Young forwards and tough defensive minutes
by Eric T. on Oct 7, 2011 11:48 AM EDT
in Philadelphia Flyers Commentary
Email
Print
As I've written elsewhere, one of my major concerns for this off-season has been where the defensive minutes will come from.
The Flyers shipped off two forwards in Mike Richards and Jeff Carter who were consistently among the team leaders in tough minutes played. In return, they got one guy (Wayne Simmonds) who plays tough minutes and one (Jakub Voracek) who doesn't.
Of course, they also got two young prospects. If Brayden Schenn and Sean Couturier can shoulder some of the load Richards and Carter used to take on, it will make their loss (and the loss of defensive stalwarts Blair Betts and Darroll Powe) easier to bear.
How likely is that? Find out after the jump.
Let's start by talking about what we mean by tough minutes. To me, a tough shift is one where you start in your own end against upper-tier competition. Shifts where you start in the offensive zone against weak competition are easy shifts, and the others are in between.
Of course, it's hard to keep track of exactly where everyone starts and who they face over the course of a game, much less a season. So we look to the box scores to keep a tally; the NHL shift charts and play by play show us who was put out for the defensive zone draws and who faced the top competition. Going through those records lets us tally up two metrics that I use a lot to assess usage:
Zone start percentages give us a measure of whether a player gets used more in the offensive or defensive zone. The average player is near 50%.
There are a few different competition metrics; the one I use most is Corsi Rel QoC, which assesses a player's competition based on whether the line he faces tends to get off more shots for and prevent shots against better than the other lines on that team. The average player has a score near 0.4, with higher numbers indicating tougher competition.
So to get a sense for how likely it is that Couturier or Schenn will be ready for tough minutes, I'm going to look at how many 18-20 year old forwards had above-average defensive zone starts and faced above-average competition over the last four years. And I'll compare that result to a group of forwards of all ages.
The result is this chart, which has a dot for each player, with dots in the top-right of the chart indicating a player who took on tough shifts and dots in the bottom-left quadrant for players who were sheltered.
There's a lot to see there, but focus on how many red and blue dots (young players) are in the top right versus bottom left, and I think you'll see that it's pretty obvious that very few players take on tough minutes before they turn 20. In case it's not obvious to you that the younger players tend to be quite sheltered, here's a tabulation of how many people are in the top-right (defensive) quadrant versus bottom left (sheltered):
Just 13% of the 18 and 19 year olds take on tough assignments. It's not impossible, and the player Couturier was most often compared to in draft profiles -- Jordan Staal -- was one of them. But Jonathan Toews was also an elite two-way forward prospect and was a Selke nominee last year, yet in his rookie year he was heavily sheltered (44% defensive zone starts, Corsi Rel QoC of 0.02).
The first game looked promising for Couturier last night. He was given four defensive zone starts and one offensive zone start, and while he made one mistake that led to an open shot ringing off the post, he also broke up several plays and generally acquitted himself well in the defensive end.
So while there is hope that Couturier, and perhaps Schenn, will follow in Staal's footsteps, we should remember that it is rare that even a defensively talented forward is ready for that role at such a young age. I am nervous about constructing a roster that relies on an 18 year old being put in those spots, but it will be very exciting if Couturier can continue his promising start.
203 comments
|
Add comment
|
0 recs |
Do you like this story?
Comments
It’s a little off-topic, so I didn’t include it in the story, but I’ll throw it in as a comment:
I watched Couturier closely because I wanted to be able to include a first-game review in this story. He was pretty successful in the defensive end — aside from the one missed assignment that led to a near goal, he was generally in the right spots and got a clear almost every time the puck was within stick’s reach of him.
However, he wasn’t so great in the other two zones. He failed to receive passes a couple of times, he once had Hartnell open for a breakaway and whizzed the stretch pass waist-high past Hartnell, and he twice got beaten in the neutral zone to allow easy zone entries by Boston.
It’s all a one-game sample that I’m not going to ascribe any predictive value to, but last night he just looked more comfortable in his own end than anywhere else on the ice.
@BSH_EricT
Writer at Broad Street Hockey
Which IMO is a stark contrast from preseason where from the games I watched and too my surprise he looked better in the offensive zone with the puck and overmatched physically and slow to react in the defensive zone.
I wonder if size has a lot to do with the success of these young players defensively. Courtier like Staal is NHL size ready 6ft 4in close to 200lbs to when the entered the league. Toews who like you said was sheltered when he was a 18 year old might have entered the league at 180 – 185 and 6ft 2in and that could have lead to his struggles physically in his own end. And now that is body has matured and he is also 210lbs and 6ft 2in he has the physical strength to match his hockey sense and is hence a Selke candidate.
I know it’s a one game sample, but generally the D zone is the hardest for young players to learn, so that’s a bright spot. On the other hand, the O zone and general puck skills come most naturally to young guys, so if they don’t have it then it’s tougher to expect them to develop it. Maybe he was just nervous (his numbers in the Q indicate he has some puck skill), but I think it’s normal that guys with high end O talent generally show that pretty quickly, even if it doesn’t necessarily result in production.
Obviously, this is all speculation.
He showed some O zone talent in the preaseason, and being the top point scorer in the Q two seasons ago indicates he has that in him.
Simon Gagne AND Mike Richards may move between towns, wear new jerseys and call different arenas home but, at the end of the day, they will both always be Philadelphia Flyers.
One day Sean Couturier will win the Conn Smythe. You heard it here first.
by PursuitOfLappyness on Oct 8, 2011 12:11 AM EDT up reply actions
He made a few nifty plays in the O-zone, one subtle pass in particular gifted Read with an unobstructed look at the net stands out. And… although he is painfully slow of foot (Prongs vs Cooter in an icing race…epic turtle vs turtle?) he showed that he’s extremely heady in getting the play moving the other way on transitions, touch-passing, outlet passing, the like. Pretty impressed with him, and pretty surprising to see Lavi dump him in the deepest part of the pool right off the hop.
Yup, the pass to Read that set up the penalty that set up the goal was nice. But that’s the only one that stood out for me, and I saw several missed opportunities.
@BSH_EricT
Writer at Broad Street Hockey
Well, it was no Lemieuxian debut for sure, but at least the kid appears to be an NHL player at 18. Let’s see how long he can keep it up, or how long Lavi is willing to stick with him after his rookie mistakes pile up a bit.
Agreed.
Don’t get me wrong, it was a positive night for me overall. Mostly because I’m more concerned about defense than offense on this year’s team, and his defense was overall a positive. And because he scored a boatload of points in juniors, so I have some faith that the offensive talent is there. It just didn’t work last night, which brings us back to the one-game disclaimer.
@BSH_EricT
Writer at Broad Street Hockey
Now I know he’s no where near as built as the comparison I’m about to make, and I know he’s a pretty different type of player (not as physical at all, lacks the speed still, etc.)….but ever since seeing him the first time and seeing his Q stats, I’ve thought “he gives me a feeling like he could be a less physical Lindros.” I know, very weird, and doesn’t seem accurate AT ALL….just my weird feeling inside since first finding out about the kid. Just wanted to post that somewhere.
Yeah, a less physical Lindros, or maybe a less skilled Crosby, or Ovechkin without the scoring. I could see that.
/just givin’ ya crap, no offense intended
@BSH_EricT
Writer at Broad Street Hockey
and pretty surprising to see Lavi dump him in the deepest part of the pool right off the hop.
I think with Lavvy it’s very important for forwards to understand, accept and handle their defensive roles. So he’s saying to Cooter — “Hey, kid, this is what we expect from you. Go see how you do!”
Maybe harsh but at least it’s forthright!
by Georgia_Flyer on Oct 7, 2011 4:28 PM EDT up reply actions
His coaches in juniors said similar things. He likes to be in the defensive zone more than the offensive, to the extent that coaches have to force him to take up scoring opportunities. It’s not very usual for a young junior player, but it really sets him up to be an elite two way forward seeing as he’s already focussed on ensuring he’s taking care of his own zone.
Simon Gagne AND Mike Richards may move between towns, wear new jerseys and call different arenas home but, at the end of the day, they will both always be Philadelphia Flyers.
One day Sean Couturier will win the Conn Smythe. You heard it here first.
by PursuitOfLappyness on Oct 8, 2011 12:12 AM EDT up reply actions
This may be a stupid question but it’s something I’ve been wondering after watching Couturier last night. It was great to see Couturier have a successful game defensively but if either Schenn or Couturier were to be primarily deployed in defensive roles all season is there any possibility that their offensive development could be stunted?
Like I said, it may be a dumb question but it is something I worry about if the torch is eventually to be passed to these two to carry the team defensively and offensively.
Opposing Ilya Bryzgalov as Philadelphia Flyers' goalie since June 23, 2011.
@Mitchman88 on Twitter
I don’t think offensive talent gets developed at the NHL level. Once you learn to control the puck and think and compete with the increased speed and physically of the NHL game your hands are your hands and you either got them or you don’t.
At this level I think everybody’s got pretty good hands… it’s really the hockey sense and determination that separates the wheat from the chaff. Rob Schremp probably has better hands than anybody outside of Pavel Datsyuk, but he wasn’t able to adjust to the speed and aggressiveness needed to put up points at the NHL level… no amount of seasoning can really change that, and you can’t really lose hockey sense and instinct… you’re more or less born with it.
So history shows youngsters are protected by not being given hard assignments. Check.
Our youngster did pretty well with a limited number of these assignments last night. Check
How did staal perform at a young age? Specifically?
Commenter formerly known as M from Pdaddy, but still just Call Me "M"!
DISCLAIMER: Information written above may not be entirely factual nor provable with the use of complex statistics. But it may induce thought, humor and possibly laughter.
Staal did great. he was 3rd on his team in defensive zone starts and 2nd on his team in quality of competition faced, yet pushed the play forward better than the average player on his team (positive Corsi Rel).
There isn’t a second example like him from the last four years (the other guy who was given tough assignments was James Wright, who got creamed), but maybe Couturier will be it.
@BSH_EricT
Writer at Broad Street Hockey
Cooter sure didn’t wilt. So I liked that.
Commenter formerly known as M from Pdaddy, but still just Call Me "M"!
DISCLAIMER: Information written above may not be entirely factual nor provable with the use of complex statistics. But it may induce thought, humor and possibly laughter.
Pretty good analysis. I was pretty impressed with Sean last night, and as advertised, he looks very very good in the defensive zone. You can see that he knows how to cut off passing lanes where the center is looking for the puck in the slot.
If it means anything, Panotch made a comment to Lavy about how he must have some confidence in Sean leaving him out there for much of the last part of the game, and Lavy answered by saying,“what do you mean, he is our best defensive forward.”
