EHM The Blue Line

I might not remember every detail, but I can go look at Sabres PGTs on hfboards and see the constant praise about how hard he carried our blueline for much of the year

Risto's numbers are skewed because of his role and a partner with a CF% of around 43% when not playing with the 4th best defenseman in the league in his last 2 full seasons before coming to Buffalo. So yea, Risto is not a top 5 dman in the league and that's ok. But drawing the conclusion that he's not a top pairing defenseman is dumb imo. If Risto is at best 2nd pairing, then so is the golden boy Ekblad.

CJ wrote:You're a bit off-limit now when talking about players who haven't been mentioned here. But I get it. A little discussion is always good.

Glad you feel that way CJ I know I certainly do (best opportunity to clarify/learn things!)

Aladyyn wrote:I might not remember every detail, but I can go look at Sabres PGTs on hfboards and see the constant praise about how hard he carried our blueline for much of the year

Risto's numbers are skewed because of his role and a partner with a CF% of around 43% when not playing with the 4th best defenseman in the league in his last 2 full seasons before coming to Buffalo. So yea, Risto is not a top 5 dman in the league and that's ok. But drawing the conclusion that he's not a top pairing defenseman is dumb imo. If Risto is at best 2nd pairing, then so is the golden boy Ekblad.

Why is Ristolainen and Ekblad interchangeable? I don't think they are, i think Ekblad is a much better player. Ekblad is a legit first pairing guy already and he's better then Ristolainen both offensively and defensively imo.

If Ristolainen was so good, why doesn't the numbers support that? Any numbers? It's not just his raw numbers that aren't good, his Relative numbers aren't good either. Not a single analytics person is saying that Ristolainen is a top pairing D-man, they all use different models and NOT ONE model has Ristolainen as good. How do you explain that? I'm not looking at just raw CF% and saying he's bad because of that, not a single number supports his claim as a top pairing D-man.

A quote from Moneyball fits here. "If he's a good hitter, why doesn't he hit good?", change it do "If he's a good player, why doesn't he play good?"

Aladyyn wrote:I can go look at Sabres PGTs on hfboards and see the constant praise about how hard he carried our blueline for much of the year

I don't give much credence to the hfboards myself; I tried reading more hfboards posts this summer and was surprised at the lack on reasoned thought/debate, the failed attempts at being witty and the level of uniformed opinion...

To give some more clarification on RIstolainen, he was not supposed to be the guy on our blueline before the season. Bogosian was supposed to be on the top pairing, because he played well after the trade and before he was injured, he was better than Myers had been so everyone got pretty excited. But Bogosian got injured and Risto was defaulted to the top pairing. And he played well. He played so well that Bylsma just couldn't stop riding him as hard as he could, even after RIsto started slumping around the midway point of the season. Bylsma prioritized winning more games above player development, which stopped him from doing lots of things and made some things take longer than they should. The main thing that he fluffed up about Risto was not playing Bogosian as 1LD for any decent period of time. When they were together for stretches during games, they always looked good, Bogo was not struggling with playing on his off side too much and he was a lot better than any LHD options on the roster (McCabe can't handle top pairing duties in his rookie season, even though he was pretty much on par with Bogo at the end of the season). Our first pairing was also buried in responsibilities due to the need to shelter the bottom-4 (not trusting Bogosian to carry rookie McCabe, bottom pairing was a tirefire outside of Pysyk and he wasn't that good after he got injured). So in the end, the Sabres' best offensive defenseman ended up getting the hardest shutdown role, which he should not be expected to handle at his age. It's a shame that he didn't have Campbell of Byfuglien to boost his numbers last season, or the best top 4 in the league to play behind and thrive in an easy role.

Basically it's Bylsma's fault

Peter_Doherty wrote:If Ristolainen was so good, why doesn't the numbers support that? Any numbers? It's not just his raw numbers that aren't good, his Relative numbers aren't good either. Not a single analytics person is saying that Ristolainen is a top pairing D-man, they all use different models and NOT ONE model has Ristolainen as good.

