaways to superstar calibur players. Once they play smarter the goals against will go down.
Baseball sabermetrics v. hockey
There are no variables in baseball. Pitcher vs. batter. Pitcher throws the ball. Batter trys to hit the ball. In hockey there are way to many variables. How can you put a # on a player when he relies on 4 teammates and 5 opposing players, and a goalie. As for us. We played horrible, absolutley horrible, in the past two games to the two other best teams in the conference, and Still finished 9-9. Defensivly there is only one way to go from here- and that is up.
by orangeandblack20 on Oct 9, 2009 8:59 PM EDT reply actions
There are no variables in baseball. Pitcher vs. batter. Pitcher throws the ball. Batter trys to hit the ball. In hockey there are way to many variables. How can you put a # on a player when he relies on 4 teammates and 5 opposing players, and a goalie.
Are you serious?
How can you judge a pitcher by win-loss record when it relies entirely on offensive production in which he has no involvement?
And Park Factors aren’t a variable in baseball?
QualComp and QualTeam specifically eliminate those variables you think are insurmountable.
One might even argue that there’s such a thing as a “rink factor” in hockey.
Broad Street Hockey - SB Nation's Philadelphia Flyers Blog. Makin' it look mean since 1967.
by Travis Hughes on Oct 9, 2009 11:38 PM EDT up reply actions
I’m not one for statistical analysis personally (I just hate math, basically), but I read something on it somewhere. If I can find the link, I’ll post it.
Broad Street Hockey - SB Nation's Philadelphia Flyers Blog. Makin' it look mean since 1967.
by Travis Hughes on Oct 10, 2009 1:37 PM EDT up reply actions
That was definitely relevant "back in the day..."
Older placed like Boston Garden and The Aud were tough to play in because of their small rink sizes. If memory serves, the Spectrum was “regulation-sized” (i.e., 200′ × 85′).
Rinks can definitely “factor” into the equation when we talk about stuff like quality of ice (Nasty in New York) or how the puck bounces off the boards/dashers (Problematic in Pittsburgh). Since the rinks themselves are all the same size now, “rink factor” cannot be as much of a variable as it is in baseball, where the parks are allowed to be differently sized.
I think the Flyers play on good ice until November, when the Sixers begin in earnest. At that point, the ice turns to crap because the Sixers usually play over it when the Flyers are not there.
Do you see what happens when you find a stranger in the Alps?
One other thing to take into consideration is how stats are compiled in different rinks.
From Gabe Desjardins at Behind the Net…
But when we look at the data recorded in each NHL rink, we do find “rink effects” – the bias of those watching the game and recording the stats for us. For several seasons, Madison Square Garden has had the shortest shot distance in the NHL (see Alan Ryder’s “Product Recall for Shot Quality” for more details.) This is not by chance; there are systematic errors in recording shot location (both for and against) at MSG.
Broad Street Hockey - SB Nation's Philadelphia Flyers Blog. Makin' it look mean since 1967.
by Travis Hughes on Oct 10, 2009 3:40 PM EDT up reply actions
Yeah, I saw that story too. Almost posted it in response to when Mario cited the official NHL stat sheet on the goal that bounced over Parent’s stick, off his skate, and into the crease. Then decided the moment had passed.
Interestingly, that story focused on how bad MSG was, but I read another one focusing on other teams too. Maybe it was in the SBNHockey email though. But yeah, shot distance recordings are known to be atrociously bad.
Broad Street Hockey -
Makin' it look mean since 1967.
by Geoff Detweiler on Oct 10, 2009 6:43 PM EDT up reply actions
Here’s something interesting: The guys at Behind the Net just posted an article on shot quality and save percentages. See the story here.
What does it say? Biron was a -3.5, Nitty was a -4.0, Boucher was a -10.8, and Emery was a +4.8. Data collected since the lockout.
Broad Street Hockey -
Makin' it look mean since 1967.
by Geoff Detweiler on Oct 10, 2009 11:47 PM EDT reply actions
It’s too late at night for me to read that right now, but how about putting in the real big qualifier that those stats are EVEN STRENGTH ONLY.
So at even strength he’s worse than he’s expected to be. Playing on a team with as good of a PK as the Flyers had, isn’t it important to see how good Biron was at even strength? As a team, they had a +11 goal differential at 5-on-5, and a -2 at 4-on-4 (6 for, 8 against).
Broad Street Hockey -
Makin' it look mean since 1967.
by Geoff Detweiler on Oct 11, 2009 1:11 AM EDT up reply actions
All I said was to clarify that it’s not a complete evaluation, it’s merely looking at even strength play.
I’m still trying to wrap my head around the two different approaches of analysis taken by the two studies.
But I will say that I don’t put much stock in numbers for Emery because the sample size is so small. He’s played 140 post-lockout games. Even for Biron I don’t like post-lockout numbers because he was the backup in Buffalo for those first two years. He played like 196 games since the lockout. By comparison, Henrik Lundquist has played 273 games because he was a starter all four of those seasons.
“Sample size” was the wrong term. It’s because Emery’s numbers include only one year as a starter and Biron’s include only two years as a starter.
I think there are great fluctuations in performance when you play three nights a week versus when you play three nights a month. Its analogous to pinch-hitting in baseball; some guys have the mental/physical makeup for it and some don’t. By no means am I being critical by making that point, just saying I think backup goaltending numbers should be treated with an asterisk.
wasn't pinch hitting
one of the things shown to have very little correlation year-in-year-out? as in, it’s crapshoot if a player is a good pinch hitter one year vs the next? except for literally a handful of players, who, like with “clutch” hitting, seem somehow to be the exception to the rule?
I don’t know specifically, but logically that should be correct, if for no other reason than sample size…
According to a google search of some bizarre website (baseball-almanac.com ??) the record for PH PAs in a season is 83 by the Mets’ Lenny Harris. Which is the equivalent of starting like 20 games?
I think there are starters who can’t pinch hit, but I think anyone on the bench is capable of being hot/lucky in a season, just as they would be capable of hitting .300 in a given month as a starter.
Comments For This Post Are Closed
Secondary Sidebar
User Tools
FanPosts
Community blog posts and discussion.
Recommended FanPosts
Recent FanPosts
SBNation.com Recent Stories
In Association With
Latest Puck Daddy - NHL from Yahoo! Sports
- Snow building up: Isles sign Frans Nielsen to 4-year deal, avoid free agency
- Puck Previews: Capitals, Panthers in Southeast showdown; Winter Classic announcement day
- Tribute to Owen Nolan: 10 moments that best celebrate his hockey legacy (Video)
- Puck Headlines: Canucks split Sedins; Ovechkin’s $4M house; Coyotes booster club folds
- The building challenges for Seattle, Quebec City as NHL markets
Certain photos copyright © 2012 by Associated Press or Getty Images. Any commercial use or distribution without the express written consent of Associated Press and Getty Images is strictly prohibited.