Its quite clear that the Flyers feel Sean can take on tough defensive minutes and that is what made Betts expandable when Schenn comes back.
Want to know Jeff Carters legacy as a Philadelphia Flyer? Game 6 Stanley Cup Finals with about 2 minutes left. The rest is history.
Cout played one hell of a game I thought. He’s strong on his skates and deceptivly fast. I remember one shift where he was skating towards the corner from center ice and his long strides made him look like he was moving slow, but he easily out paced the (seemingly) faster Boston player.
He seems like one of those guys that you look at and say “he gets it”.
by ThatsGoodKoolAid on Oct 7, 2011 12:43 PM EDT reply actions
Anyone else ready to watch Mike Richards not hit anybody, not drop the gloves, and try to pass the puck through 5 players on the PP.
I sure am.
Want to know Jeff Carters legacy as a Philadelphia Flyer? Game 6 Stanley Cup Finals with about 2 minutes left. The rest is history.
Anyone else ready to watch Mike Richards not hit anybody, not drop the gloves, and try to pass the puck through 5 players on the PP.
Not sure which Mike Richards you mean. The one who used to be on the Flyers was 1st, 2nd, 2nd, 3rd, 4th, and 6th among Flyers forwards in hits in his six years here, was 2nd, 3rd, 5th, 5th, 5th, and 6th on the team in fighting majors in his six years here, and not only led the team in power play points each of the last three years, but was among the best in the league.
If you just don’t like the guy, that’s fine. But it would seem you’re making up reasons to support that dislike.
@BSH_EricT
Writer at Broad Street Hockey
by Eric T. on Oct 7, 2011 1:23 PM EDT up reply actions 3 recs
Yea I was a little confused on that comment myself. If anything That was not true of Richards. He might have meant Carter.
Commenter formerly known as M from Pdaddy, but still just Call Me "M"!
DISCLAIMER: Information written above may not be entirely factual nor provable with the use of complex statistics. But it may induce thought, humor and possibly laughter.
Yes, I assumed it had to be anti-management sarcasm.
by flyersfaninchicago on Oct 7, 2011 1:44 PM EDT up reply actions
Thank you for making my point Eric. What I was trying to insiuate was that Mike Richards USED to be an all around player who could do almost everything pretty damn efficiently.
I actually was a fan of Richards until we gave him a contract and his game changed over the last year. You have your opinion, which i respect and understand, and I have mine.
Bringing up stats from 2008 is not helping your case, but hey, to each their own.
If you dont feel that Richards underachieved last year, than we werent watching the same hockey game.
Want to know Jeff Carters legacy as a Philadelphia Flyer? Game 6 Stanley Cup Finals with about 2 minutes left. The rest is history.
You would assume that Eric and Geoff watch the games. They look at the stats afterwards.
Commenter formerly known as M from Pdaddy, but still just Call Me "M"!
DISCLAIMER: Information written above may not be entirely factual nor provable with the use of complex statistics. But it may induce thought, humor and possibly laughter.
What he gave you is a link the NHL.com, where you can find your own damn evidence by scrolling to a different year. If you did, or even if you read Eric’s comment correctly, you would have figured out that Richards has consistently been near the top of his team in all of these measures, regardless of when he got the contract.
Keeping alive the old Vaudeville joke, "I'd rather be dead than play Philadelphia."
Listen, I understand that you are all loyal fans and appreciate what he did while he was here, but if you take your heart out of it, Mike Richards was not the same player regardless of his stats. He is not a superstar, stop treating him like one.
He had 5 powerplay goals last year, that is not good enough. Nowhere near good enough.I guess I am on my own, but I saw Richie play some very very uninspiring hockey last year, coasting hockey, and that infuriates me and should infuriate you as well.
In his last 43 games, when we needed him most, he had 30 points, and 10(!!) goals. If that is good enough for you guys than that is fine, but to me, that just wasnt good enough.
Forget all that drinking and locker room bull. He got traded because his game dipped tremendously in the last year, he knows it. He even admitted while being interviewed saying, “I need to get my game back to where it USED to be.”
Want to know Jeff Carters legacy as a Philadelphia Flyer? Game 6 Stanley Cup Finals with about 2 minutes left. The rest is history.
Assuming you’re right about the last 43 games, that means that in the previous 38 games he had 36 points and 13 goals.
So your claim is that he was a great player for several years, was at a point per game pace over the first half of the season, then slipped to a 60-point, 20-goal pace over the second half of the season and that half-season downturn gets him traded away. Do I have that right?
@BSH_EricT
Writer at Broad Street Hockey
Unfortunately, yes, you have that right. Again, I am not a Richie hater, far from it. But when you are the best team in the conference for the first half, then go .500 for the second half, people start pointing fingers. And Richie is someone who you look towards to get you out of that slump, whether it is fair or not.
I dont care if he only had 30 points last season, if he was skating his butt off every shift than I am ok with it. But he was not.
Richie did not get traded because his stats were not there. He got traded because he played uninspired hockey for much of the season. You watched the same team. How can you not admit that a small part of you at least agrees with that statement?
Want to know Jeff Carters legacy as a Philadelphia Flyer? Game 6 Stanley Cup Finals with about 2 minutes left. The rest is history.
a) I don’t believe “he played uninspired hockey for much of the season”.
b) Even if he did, I care more about the quality of his performance than about his inspiration, and his performance was fine.
c) Even if his performance had dipped, I wouldn’t dump a star player every time they have a down half-season.
@BSH_EricT
Writer at Broad Street Hockey
If we dont agree that he played uninspired hockey for much of the year, than we will never agree anyways.
Im sure there are alot of people out there that agree with me, I know there has to be some.I am just not afraid to be one of the few to actually admit it, for fear of being one of the few who gets criticized for not agreeing with everyone else.
I respect your opinion, I really do. But I dont feel like he lived up to his potential last season, and am not afraid to be criticized for feeling that way.
Want to know Jeff Carters legacy as a Philadelphia Flyer? Game 6 Stanley Cup Finals with about 2 minutes left. The rest is history.
An entire conversation on a player’s inspiration level.
A player who dressed for 81 games despite having torn cartilage in his left hand the entire season, and then who, with 30 seconds left in his season, down by four goals, about to be swept, dove head-first to block an empty net goal while the rest of his teammates stopped skating.
That player.
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 Oct 7, 2011 5:33 PM EDT up reply actions 8 recs
wow
You have to be joking right? First of all, he sustained that injury in training camp. If it was really as serious as you are making it out to be, he would have had surgery and been done with it. He wouldnt have had a choice. Players play hurt every single game of the season, get over it.
And that play you are referring to, is the worst example of heart that you could have shown. How does that show heart? He didnt dive head first to block that shot, he obviously was trying to knock the puck of Paille’s stick. In what way was he trying to block the puck. Look at it again.
Again, Richie is a very solid player who will be very good with the Kings, but if you ask me would you do that trade all over again, I would without blinking an eye.
Want to know Jeff Carters legacy as a Philadelphia Flyer? Game 6 Stanley Cup Finals with about 2 minutes left. The rest is history.
I’m trying really hard to be nice this season, so as to not scare away the new people.
I’m trying really hard to be nice this season, so as to not scare away the new people.
I’m trying really hard to be nice this season, so as to not scare away the new people.
I’m trying really hard to be nice this season, so as to not scare away the new people.
I’m trying really hard to be nice this season, so as to not scare away the new people.
Okay. Woo-saaaa.
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 Oct 7, 2011 10:02 PM EDT up reply actions
Its fine I wont post anymore.
I know that I understand the game of hockey very thoroughly. One conversation with me in person would prove that.
I also have been reading the blog for about a year now, and as much as I like it , actually love it, I also notice how everyone likes to agree with the three editors so I will never win.
I may be new to posting on the site, but I most definately am not new to the game of hockey. Unfortunately, any opinion that is not your own is considered wrong.
Thank god Paul Holmgren is the GM and not you. Boucher might have been the starter in net last night
I
Want to know Jeff Carters legacy as a Philadelphia Flyer? Game 6 Stanley Cup Finals with about 2 minutes left. The rest is history.
Don’t leave the site and don’t stop posting. You are not the only one here who does not see eye to eye with people on here. Don’t get discouraged everyone is entitiled to their opinion.
"When the Flyers win a playoff series, as they did this past year, is when you yell at me for being wrong? Because I said the Flyers won’t win a round'?
Geoff Detweiler.
Agreed on all counts.
I don’t agree with much you said here today, but that doesn’t mean your opinion isn’t welcome.
@BSH_EricT
Writer at Broad Street Hockey
Seriously. With how often I argue with people, it’s amazing that so many people think we hate differing opinions.
If we all agreed, what would I do with my time?
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 Oct 8, 2011 1:02 AM EDT up reply actions
I had the same thoughts when I first came here. You learn that people just have strong opinions and their not being a dick or anything even though it may come off like that because things are typed and not spoken. Everyone here are decent people.
"When the Flyers win a playoff series, as they did this past year, is when you yell at me for being wrong? Because I said the Flyers won’t win a round'?
Geoff Detweiler.
Totally understandable. It’s not easy to come in and voice a dissenting opinion in a group of strong-willed people.
But the mix of views is important. If everyone sat around and agreed about everything, it would get dull.
@BSH_EricT
Writer at Broad Street Hockey
Try again:
June 7th, 2011, 9:40 PM:
My view is that Boucher should not be re-signed.
May 13th, 2011, 12:47 PM:
Seriously, it’s shocking how much crap I get for being blind to Boucher’s fault, yet I’ve never once said he should be re-signed.
May 13th, 2011, 12:18 AM:
I’m not even arguing for Boucher. I do not want Boucher re-signed.
May 12th, 2011, 10:08 PM:
As someone who does not want Leino re-signed, and does not want Boucher re-signed, I definitely do not want Bob in the AHL, nor do I want Carter traded.
May 12th, 2011, 9:56 PM:
It is why I have constantly said I do not want Boucher re-signed.
March 22nd, 2011, 11:36 PM:
I’m fine with him this year (he’s a flippin’ steal) but I do not want Leighton/Boucher next year.
March 15, 2011, 9:10 PM
But I don’t want to re-sign Boucher, then have Leighton and Backlund in the AHL, with Kovar and Riopel (and maybe Stewart and Morrison) not having any place to play.
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 Oct 8, 2011 12:35 AM EDT up reply actions
Geoff quick ? who would you start 2morrow Mezzy or Kaberle
"When the Flyers win a playoff series, as they did this past year, is when you yell at me for being wrong? Because I said the Flyers won’t win a round'?
Geoff Detweiler.
haha, seriously? Not that this is based on facts or analysis, but I don’t like Kaberle. So Mezzy.
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 Oct 8, 2011 12:50 AM EDT up reply actions
Yea thats what I was thinking Its fantasy but Kaberle is playing in Washington 2morrow so I was thinking I would take Mezzy. First time I ever had a fantasy team but I am doing good. Never thought I would like it but suprisingly I find pretty cool.
"When the Flyers win a playoff series, as they did this past year, is when you yell at me for being wrong? Because I said the Flyers won’t win a round'?
Geoff Detweiler.
It’s a lot harder in the beginning. But once guys start giving you enough data to go off of, it’s really just looking at numbers and seeing what your team needs.
Oh, and I get addicted to fantasy sports quickly.
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 Oct 8, 2011 12:54 AM EDT up reply actions
Yea I got a nice team. I had no idea about how to make a plan about drafting but I think my team is strong. A little weak on the wing but strong D and solid centers.
"When the Flyers win a playoff series, as they did this past year, is when you yell at me for being wrong? Because I said the Flyers won’t win a round'?
Geoff Detweiler.
Last year was my first fantasy hockey league in about 8 years. I drafted completely different this year than I did last year. Everyone has their own strategy, and I’m not one to put a lot of worth in goaltending (shocker, I know). Others are.
It’s what makes it great.
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 Oct 8, 2011 1:02 AM EDT up reply actions
Yea I went with one very good goaltender (Rinne) and a very good backup. I really wanted a good D that puts up a lot of points.
"When the Flyers win a playoff series, as they did this past year, is when you yell at me for being wrong? Because I said the Flyers won’t win a round'?
Geoff Detweiler.
I’ve got Rinne on my team, and I had him last year, too. He should treat you well. :-)
Lightning strikes once, Hextall strikes twice!
"I think there is virtue in pissing off idiots." - Fehr and Balanced
Geoff (on this post I am not looking for a bunch of stats just your gut feeling) when the game was in the third last night you didn’t feel a little bit different than in years past. You didn’t feel a little more confident with Bryz back there.
"When the Flyers win a playoff series, as they did this past year, is when you yell at me for being wrong? Because I said the Flyers won’t win a round'?
Geoff Detweiler.
I actually couldn’t watch last night because I was at work. Ask again when I can actually watch a game :)
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 Oct 8, 2011 1:20 AM EDT up reply actions
Work is good. Yea I commented last night about it with like 10 min to go or so. I was thinking to myself when I said it, “If I feel more confident what are the guys on the ice thinking”.
"When the Flyers win a playoff series, as they did this past year, is when you yell at me for being wrong? Because I said the Flyers won’t win a round'?
Geoff Detweiler.
You see that the two people who just agreed with me (Eric and Geoff) are two guys that I have had long and torturous disagreements with.
"When the Flyers win a playoff series, as they did this past year, is when you yell at me for being wrong? Because I said the Flyers won’t win a round'?
Geoff Detweiler.
I don’t think describing them as “torturous” will make him want to try again. :)
@BSH_EricT
Writer at Broad Street Hockey
But its a fun torture:)
"When the Flyers win a playoff series, as they did this past year, is when you yell at me for being wrong? Because I said the Flyers won’t win a round'?
Geoff Detweiler.
It’s not just about opinion all the time. You came in with that signature claiming Carter to suck because of one play, when it takes a team to win or lose. Then with one of your first comments, you stab at a great player—who tons of people with more hockey knowledge than you or me such as coaches and other players have repeatedly said they’d love to have on their team—with a flippant statement and hyperbole. And if you’ve been reading the site for a year, you would know making a comment like that will bring disagreement and outright throw down with some members, not just the editors.
As Eric pointed out, you didn’t even get that right since there are only 4 opposition players on the ice during a PP; or maybe passing it through the goalie was one of his 5 PP goals. You neglect the fact that the whole team struggled in the second half last year. You talk about how JvR had more hits, when he’s finding his game and is becoming a power forward so that should be a good thing.
No one minds a different opinion, but don’t come in and say something that seems clearly meant to gain a reaction. That’s what Philly.com is for.
G, the second coming of Foppa.
Embrace the Jagr.*
I know that I understand the game of hockey very thoroughly. One conversation with me in person would prove that.
Irrelevant. No one is questioning your hockey knowledge. And boasting about it isn’t going to impress anybody. I know hockey too, so what.
I’ll agree that his numbers were down at the end of the season last year, absolutely. (For that matter, so were everyone else’s on the team, but I’ll stick to just Richards.) I guess what I don’t see is why that slump warrants throwing away the multiple years of consistently being among the best on the team in multiple categories. And let’s drop the “agreeing with the editors” and “loyal fan” bullshit, that’s moving the conversation away from hockey and stats and into personal garbage. I had Richards on my fantasy team for a long time (someone drafted him ahead of me last year, but I got him back this year), simply because he put up good numbers year after year. If I based my team on who the guys are or who they play for, I could get lucky, but most likely I would suck. So I couldn’t care who Richards was or that he was a Flyer, I kept him for his consistently good performance. Because of that, I’m willing to forgive the last-season slump last year and give him another shot on my roster this year. And I understand you don’t see it that way, for you the half-season outweighs everything before it and that’s justification for the trade. So we can agree to disagree, that’s fine. I’m just putting my thoughts out there, same as you.
Sorry if I’m coming across as an asshole. You stated your opinion and someone had the audacity to question it and show stats that support their contrary opinion. Instead of responding with your own supporting evidence, which would be perfectly fine, even encouraged, you go for “but I know hockey” and “you all love the editors and Richards and will never listen to me anyway” and a completely irrelevant crack at Geoff being a fan of Boucher. That’s a bunch of shit and detracts from the real conversation we’re trying to have. If you’ve got hockey knowledge, bring it. Let’s have a conversation using that instead of personal crap. Then you’ve got a real shot of having your opinion heard and respected.
Lightning strikes once, Hextall strikes twice!
"I think there is virtue in pissing off idiots." - Fehr and Balanced
Not True
I disagree with most of what is said on this site. For the most part, they are stat nerds who refuse to acknowledge intangibles. They’re like the hipsters of the the sports fan world. (which is sad on many levels) Pretentious and generally unlikable. I’m fairly certain that I wouldn’t be able to sit in a room with Geoff for more than 5 minutes before I started daydreaming about watching him explode in slow motion to the soundtrack of O Fortuna.
Think about it.
Yeah. Nice, isn’t it?
Divergence is needed to see something from different perspectives. Without that, you don’t get the full picture. Learn to skim so you don’t get a full dose. Also, if you know a whole bunch of WIP listeners, encourage them to start posting here. Maybe then, I’ll get to see Geoff’s head explode.
stat nerds who refuse to acknowledge intangibles
I vote this when it’s time for the next BSH banner change. ;-)
by flyersfaninchicago on Oct 8, 2011 7:27 PM EDT up reply actions
And that play you are referring to, is the worst example of heart that you could have shown. How does that show heart? He didnt dive head first to block that shot, he obviously was trying to knock the puck of Paille’s stick. In what way was he trying to block the puck. Look at it again.
The official blocked shot label would depend on whether Richards blocked it to the corner before or after Paille’s stick hit the puck. But regardless of whether it counts as a blocked shot, he dove head-first to block a goal, which is what Geoff said.
If you have to ask how diving head-first onto the ice to prevent a goal in the waning seconds of the final game shows heart, then I’m not sure what definition of heart you use.
1. The chambered muscular organ in vertebrates that pumps blood received from the veins into the arteries, thereby maintaining the flow of blood through the entire circulatory system.
Pretty sure you need one of those to skate.
The repository of one’s deepest and sincerest feelings and beliefs
I’d say diving to stop a goal in the waning seconds of the final game shows a deep and sincere desire to compete.
The firmness of will or the callousness required to carry out an unpleasant task or responsibility
I don’t think diving face-first onto the ice is a pleasant task; I’d say he showed the firmness of will to carry it out.
Capacity for sympathy or generosity; compassion
I guess that’s the one you’re looking at? Admittedly, someone who was more generous might have allowed the Bruins to cap their celebration with one last goal. Personally, that’s not a quality I look for in my hockey players, but to each his own.
@BSH_EricT
Writer at Broad Street Hockey
Geoff is burning incense and Eric….not so much. Salmon box mania on the first night of the season.
Commenter formerly known as M from Pdaddy, but still just Call Me "M"!
DISCLAIMER: Information written above may not be entirely factual nor provable with the use of complex statistics. But it may induce thought, humor and possibly laughter.
Eric’s pulling out the dictionary, too.
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 Oct 7, 2011 10:25 PM EDT up reply actions
he obviously was trying to knock the puck of Paille’s stick. In what way was he trying to block the puck
Are you serious? Trying to knock the puck away from an opposing player who is about to score IS trying to block the puck. It’s the same goddamn thing, just using a different word! Don’t try to duck out on semantics. The point is he was still trying, unlike everyone else on the ice.
Lightning strikes once, Hextall strikes twice!
"I think there is virtue in pissing off idiots." - Fehr and Balanced
Giroux had nine goals over the last 44 games.
There’s going to be a sample size problem with this argument, however you restrict it, and there’s going to be a selection bias with this argument, however you explain the stats.
Keeping alive the old Vaudeville joke, "I'd rather be dead than play Philadelphia."
JVR had more hits than Richie last season, give me a friggin’ break.
WAKE UP PEOPLE!!
Want to know Jeff Carters legacy as a Philadelphia Flyer? Game 6 Stanley Cup Finals with about 2 minutes left. The rest is history.
Year by year total, since the lock-out: 63, 78, 110, 147, 145, 104. I don’t think you can say his game changed last year because he returned to 08-09 levels. I think it says more about line usage, variance, and subjective scorekeeping than it does about Richards being less physical. Even if he did hit less, I’m not convinced that’s a bad thing, and that wasn’t even your argument: saying that he wasn’t hitting anything is clearly wrong because he was stilling hitting a lot.
Keeping alive the old Vaudeville joke, "I'd rather be dead than play Philadelphia."
Subjective Scorekeeping?
LOL I guess maybe I have no clue what I am talking about.
But I have not lost one ounce of sleep since Richie got traded, and I dont think I will.
The fact of the matter is Richie got traded because of his contract and our terrible systems with no elite prospects. I am not arguing whether that is right or wrong, but it is what it is.
Want to know Jeff Carters legacy as a Philadelphia Flyer? Game 6 Stanley Cup Finals with about 2 minutes left. The rest is history.
I actually hated Richards on the PP but he hit and dropped the gloves. Don’t see where the hate’s coming from in those areas.
by mantis toboggan on Oct 7, 2011 4:01 PM EDT up reply actions
Anyone else ready to watch Mike Richards not hit anybody, not drop the gloves, and try to pass the puck through 5 players on the PP.
For those who didn’t get to see it, here’s Richards passing the puck through 2 players on the PP for the game-winner. Sorry, no video of the goal he scored to tie the game late in the third period.
@BSH_EricT
Writer at Broad Street Hockey
He doesn’t even look excited. He probably doesn’t care.
Keeping alive the old Vaudeville joke, "I'd rather be dead than play Philadelphia."
nice to see Richards still gets to beat the Rangers
by flyersfaninchicago on Oct 7, 2011 4:41 PM EDT up reply actions
Here’s the video of Richard’s tying goal.
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 Oct 7, 2011 5:35 PM EDT up reply actions
Wow. Never once did I say he couldnt pass the puck on the goal line to a guy 5 feet away for a tap in goal. You could probably make that play.
And I see one defensemen there, not 2. Dont take my comment out of context. If you watched any flyers games last year, you would know which play I am talking about. When He is up above the circle trying to pass it to the opposite goal line.
Want to know Jeff Carters legacy as a Philadelphia Flyer? Game 6 Stanley Cup Finals with about 2 minutes left. The rest is history.
As did many other players on the team all year. And a PP that never changed the whole year even though it struggled the whole time. That’s Richard fault too, correct? Not a coach or anyone else? Just want to know the width and breadth of what his contributions to the downfall of the entire team were. Oh, wait, Timmy P already told us all.
G, the second coming of Foppa.
Embrace the Jagr.*
To be fair, that would be a lot of players fault, and they got rid of a lot of players including Richards.
Personally, I don’t really like or dislike Richards & Carter. If they were still here, great, if not, I don’t care. They (and that whole team) were a little too Jekyll and Hyde for my liking. Somedays they would look unstoppable, other days they looked like they didn’t give a damn and were only there because their contract said they had to show up. They were a lot of fun to watch when they were on, and painful to watch when they weren’t. As a result, I found it hard to care about the team. I can honestly say that I have been more interested in the Flyers this summer & preseason than I was at any point during the last season including the playoffs. Even when they went to the finals I had very little faith that they would win it, and expected them to lose every series leading up to the finals.
Completely disagree. The players go out an play, but the coaches, starting with whomever is crafting the PP and ending with the head coach, should be in control of what is happening. I was literally shocked on Thursday night when two or three guys away from the puck were all moving to open spots. That just wasn’t happening last year consistently, and if your PP sucks, you need to change something up. A bunch of different people here were saying drop a guy (Hartnell or JvR) in front to screen. Put Z out there. Do something. The players have no control over that. Could have done more with what they had possibly? Sure, but I don’t believe coaching isn’t still in charge of what players are doing overall.
G, the second coming of Foppa.
Embrace the Jagr.*
And like I said, I think that team was so Jekyll and Hyde that, in my opinion, somedays they did what the coach wanted and other days they were skating around with their heads up their asses. I don’t think any coach could have done any better with that group. Some days they were unstoppable regardless of who the coach and/or system was. Other days there was no chance of them doing anything right. I put that blame on the players. If they could do what he told them for the first half of the year, why couldn’t they in the 2nd half? What is worse, was that it wasn’t just a 1st half vs. 2nd half issue. It was just more obvious in the 2nd half.
If you saw it differently, that’s fine. That’s why Baskin-Robbins have 31® flavors. We don’t have to agree. That’s what makes opinions so wonderful.
I agree with a lot of what you are saying. I found it hard to get into the past couple of years teams. If they were not playing well they were hard to watch. Many times last year they had no spark or hunger. I don’t mind loosing but at least play with some desire.
"When the Flyers win a playoff series, as they did this past year, is when you yell at me for being wrong? Because I said the Flyers won’t win a round'?
Geoff Detweiler.
Exactly! Nobody is going to win every night, but I still expect it to look like they wanted to. There was way too much talent on the team the last few years for them to be dominated the way they were some nights. They proved they were capable. They just didn’t show up some nights and that drove me crazy.
And I see one defensemen there, not 2.
Since you said “pass through five players on the PP”, I assumed you were including the goaltender. I, too, included the goaltender in counting his pass as going through two people.
Look, it’s very clear that you just don’t like the guy. That’s fine. You don’t have to. I wish your criticisms could focus on some of the things he actually doesn’t do well, but since nobody is going to be influenced by anything that’s said in this debate, I think I’m going to be done here.
@BSH_EricT
Writer at Broad Street Hockey
haha
well i have to admit, if I do post in the future, proof reading should be a priority lol
I can assure you one thing, I am aware that there are 4 players out there on a PK.
Want to know Jeff Carters legacy as a Philadelphia Flyer? Game 6 Stanley Cup Finals with about 2 minutes left. The rest is history.
Lavy said ‘he is our best defensive player’. Not was. Not will be. Is. That’s serious. I think the subjective analysis on his game was poor. Less the turnover, he looked impressive and deceivingly fast. His big body makes him appear to look slow, but during the neutral zone play in the third period where he played the puck off the boards and got off a wicked slapper. If he plays like this I’d be happy for him to stay for more than ten games.
by OrangeNblacK on Oct 7, 2011 1:14 PM EDT via iPhone app reply actions
Schenn
The only thing I am worried about is is Coturier sticks around, which it seems like he will, what does that mean for Schenn? Where does he fit in those top 9 forwards? Im not sure if he does, barring a trade of course.
Want to know Jeff Carters legacy as a Philadelphia Flyer? Game 6 Stanley Cup Finals with about 2 minutes left. The rest is history.
Stick one of them on the wing or one of them is the fourth line center. Those are your options, basically.
Keeping alive the old Vaudeville joke, "I'd rather be dead than play Philadelphia."
well even if you stick Schenn on the wing, that means one of your top 9 have to go to the fourth line.
And I would rather keep Schenn with the Phantoms or send Sean back to the Q, before putting either of them on the fourth line.
Want to know Jeff Carters legacy as a Philadelphia Flyer? Game 6 Stanley Cup Finals with about 2 minutes left. The rest is history.
It’ll be Nodl, but yeah, I’m working under that assumption. And yes, I agree neither should play on the fourth line.
Ultimately, I think we just have to wait to see when Schenn gets called up and if Couturier plays 11 games.
Keeping alive the old Vaudeville joke, "I'd rather be dead than play Philadelphia."
Agreed. But Nodl is already playing on the fourth line which means it would have to be one of the TOP 9 who would have to.
But I agree with you 100%, Schenns spot on the team is in Coturier’s hands.
Want to know Jeff Carters legacy as a Philadelphia Flyer? Game 6 Stanley Cup Finals with about 2 minutes left. The rest is history.
Der, that was stupid of me.
I could certainly see a scenario where Hartnell-Couturier-Read and Nodl-Schenn-Talbot are the bottom six. I just don’t want to see it.
Keeping alive the old Vaudeville joke, "I'd rather be dead than play Philadelphia."
Healthy scratch one so neither of them get worn out in their first season in the NHL?
by KornontheKobb on Oct 7, 2011 10:44 PM EDT up reply actions
Eric,
So the Corsi Rel QoC is what factors in line-matching by the coaches.
by flyersfaninchicago on Oct 7, 2011 1:44 PM EDT reply actions
2/3 of the coaches in this league would probably not know what you just said.
Commenter formerly known as M from Pdaddy, but still just Call Me "M"!
DISCLAIMER: Information written above may not be entirely factual nor provable with the use of complex statistics. But it may induce thought, humor and possibly laughter.
I dunno, M. I mean, it didn’t seem like Lavvy knew about it in the playoffs last year, but I’d assume more than 1/3 have heard of “line-matching”.
@BSH_EricT
Writer at Broad Street Hockey
No I was actually not being a smart ass. I would bet you if you polled all 30 coaches 20 would not know what you were talking about when you said Corsi or Fernwick or Vigota. Seriously.
Commenter formerly known as M from Pdaddy, but still just Call Me "M"!
DISCLAIMER: Information written above may not be entirely factual nor provable with the use of complex statistics. But it may induce thought, humor and possibly laughter.
And yet they are NHL head coaches…just who do they think they are basing their decisions on years of experience playing and coaching the game and not the brilliance of a blogger who has never spent one day being paid to GM/Coach/ or Play Ice Hockey. If I have to have surgery I’ll take an actual surgeon over some guy who has seen every episode of House and ER anyday of the week.
by Highaltitude784 on Oct 7, 2011 5:13 PM EDT up reply actions
This worked out great for baseball. Thinking that since they “played the game”, they knew more about people who study the game.
Eventually, people admit that being a joke doesn’t automatically make you the best at running a sports team.
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 Oct 7, 2011 5:37 PM EDT up reply actions
Since I am sure you are making a Moneyball/ sabermetrics reference remind me again just how well the 2002 A’s did in the playoffs…oh that’s right they got knocked out in the first round of the playoffs…
Since being the GM Paul Holmgren has built teams that went to the ECF and the SCF…Laviolette…Head Coach of a Stanley Cup Championship team…going to say they probably are a little more successful at building and running a hockey team then the local starbucks barista
by Highaltitude784 on Oct 7, 2011 6:01 PM EDT up reply actions
Because Moneyball died in 2002? Oh… that’s right, sabrmetrics is about a lot more than Moneyball. Moneyball focused primarily on OBP, which is hardly sabrmetrics.
But since you don’t understand Moneyball – or sabrmetrics – let’s just stick to the notion that people assume those who played the game understand how to win better automatically. Just because the Starbucks barista worked in a coffee shop for a couple years doesn’t mean I trust them to run a Starbucks better than a guy who ran a gas station.
And yet this is exactly what people do in sports: The guy who never played above minor league hockey and stopped going to high school at age 16 is automatically better qualified to write contracts, scout players, analyze players, and evaluate coaches.
I’d much rather hire the CEO of Starbucks as Assitant GM than hire the organizational grunt. I’d rather hire the guy who proves he knows something about hockey nobody else does, than hire the guy who knows how to play the best practical joke on the rookie teammate.
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 Oct 7, 2011 6:19 PM EDT up reply actions
On this topic Geoff we are in complete agreement. I have long said the biggest mistake in sports is that too many ex athletes are running them. The majority of these guys didn’t make in sports because Mensa was another option. The better gms in the league for the most part never played at a major level.
Commenter formerly known as M from Pdaddy, but still just Call Me "M"!
DISCLAIMER: Information written above may not be entirely factual nor provable with the use of complex statistics. But it may induce thought, humor and possibly laughter.
by MJDII on Oct 7, 2011 6:39 PM EDT via mobile up reply actions
Even though it’s rare, I do like when we agree. I just don’t like the black and white “they played the game, they know what they’re doing” claim. There are tons of dumb players and smart fans. There are so many guys who have incredible hockey intelligence but lack any athleticism, to the point where they never played past high school. There are so many guys who played hockey that don’t know what supply and demand means.
Simply put, being a former player qualifies you to maybe coach – but since even Wayne Gretzky can’t coach, what does it really qualify you for? Then expecting someone to know how to be a GM, simply because they played?
Is Sean Avery more qualified to be a GM than Travis? Than DragonGirl? Than Eric? No, he’s not.
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 Oct 7, 2011 6:49 PM EDT up reply actions
Sean Avery is more qualified as a GM to manage professional hockey player personalities simply because he’s more familiar with them. But he’s not more qualified to be a cap expert.
Keeping alive the old Vaudeville joke, "I'd rather be dead than play Philadelphia."
How much of a GM’s job description is managing player personalities? Beyond how much, how important is it? And do you need to play professionally to be qualified for that?
Can someone who played in college – think PSU – be qualified to manage professional hockey player personalities? What about the Staal brother who hasn’t played in the NHL yet? What about a Staal sister?
This probably seems argumentative, but I don’t understand where the line is.
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 Oct 7, 2011 6:56 PM EDT up reply actions
We all know the GM is heavily involved in managing player personalities, remember how the Phantoms problems were all because it was too far for Holmgren to drive to Glens Falls and handle the locker room issues?
@BSH_EricT
Writer at Broad Street Hockey
heh I do.
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 Oct 7, 2011 9:54 PM EDT up reply actions
There isn’t a line. I just wanted to point out that being a GM is more than understand the CBA. Some of that relies on insider knowledge/connections, because it’s still a good ole boys club.
Keeping alive the old Vaudeville joke, "I'd rather be dead than play Philadelphia."
Certainly. But that’s why I asked how much of the job requires him to know how to deal with professional hockey personalities. Can you know how to do that without having played? Look at Guy Boucher, who never played in North America (only one year in Europe), who managed professional hockey personalities to the Eastern Conference Finals.
Is Jim McCrossin capable of managing professional hockey personalities, thus capable of being a GM, because he was an NHL trainer?
Basically, I don’t see how “managing professional hockey personalities” is a skill either unique to professional players, or even evident in a majority of them.
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 Oct 7, 2011 9:58 PM EDT up reply actions
I don’t agree because ex players see it through there experience and skill level. They expect what they gave out of players that just aren’t capable but have a roll in the league. And of course I don’t mean every situation.
Commenter formerly known as M from Pdaddy, but still just Call Me "M"!
DISCLAIMER: Information written above may not be entirely factual nor provable with the use of complex statistics. But it may induce thought, humor and possibly laughter.
by MJDII on Oct 7, 2011 7:15 PM EDT via mobile up reply actions
Look at Wayne Gretzky. He proved dismal at managing a hockey team. Look at John MacClean, who failed in New Jersey.
Basically, being a professional player isn’t proof of competence, just like failing to be one isn’t proof of incompetence.
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 Oct 7, 2011 10:00 PM EDT up reply actions
Shero, holland and several others never even played college or juniors
Commenter formerly known as M from Pdaddy, but still just Call Me "M"!
DISCLAIMER: Information written above may not be entirely factual nor provable with the use of complex statistics. But it may induce thought, humor and possibly laughter.
by MJDII on Oct 7, 2011 7:05 PM EDT via mobile up reply actions
Scotty Bowman never played in the NHL or the minors…
"The characters in this picture are all fictitious. Anyone resembling them is better off dead"
Karate' Jerry..... Karate'
Semper Fi...
Bingo!
Commenter formerly known as M from Pdaddy, but still just Call Me "M"!
DISCLAIMER: Information written above may not be entirely factual nor provable with the use of complex statistics. But it may induce thought, humor and possibly laughter.
by MJDII on Oct 7, 2011 7:57 PM EDT via mobile up reply actions
:)
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 Oct 7, 2011 10:00 PM EDT up reply actions
This is one of the reasons that The Eagles manage there cap so well, is that Joe Banner is not a football guy, but is an expert in the CBA (I doubt you find another person in the world there know how to work the CBA)
by Anders Jensen on Oct 7, 2011 7:19 PM EDT up reply actions
While I am definitely a novice when it comes to statistics in hockey, I personally think the Moneyball reference is a generalization fallacy. That is, while sabremetrics have been quite effective in baseball, I believe a lot of this effectiveness is due to the nature of the game and therefore does not translate as nicely to other types of games such as hockey.
In particular, baseball is a game largely made up of independent events (a single pitcher pitches to a single batter, etc.). As a result, one can use individual level statistics to fairly accurately predict team level outcomes (such as wins). However, I believe that hockey is a game largely made up of interdependent events (i.e., much more of a "team" game than baseball). As a result, it is very difficult to predict criteria of interest (how well the team will do in a season) based on a collection of individual level player data (like they can do in baseball).
While I don’t know much about them, I am all for statistics in hockey and believe that they can be useful. However, I also believe that people go much too far in making inferences based on these statistics, when there is no data to back up these inferences (as there is in baseball) – I expand on this point in a following post.
The Problem with "Advanced Statistics"
Again, I believe that statistics in hockey is definitely a good thing and there is a lot to be learned from taking an analytical approach to the game. However, while I am certainly a novice and lack understanding of statistics in hockey, my crude interpretation is that a lot the "advanced statistics" in hockey haven’t been shown to have a meaningful team level outcome (if I am wrong on this point, I look forward to being corrected).
In particular, I think a major problem (from my lay perspective) is that most of the "advanced statistics" are really just descriptive statistics (summary and description of a player such as Corsi, etc.). However, what is needed in my opinion are inferential statistics, or statistics that make predictions and conclusions (with probability of error).
To me, this is a key distinction. If I was a GM, I am really worried about how many games my team will win this year. The "advanced statistics" crowd suggests that GMs should make their decisions based on advanced statistics of individual players. However, this is based on a descriptive statistics (Corsi) with no data to make a prediction (inferential statistics) about how this addition of an individual with a certain Corsi (over an individual with a different Corsi) will result in a number of wins for my team. I believe this is a huge problem that I haven’t seen addressed in hockey, but it has been addressed in baseball as they use very complex inferential statistics.
In a nutshell, I believe the "advanced statistics" in hockey are only "advanced" in one type of statistic: descriptive statistics. I personally feel that "advanced statistics" are very inadequate in the other type of statistic: inferential statistics. This is a main problem for the hockey stats community as ultimately GMs (and fans) care primarily about what can be done with inferential statistics (how a combination of individual players’ corsi predicts team wins).
I want to be very clear that I very much value "advanced statistics" in hockey. However, I feel that these approaches have major limitations in terms of inference, which is very problematic in my opinion.
I realize that I have provided somewhat negative opinions regarding "advanced statistics", which I understand may cause contention among individuals who put a lot of weight in them (especially on this board based on my lurking reading of it). As a result, I am willing to back up my opinions against the "advanced statistic" crowd with my challenge below.
A Challenge for "Advanced Statistics"
Per my post above, I believe that "advanced statistics" are woefully inadequate when it comes to predicting important team level outcomes in terms of wins. This is due to my belief that "advanced statistics" are based on descriptive rather than inferential statistics. As a result, advanced statistics are not very useful or valuable when predicting important outcomes such as wins (what really matters to GMs and fans).
I am willing to back up my belief with the following challenge. I will bet $500 that I will be able to better predict the total points for this season for each NHL team better than any "advanced statistics" model or prediction for points. Here would be the parameters of this bet:
1. The challenger can be any individual or combination of individuals. Thus, the challenger could be a collection of the brightest hockey statisticians in the world.
2. The challenger would need to predict team point totals for the 2011-2012 season based on any "advanced statistics" metrics that are used (corsi, fenwick, etc.).
3. The challenger would need to publish their statistical model that is used to predict team point totals. This model would need to be able to be replicated. That is, the prediction of the challenger must be based entirely on statistics (not subjective opinion) and this statistical approach must be transparent.
4. I would predict team points for the season based on my intuition and opinion. I would not use any advanced statistics; however, I could talk with other people to get their opinion and perspective (such as opinions on message boards). I could also look at predictions made by "experts" (such as newspaper columnists, bloggers, etc.). In the interest of full disclosure, I have been following hockey for a few years and only have an average to below average level of understanding of the game.
5. If the "advanced statistics" model is more accurate than my opinion in terms of predicting points for each team for the 2011-2012 season based on differences from prediction then I will pay $500 to the challenger. If I am more accurate based on my opinions and guesses than the statistical models, then the challenger will pay me $500.
I raise this challenge to highlight (or test with a bet) the limitations of "advanced statistics" in hockey. In fact, I am confident enough in my beliefs of the limitations of advanced statistics that I am willing to risk $500 of my own money to show that an average casual fan’s opinion is better at prediction of team outcomes than the best "advanced statistics". Note that I would never make this bet for baseball as I know that their statistical models would be much better at prediction than any fan’s opinion…given the independent nature of baseball (as compared to hockey per my earlier post).
I hope that someone will take me up on this bet as I would like to see how the "advanced statistics" crowd puts their descriptive statistics into practice in an inferential way in order to make predictions…and more importantly, see how accurate those predictions really are as compared to opinion and conjecture.
I believe that this challenge will show the main limitations of "advanced statistics" in hockey – Mainly that "advanced statistics" are not very good when it comes to predicting important outcomes such as team based performance. If one disagrees with this assertion, I hope that they will take me up on this challenge so that we can test it out and see who is right.
Challenge Hockey Prospectus. Just got the book yesterday, and they have predicted points for every NHL team.
And since their’s is already published, get yours in ASAP.
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 Oct 9, 2011 12:20 AM EDT up reply actions
Hockey Prospectus Reinforces My Original Point
Very interesting that you bring up the idea of challenging the predictions of Hockey Prospectus (HP). Indeed, I think that the HP is a great resource (already have a copy of it) and the authors of HP are some of the best at using inferential statistics in terms of predicting performance and future outcomes (which was the main focus of my original posts).
However, while I think HP does a great job at using inferential statistics (although they don’t publish their specific models as far as I know, so it is hard to evaluate just how complex and appropriate they are), I do not think that they use "advanced statistics", at least based on my understanding of their projections and the range of advanced statistics. That is, I believe HP’s predictions are based on traditional basic statistics, albeit they do a very good job at using them in an inferential way. Below I explain why in my estimation HP uses traditional statistics and does not utilize "advanced statistics".
The point totals HP predicts for each NHL team is based on their VUKOTA calculation which is based on their GVT calculation for players. Simply put, I would contend that VUKOTA and GVT are based on traditional hockey statistics and do not use any "advanced statistics". Indeed, HP affirms this point and states that "GVT is calculated based on statistics the NHL has kept for over 40 years" (p. viii), which to me suggests that GVT is not based on any of the "advanced statistics" that are typically discussed in recent years.
In particular, GVT is based on several traditional (not "advanced") statistics: (1) goals, assists, and ice time for offensive GVT, (2) number of shots allowed by a team (not an individual player), plus-minus, and penalty killing ice time for defensive GVT, (3) essentially save percentage for goalie GVT, and (4) essentially shootout percentage for shootout GVT. All of these statistics are basically normalized based on the average replacement level player to come up with GVT. Thus, I would assert that GVT is based on very traditional (and not very "advanced") statistics: goals, assists, ice time, plus-minus, save percentage, etc.
HP takes their GVT projections and creates the VUKOTA statistic which uses GVT and then adjusts it based on comparable players from history (e.g., taking into account age of a player). Note that VUKOTA is based entirely on GVT and therefore also does not use "advanced statistics", although they do a great job with the inferential statistics in terms of using historical metrics of prior players (in terms of their traditional stats) to predict future performance of current players.
HP then takes these VUKOTA ratings for GVT for each individual player and sums them for a team VUKOTA GVT, which then is arithmetically transformed to calculate total points for the team (every three goals is worth one point in the standings).
Hopefully this description clearly shows that HP’s projected point totals are based entirely on traditional statistics that have been around for a long time (40 years): goals, assists, plus minus, save percentage, etc. HP does not include any "advanced statistics" in their calculations of GVT, VUKOTA, or projected points – That is, HP does NOT incorporate or use any "advanced statistics" such as Corsi, Corsi QoC, Corsi Rel QoC, Relative Corsi, Fenwick, QualComp, sheltered minutes, zone start percentage, etc.
To sum up a very long winded post (for anyone who is still reading…or less likely cares), I believe Geoff’s reference to my challenging HP essentially further reinforces my original point. HP does not use any "advanced statistics" in their predictions. Instead, HP uses very traditional statistics in their prediction. What HP does very well is use inferential statistics in terms of calculating complex models (by using comparison to former players, etc.), but all of these models are fundamentally based on very traditional basic descriptive statistics (goals, assists, plus-minus, save percentage, etc.).
What I would like to see, as evidenced by my challenge above, is someone use "advanced statistics" in an inferentially way to predict meaningful team level outcomes. I don’t think HP meets this criterion as they very clearly use basic statistics that have been around for 40 years in their inferential predictions.
In short (for a very long post), I am a believer in "evidence based" prediction and I personally haven’t seen much convincing evidence (at least by my standards of inferential statistics) that "advanced statistics" predict team outcomes such as wins over the season. In fact, I would very much like to see some inferential statistics that are based on "advanced statistics", which is why I put forth my $500 challenge above in the hopes that someone will take me up on it. I find "advanced statistics" very appealing and useful, so I would very much like to see evidence of their efficacy in predicting team outcomes. Until that time, I personally will continue to use temperament and restraint in advocating the use of them until I see some sort of predictive study that clearly illustrates their utility in an inferential way.
Just my two cents (which I am sure that many people will think is worth only about a half a cent or less).
You’re saying GVT doesn’t count as a new statistic because it’s based only on traditional stats, including things like plus/minus, shots on goal and time on ice.
First of all, that’s a misleading statement. Yes, SOG is a traditional stat, but to calculate GVT, you need to know how many SOG for and against were recorded while a given player was on the ice, and that’s something we didn’t have until recently.
But that’s really beside the point — I think you’re setting up a false dichotomy. What makes Fenwick a non-traditional stat isn’t that it includes missed shots (which weren’t tabulated until recently), but that it’s…not a traditional stat.
To compare to baseball, FIP and OPS are calculated from numbers that were in traditional box scores; do you think the average fan considers those to be traditional stats? The essence of advanced analytics is not finding new things to track (though that often helps); it is understanding how much value to put on each of the things we have access to.
Basing your categorization of GVT on whether it uses numbers that were available 20 years ago not only misses the point that it doesn’t, but sets up a false dichotomy. The right line to draw is not whether the information was available to people 10 years ago; it is whether people thought that information was important ten years ago.
@BSH_EricT
Writer at Broad Street Hockey
by Eric T. on Oct 9, 2011 6:32 PM EDT up reply actions 2 recs
To clarify, as I noted in my original post, GVT does not include individual level SOG, it includes “shots allowed by a player’s team” (p. viii). Thus, all the statistics included in GVT have been available for quite some (as noted by HP in their discussion of the measure).
Essentially, my understanding is that GVT is just a multiple regression that predicts a single criterion of overall performance based on a number of common factors such as goals, assists, and plus-minus. I guess we can disagree, because I don’t consider this “advanced” in any sense. That is, people have known for a long time that goals, assists, and plus-minus were important and have evaluated players as such. While a casual fan may not have created an omnibus metric like GVT, they have always known that these factors added up to better performance.
Thus, a key point is that this GVT measure fundamentially does not exploit any inefficiencies in the market – There is no “moneyball” here as everyone already knows that goals, assists, and plus-minus are valuable metrics and statistics. I would contend that people have always evaluated players using the same metrics that GVT is based upon, so just creating a omnibus measure is definitely not “advanced” in my opinion.
If you consider this perspective to be a false dichotomy then we can agree to disagree. Thanks for your perspecitves – I appreciate reading these different points of view.
Ugh, you’re right — it’s not SOG allowed while the player is on the ice; it’s SOG allowed by the team regardless of whether the player is on the ice. Yet another reason I hate GVT, I guess.
Everyone always knew that getting hits and walks and extra base hits was valuable in baseball too, but most people consider OPS to be an advanced statistic because it combines them in a non-traditional formula. Everyone always knew that getting strikeouts and not giving up walks or home runs was valuable, but most people consider FIP to be an advanced statistic for the same reason.
If you disagree with those perspectives, that’s fine, but then you’re really just disagreeing with the common definition of advanced stats, not showing anything about the merits of advanced stats.
@BSH_EricT
Writer at Broad Street Hockey
Good points about OPS and FIP.
I think that this may be a matter of semantics and definitions (i.e., what falls under the rubric of advanced statistics). Ultimately, I guess it doesn’t really matter whether one considers GVT to be an "advanced statistic", as this is just a subjective label anyway (for the record, I do not consider GVT to be an advanced statistic). What matters is whether GVT is related to a meaningful outcome.
If one believes that there is evidence that GVT is related to a meaningful outcome, then this individual is really just advocating for the use a weighted combination of goals, assists, TOI, and plus-minus as a measure of a player and prediction of performance. While this is definitely useful and valuable, I personally want to see evidence beyond these metrics (either in isolation or in some weighted combination as in GVT) in terms of the value of other "advanced statistics" such as Corsi, Fenwick, etc. (which GVT does not consider).
Thanks again for your thoughts and perspectives – I find them valuable.
Thus, a key point is that this GVT measure fundamentially does not exploit any inefficiencies in the market – There is no "moneyball" here as everyone already knows that goals, assists, and plus-minus are valuable metrics and statistics.
Moneyball focused almost primarily on on-base percentage, a statistic that was tracked for over 100 years in baseball. It’s a statistic that everyone knew was important – getting on base.
Just so we’re clear.
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 Oct 9, 2011 8:36 PM EDT up reply actions
Unfortunately, I do not think that we are clear…and I think your post muddies the waters even further.
I agree with you that OBP is a statistic that has been around for a long time. However, I strongly and fundamentally disagree with you when you state that OBP is "a statistic that everyone knew was important". My interpretation of moneyball is essentially about how people did NOT know how important OPB was and therefore highly undervalued it. Billy Beane recognized this market inefficiency (he valued OPB much more than most other GMs) and exploited it (able to assemble a high performing team with little money). If everyone really knew how important OBP was, then Beane could not have utilized it to his advantage (e.g., everyone would select players based on OBP). Thus, I believe your assertion that OBP was a statistic "that everyone knew was important" is wrong and fundamentally misses the idea of moneyball.
Ultimately, I want to find an individual hockey statistic that reliably predicts team level performance (like OBP). I also want this statistic to be undervalued in the market so that a GM could exploit its inefficiency (like Beane did with OBP). I don’t think that goals, assists, and plus-minus fits this bill as I think that everyone already values these metrics (whether in isolation or some weighted combination) and therefore they are efficient. My point of my original posts were to see if there is evidence that Corsi (and related concepts) is such a stat that can be leveraged for predicting performance beyond traditional metrics. If so, I believe that Corsi could be leveraged due to inefficiencies in valuing it (but I first want to see that it does predict team level outcomes likes OBP).
Yes, saying OBP is “a statistic everyone knew was important” is incorrect. It wasn’t the statistic everyone knew was important, it was the act behind it. Everyone in baseball wanted guys who hit for average. They wanted guys who hit for power. They wanted guys who had “the five tools”.
What Beane did was use OBP while everyone else was focusing on batting average and power. It’s not that people didn’t value getting on base, it’s that they didn’t value a 250 hitter who had a 340 OBP. So no, OBP wasn’t what everyone knew was important, but “getting on base” was important. They simply didn’t look at walks.
Secondly, Eric and I have already linked to numerous sites detailing how Corsi does predict team level outcomes.
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 Oct 10, 2011 9:20 AM EDT up reply actions
Got it – Thanks for your clarification on the interpretation of moneyball.
In terms of the sites showing that Corsi predicts team level outcomes, I appreciate the links provided. I have been reading through these and others articles in order to digest the evidence for this statistic relating to team outcomes. So far, I am not overly convinced as I personally think the evidence is not very strong – I understand that we will most likely markedly disagree in this assessment, which is fine. Ultimately, I hope to be able to run some analyses on my own in order to best satisfy my curiosity.
Right on. Sorry I misrepresented my position.
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 Oct 10, 2011 12:12 PM EDT up reply actions
If you don’t think GVT and VUKOTA are “advanced statistics”, I’m really, really curious what you consider “advanced statistics”.
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 Oct 9, 2011 8:24 PM EDT up reply actions
I will answer your question below about what I consider "advanced statistics", but I would first like to turn the question back to you:
"If you DO think that GVT and VUKOTA are "advanced statistics", I’m really, really curious what you consider NOT to be "advanced statistics?"
That is, do you really think that a weighted combination of goals, assists, plus-minus, and TOI is an "advanced statistic"? If so, I would contend that just about anything could be an "advanced statistic", which dilutes the value of the label and definition, in my opinion.
Thus, my question to you is what is not an advanced statistic? For example, are just goals and assists in isolation not an advanced statistic? However, what if I create a weighted metric of goals and assists in combination? Is this combined measure statistic really "advanced" (where as goals and assists in isolation is not advanced)? I don’t think this combined measure is advanced.
To answer your question, I consider "advanced statistics" to be measures that I highlighted in my prior post above: Corsi, Corsi QoC, Corsi Rel QoC, Relative Corsi, Fenwick, QualComp, sheltered minutes, zone start percentage, etc. My view of "advanced statistics" overlaps fairly highly with the BSH Advanced Stats Glossary
I do not consider traditional measures of goals, assists, and plus-minus to be advanced. Therefore, I do not consider a weighted combination of these same measures to be advanced. As a result, I do not consider GVT an advanced statistic (it is harder to evaluate VUKOTA as HP does not publish their specific calculation of it, so it cannot be dissected).
What I don’t consider advanced statistics are goals, assists, points, PIM, shots, takeaways, giveaways, hits, plus/minus, and more. All of those are simply counting stats (save plus/minus) and don’t require anything more than adding.
Advanced statistics are those that require more than simply counting. They are those that aren’t found in a boxscore or fantasy hockey. They are those that weren’t around 10 years ago, even if the events they track were.
But I’m still confused as to where you draw the line between traditional and advanced. If GVT isn’t advanced because it weights traditional measures, how are Corsi and Fenwick advanced, since all they do is count and divide shots? Shots aren’t traditional? Five of your examples are measures that derive almost exclusive from shots, which I thought were quite traditional.
A sixth example is QualComp, which is a weighted measure of goals by time on ice, which you’ve already explained is a traditional statistic, thus not advanced. I’m not sure what you are referring to by “sheltered minutes”, but the only stat I recognize in your list that comes from something other than traditional statistics is Zone Start percentage.
So I’m definitely having trouble with where you draw the line between what is “traditional” and what is “advanced”, if goals are traditional, but shots are advanced.
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 Oct 10, 2011 9:08 AM EDT up reply actions
Thanks for the clarification – I think this may be a matter of semantics, which isn’t all that important. Indeed, I greatly dislike the term "advanced statistics" given that I don’t believe that one set of statistics can be more "advanced" than another. Essentially, I see these all of these statistics as different tools, and the appropriateness of them depends on the context or the job required in terms of the intended purpose and conclusions.
Ultimately, I don’t care if a statistic is classified as "advanced" or not – In fact, I would prefer the term to disappear altogether. Instead, I am interested in finding highly predictive statistics that are inefficient in the market. That is, I want to find statistics that really predict outcomes very well, but not that many people realize that they predict outcomes. If a statistic is very predictive and efficient, that is great…but I (or a GM or anyone) wouldn’t be able to leverage it for value as everyone would already know that it is indeed predictive.
Ultimately, I see GVT as inefficient – It definitely does predict important outcomes, but I think that people already fundamentally know that it predicts outcomes (in terms of how people essentially know that a combination of goals and assists predict wins). Thus, to me, the value of GVT isn’t as great as the potential value of Corsi…as most general hockey fans haven’t really considered if Corsi is related to outcomes and therefore this information can be leveraged for gains (if it is indeed really related to outcomes, which is what I am trying to understand with my original posts).
I also avoid “advanced stats” whenever possible. The terms I use are “traditional” and “non-traditional”.
I classify GVT as clearly non-traditional. But I also kind of hate it, so I’m hesitant to call it advanced — at least until someone shows that it correlates with positive outcomes better than the alternatives (and I’ll lump goals and points in as alternatives).
@BSH_EricT
Writer at Broad Street Hockey
Okay, now I understand. I’m definitely with you and Eric on the naming problem. Advanced stats is not something I like, even though I use the phrase often. When people attack Corsi as an advanced stat, I have a hard time understanding how a plus/minus of shots is advanced, but a plus/minus of goals is preferred.
So I don’t think we disagree much at all. Hockey statistics are definitely in their infancy, and they aren’t very “advanced”, even if there are non-traditional numbers now. Predicting outcomes is something I’m always skeptical of – how do you predict outcomes in any sport, really? – but that’s what all statistics strive to do. They just do it based on descriptive numbers.
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 Oct 10, 2011 12:18 PM EDT up reply actions
While it is true that baseball’s individual nature is easier to analyze, hockey’s advanced statistics are beginning to get a foothold in the sport. Corsi has been shown to correlate with zone time, scoring chances, and is better at predicting future success than previous winning percentage.
But let’s not forget that the big new advanced sabrmetric stat in Moneyball was on base percentage. That’s hardly very advanced.
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 Oct 8, 2011 1:14 PM EDT up reply actions
Except that OBP wasn’t big and new when Alderson began focusing on sabermetrics in 1995. It was an official MLB-reported stat in 1984. It just took 11 years for anyone to use (or at least admit using) it in a significant manner.
Bob.
Note that your post says that MLB started officially reporting them in 1984. IBM PCs and Macintoshes, ftw! (or in 1984 maybe CPM?!!!!?)
I specifically recall calculating OBP off newspaper boxscores in the early 1970s. I calculated OPS on my own at that time (although usually would separate out OBP and SLG since I didn’t know to add them together!).
These were stats well known many years ago. Could take awhile for the NHL as well.
Hunter Pence did not guarantee a WS, but, then, neither did Carlos Beltran.
This (in combination with your other posts below) is by far the most interesting critique of NHL advanced stats that I’ve heard.
HOWEVER, please note that most sabrmetrics are still NOT predictive in nature. In fact, while you mention that much of Corsi et al are “descriptive statistics”, the regression lines that Eric T. creates are actually much more predictive than most baseball stats.
I want Eric T. to weigh in on this.
Hunter Pence did not guarantee a WS, but, then, neither did Carlos Beltran.
I agree.
I’m not making a claim about whether the critique is accurate, but it is articulate and captures much of my own skepticism. I appreciate someone taking a run at it – my own knowledge of statistics in general is average and my understanding of advanced hockey statistics is grossly insufficient to challenge. A good discussion can only move things forward.
by flyersfaninchicago on Oct 8, 2011 7:35 PM EDT up reply actions
As I understand the critique, it is basically “these stats are fine for describing what has happened but useless for making predictions.”
I wholeheartedly disagree. Need some examples?
In this article, Hawerchuk explained why stats said that New Jersey (who had 20 points at the time) was a better team than Dallas (who had 46 points), something that was certainly not conventional wisdom, and almost half of the readers predicted Dallas would outscore New Jersey by at least 10 points the rest of the way. The result? New Jersey outscored Dallas by 12 points as they made a shocking playoff run and Dallas fell out of the playoffs.
In this article, Derek Zona predicted marked changes for many of the Oilers players, which were almost all borne out by the actual results in the following season.
In 2010, pretty much every stats guy (I think the first was Hawerchuk) was saying the Avs were a below-average team that would squeak into the playoffs because their early start put enough points in the bank to get them there. The Avs fans rebelled, of course, and even after three months of .500 play predicted a division win and wanted to stay pat at the deadline — not because they knew their team wasn’t going anywhere, but because they were afraid of changing anything on a team they perceived as playing very well.
If you want something from the writers on this site in particular, how about the repeated predictions that momentum didn’t matter come playoff time (the Flyers, you may recall, lost 5 of their last 6 and 14 of their last 20, yet ousted the Sabres who had won their last four and 8 of their last 10)?
I’m probably missing literally 100 other examples. SB Nation’s archives are filled with examples of articles making predictions based on stats that were contrary to popular thought and proved to be true. Hell, Geoff’s comment right above this one links to an article that shows that if you want to predict how many points a team will have in the remaining portion of the season, you’re better off looking at their Corsi Tied than at their points scored thus far in the season.
@BSH_EricT
Writer at Broad Street Hockey
by Eric T. on Oct 8, 2011 11:44 PM EDT up reply actions 1 recs
Thanks for these links and examples – I will check them out as well as search through the SB Nation’s archives for some of the 100 other examples to which you alluded. Indeed, I am looking for articles that clearly show based on inferential statistics (predictive statistics that produce a confidence interval based on standard error estimates) how individual level advanced statistic metrics are related to team performance. If you or anyone else thinks of any other examples along these lines (I will search the archives on my own to see if I find anything), I would appreciate if you could pass them along as I truly am interested in reading them to see what support is out there for the predictive validity of advanced statistics (I am especially interested in the standard errors around the prediction and the resulting confidence intervals).
Since Geoff originally mentioned the article on Corsi being a better predictor of future success then prior winning percentage and you highlighted it again, I did read through this article as it seems to be directly related to my interest (how an individual metric is inferentially related to team performance). While I did find the article informative and I applaud the author for all of his work on the analyses, I personally found the method and approach flawed. While I think the author utilized a very valuable approach in terms of bootstrapping (i.e., how he ran the analysis 1000 times), this approached ignored any time series based relationship of data (which is a fundamental flaw in my opinion).
In particular, with his approach, he examined how Corsi (actually just Corsi tied) in one set of games predicted winning in another set of games. Further, this was sampled randomly (actually a stratified random sample) from the season meaning that a team’s Corsi in game 80 could have been used to predict an outcome of win/loss in game 4. While one can argue for this approach, I find it problematic for a number of reasons (mainly because of time series questions in casual direction). There are also a number of other problems from a statistical perspective (violation of statistical assumptions, appropriate of statistical tests employed, etc.) that I won’t go into as it is more of a technical discussion for another day.
While there are some definite positives with this approach and I think the author did some very good things, I personally wouldn’t hang my hat on this study as evidence of a causal relationship between Corsi and game outcome. Just my lay perspective and I am sure that a number of people will disagree with my assessment, which is great.
JLikens uses this method quite often, and a few others use it on occasion as well. I’m not enough of a statistician to know whether its flaws render its results untrustworthy, but I strongly encourage you to do the analysis an alternative way if you think the selected method is poor — if you get the same answer, then you’ll have confidence in his results, and if you get a different answer then you’ll have a major impact on the community and advance our understanding.
But until then, I’m going to have to use the best available information, which does seem to consistently point towards shot-based metrics being more predictive than goal-based metrics.
@BSH_EricT
Writer at Broad Street Hockey
Thanks for your encouragement. I would love to do these analyses for myself. Indeed, I have a few ideas about how to better model some of these outcomes by using some multilevel analyses such as HLM or RCM.
However, my problem has been that I haven’t been able to find any of the individual level data in a useable format. Do you know where I can get my hands on a file of the “advanced statistics” for every player for every game (from last year and prior years)? Ideally, this would be in a .dat format, but I could also work with .xls or other data formats.
If you can let me know where I can download this data, I would love to run some analyses and would be happy to post my results and method, so that it can be critiqued and evaluated as I have done to these other studies. I would be very interested to see what we find as my original posts have ultimately been trying to better understand the utility of advanced statistics.
The data normally have to be scraped from the NHL’s play by play files, one game at a time.
I have files for the last couple of years that I’d be willing to forward if you want. Email me (contact info at bottom of page).
@BSH_EricT
Writer at Broad Street Hockey
Your offer to share your data files is very kind and generous, Eric. I can only imagine how much effort and long hours it took to pull together the raw data. While I would certainly very much appreciate receiving the files so that I could run some analyses on them, I would obviously understand if you didn’t want to share them (as they may be proprietary, etc.). I will email you to discuss further. Thanks again for your generosity.
I don’t mind sharing. Yeah, I put a fair bit of time into it, but a) maybe your study really will advance the community in some important way, which I’d be happy to help with, or b) maybe you’ll be in a position to do a favor for me sometime, Enzo.
I’ll follow up off-list.
@BSH_EricT
Writer at Broad Street Hockey
Let’s see if I got this straight.
The A’s had the lowest payroll in the league that year, and used sabermetrics to construct their roster.
That roster won 103 games, the most in the league despite having the lowest payroll. Since then, every team in the league has started using sabermetrics. Many of them have turned their entire operation over to a sabermetric expert.
And yet you think that a single inning of play — the A’s lost the series in 5 games, losing game 5 by a single run when their closer with conventional stats of 11-4, 44 saves, and a 3.27 ERA gave up three runs in the 9th inning — you think that single bad inning by a single reliever who had good traditional numbers proves that sabermetrics don’t work.
Is that right?
@BSH_EricT
Writer at Broad Street Hockey
by Eric T. on Oct 7, 2011 6:52 PM EDT up reply actions 1 recs
Geoff it wasn’t even necessary. Really
Commenter formerly known as M from Pdaddy, but still just Call Me "M"!
DISCLAIMER: Information written above may not be entirely factual nor provable with the use of complex statistics. But it may induce thought, humor and possibly laughter.
by MJDII on Oct 7, 2011 6:02 PM EDT via mobile up reply actions
Yeah, that’s the idea.
It shows whether you were matched up against the skaters who tended to control play for the other team (usually their top lines) or the ones who tended to get pushed back (their bottom lines).
@BSH_EricT
Writer at Broad Street Hockey
Has anyone been playing around with how zone starts effect Corsi of competition? Probably, I just forget.
Keeping alive the old Vaudeville joke, "I'd rather be dead than play Philadelphia."
You mean like with out being coersed.
Commenter formerly known as M from Pdaddy, but still just Call Me "M"!
DISCLAIMER: Information written above may not be entirely factual nor provable with the use of complex statistics. But it may induce thought, humor and possibly laughter.
by MJDII on Oct 7, 2011 6:33 PM EDT via mobile up reply actions
"Defensive zone" vs. PK
Good analysis as always Eric. I’m happy Cou seems to have held his own last night.
However, I mentioned at the end of the game thread I am not so happy he was regularly on the PK. This works against Boston. It would not work against many other teams (Pittsburgh? Tampa? LA?). I truly fear that his precocious defensive-forward capabilities may have partially influenced the waiving of Betts (though as I’ve posted, I think other things were at play there).
I’m ok with Read on PK; he is a very different story as a 25 year old (physically mature as well as experienced).
The question is, after Nodl/Read, who is going to pair with Talbot on PK? Given the changes in personnel and responsibilities, I am way-less thrilled about G taking a regular PK shift.
Hunter Pence will not guarantee a WS, but, then, neither does Carlos Beltran.
The New Phone Book Is Here! The New Phone Book is Here!
Attention all lawyers and members of law enforcement:
Here is the .pdf of the 2011-2012 NHL rulebook.
http://www.nhl.com/nhl/en/v3/ext/pdfs/2011-12_RULE_BOOK.pdf
and the URL
by flyersfaninchicago on Oct 7, 2011 2:04 PM EDT reply actions
have not found the “summary of changes” page for ’11-12.
by flyersfaninchicago on Oct 7, 2011 2:08 PM EDT up reply actions
hah! took me a minute, forgot my own reference.
by flyersfaninchicago on Oct 8, 2011 7:36 PM EDT up reply actions
I think I read somewhere over the summer that Toews was sort of an upper-limit, best-case-scenario comparison for Couturier, and that Staal was a safe comparison. So it’s interesting to see that, based on how he was deployed in this game, they’ve got him going more towards Staal so far.
It’ll be real nice if he gets his touch on offense going soon.
by everybodyhitswoohoo on Oct 7, 2011 2:26 PM EDT reply actions
Bryz
“I took my gear off and return to the hotel, took a 20-minute nap and came back again,”
I may hate his contract and not want him here, but if he’s going to be here, it’s nice that he’s funny. And to be honest, this is what the Flyers need to keep them buzzed through the not so much Hockey fans around here. You bring in a real personality, people will take notice.
It's in his wheelhouse!!
Carlos Ruiz, My Nickname is Chooch.
I didn’t like his contract and still don’t like it, but if he plays the rest of the season how he played yesterday, then you won’t here a chirp out of me about it.
I don’t think attracting new fans has anything to do with why he’s here. Winning generates the buzz. He’s here to make us winners — and by ‘winners’ I mean Stanley Cup Champions. You don’t blow up a team that went to the Finals two years ago and was the best team in hockey for the first half of last year unless you think they overachieved and/or didn’t have what it takes to get over the hump. Was Richards a locker room problem? I don’t know. Was he a beast on the ice? Yup. He’s still my favorite NHL-er, but if Bryz makes us better, then I’m on board. If I thought all the moves last year made us a significantly weaker team, then I would be all over Homer for it. But I can’t say it does, nor can I say it makes us significantly better. From preseason and game 1, the changes look promising at least. So, I’m up in the air about Bryz being brought in and Richards being shipped out. But as a Flyers fan, I’m geeeked up at the potential, so I cannot say it’s a bad deal — or I wouldn’t be geeked up.
by fiveredapples on Oct 7, 2011 3:56 PM EDT up reply actions
Richards had 1 clutch G, 1 clutch A for the Kings today
Fixed.
Keeping alive the old Vaudeville joke, "I'd rather be dead than play Philadelphia."
You’ve come over to the dark side!
Commenter formerly known as M from Pdaddy, but still just Call Me "M"!
DISCLAIMER: Information written above may not be entirely factual nor provable with the use of complex statistics. But it may induce thought, humor and possibly laughter.
by MJDII on Oct 7, 2011 5:48 PM EDT via mobile up reply actions
I almost tagged that comment at the end with “/MJD bait”
Keeping alive the old Vaudeville joke, "I'd rather be dead than play Philadelphia."
So your not coming over to the darkside????
Commenter formerly known as M from Pdaddy, but still just Call Me "M"!
DISCLAIMER: Information written above may not be entirely factual nor provable with the use of complex statistics. But it may induce thought, humor and possibly laughter.
by MJDII on Oct 7, 2011 6:06 PM EDT via mobile up reply actions
Dynamite stuff, Eric
I’ve had this same concern with Nashville, in light of Joel Ward and Marcel Goc being replaced by kids.
Managing Editor of On the Forecheck, SB Nation's blog covering the Nashville Predators, and founder of HockeyGearHQ, a site devoted to hockey equipment and accessories.
The comments in this post really took some odd turns. It was a little crazy.
Commenter formerly known as M from Pdaddy, but still just Call Me "M"!
DISCLAIMER: Information written above may not be entirely factual nor provable with the use of complex statistics. But it may induce thought, humor and possibly laughter.
by MJDII on Oct 7, 2011 5:51 PM EDT via mobile reply actions
You’re right. Your comment made me look up at the title, and I was surprised to see that the whole Richards discussion made me completely forget that we were actually talking about defensive minutes for young forwards.
Lightning strikes once, Hextall strikes twice!
"I think there is virtue in pissing off idiots." - Fehr and Balanced
Eric,
Great work as usual.
Something I wonder about is how much of the lack of tough time is due to the players not being ready versus just having better guys available. The Pens were pretty awful when Staal came in so I’m not surprised he’s an exception his rookie year. If he’d somehow slipped in the draft to a decent team I’m sure he would have been protected as well.
The Flyers aren’t in nearly as bad a situation, but it’s clear that they’ll have to give some guys tough time that haven’t gotten it before.
Driving Play - The Blog with Three First Lines
Yeah, there’s no way to know, but most of the people who play at 18 years old were picked in the top ten and most of the teams that draft in the top ten aren’t very good. So I’m not inclined to suspect that it’s just an issue of who else was available, but I can’t rule it out.
Of course, even if you’re right, then the correct conclusion would be that the Flyers are uniquely lacking in defensive forwards, which would still make me nervous.
@BSH_EricT
Writer at Broad Street Hockey
off topic
Watch alittle of the the LA game. Got to watch the West Coast Flyers and Richie was dishin some sweet offerings and Gags was punchin people in the face. Yeah. Gags got a roughing minor. WTF. Guess I was walking down memory lane, but it was nice to see them being more than what the media has made them out to be.
Leino = $$$
by flyerfreek on Oct 8, 2011 2:08 AM EDT via mobile reply actions
Interesting Analysis
Very interesting article – Thanks for all of your analysis, Eric. I imagine that there was a lot of work in collecting and cleaning the data, so I appreciate all of your efforts in summarizing it. These are the type of articles that I personally think makes this site valuable and informative.
Something to say? Choose one of these options to log in.
Secondary Sidebar
User Tools
Related FanPosts on Broad Street Hockey
Related FanShots on Broad Street Hockey
FanPosts
Community blog posts and discussion.
Recommended FanPosts
Recent FanPosts
SBNation.com Recent Stories
In Association With
Latest Puck Daddy - NHL from Yahoo! Sports
- NHL 13 cover final 4: Claude Giroux, Evgeni Malkin, John Tavares, and Pekka Rinne
- Watch massive hit in Australian League; hear player’s mom discuss concussion (VIDEO)
- Wednesday’s Three Stars: Devils fourth line steps up; Watch Prust dive
- Stephen Gionta leads Devils in Game 5 triumph over the New York Rangers
- Rangers, Devils back at MSG for pivotal Game 5; NHL 13 screenshots (Playoff Puck Previews)
Latest stories and headlines from NHL.com
- NHL 13 final four set - Giroux, Malkin, Tavares, Rinne
- Carter rescues Devils; New Jersey moves within 1 win of Cup finals with 5-3 win over Rangers
- Rangers' best game fails; Devils once again find holes in New York G Lundqvist in 5-3 victory
- Lundqvist struggles as Rangers lose Game 5, 5-3
- Devils-Rangers Sum