Bah! raw numbers! analytics! Wait until the TBL Rosters are released and CJ will tell us how good he is

So, what ur saying is that Ristolainen got overplayed? Which is exactly what i said to begin with He can't handle a top pairing role yet, hence his numbers are bad.

nino33 wrote:

Peter_Doherty wrote:If Ristolainen was so good, why doesn't the numbers support that? Any numbers? It's not just his raw numbers that aren't good, his Relative numbers aren't good either. Not a single analytics person is saying that Ristolainen is a top pairing D-man, they all use different models and NOT ONE model has Ristolainen as good.

Bah! raw numbers! analytics! Wait until the TBL Rosters are released and CJ will tell us how good he is

Hahaha

Peter_Doherty wrote:

Aladyyn wrote:I might not remember every detail, but I can go look at Sabres PGTs on hfboards and see the constant praise about how hard he carried our blueline for much of the year

Risto's numbers are skewed because of his role and a partner with a CF% of around 43% when not playing with the 4th best defenseman in the league in his last 2 full seasons before coming to Buffalo. So yea, Risto is not a top 5 dman in the league and that's ok. But drawing the conclusion that he's not a top pairing defenseman is dumb imo. If Risto is at best 2nd pairing, then so is the golden boy Ekblad.

Why is Ristolainen and Ekblad interchangeable? I don't think they are, i think Ekblad is a much better player. Ekblad is a legit first pairing guy already and he's better then Ristolainen both offensively and defensively imo.

If Ristolainen was so good, why doesn't the numbers support that? Any numbers? It's not just his raw numbers that aren't good, his Relative numbers aren't good either. Not a single analytics person is saying that Ristolainen is a top pairing D-man, they all use different models and NOT ONE model has Ristolainen as good. How do you explain that? I'm not looking at just raw CF% and saying he's bad because of that, not a single number supports his claim as a top pairing D-man.

A quote from Moneyball fits here. "If he's a good hitter, why doesn't he hit good?", change it do "If he's a good player, why doesn't he play good?"

Ekblad's CF% when not playing with Brian Campbell was firmly in Ristolainen territory last season.

It's very hard to contextualize stats in hockey and it hasn't been done well yet. Zone starts are not useable as hard data and no good QoC metric has been found yet. The models I've seen also generally use more than 1 season of data which makes them basically useless when talking about Ristolainen as he only really has 1 season of relevant data and in that season he was set up to fail. I explain that as bad usage and coaching.

Peter_Doherty wrote:So, what ur saying is that Ristolainen got overplayed? Which is exactly what i said to begin with He can't handle a top pairing role yet, hence his numbers are bad.

He can handle a top pairing role. He can't handle a shutdown top pairing role with Josh Gorges as his partner. If he played with someone as good as him, they'd make a good top pairing.

Ekblad do have other numbers that support his claim as a top pairing D-man, it's not all CF%. I'm definitely not just looking at CF% when i say that i don't think Ristolainen is a top pairing guy.

"He can handle a top pairing role. He can't handle a shutdown top pairing role with Josh Gorges as his partner. If he played with someone as good as him, they'd make a good top pairing."

I don't think he would be a good top pairing guy with a good partner, there are no numbers available to support that theory, like at all.

Seriously though, we're not going anywhere with this, you are clearly not changing your mind here and neither am i. We had a good discussion so let's end it before we take a bad turn and it stops being a good discussion

CJ wrote:I think we have enough votes for the first poll:

Here are the results (based on a 3 points, 2 points, 1 point for the positions given)

1) DEFENSEMEN

POLL 1 - ANA
1) Hampus Lindholm 31
2) Sami Vatanen 23
3) Cam Fowler 12

POLL 2 - NYI
1) Travis Hamonic 27
2) Nick Leddy 25
3) Johnny Boychuk 14

POLL 3 - NYR
1) Kevin Klein 27
2) Marc Staal 25
3) Dan Girardi 14

____________________________________________________________________

And here's the next poll for you. The final Poll featuring defensemen.

2) DEFENSEMEN #2

POLL 1 - CBJ
1) Jack Johnson
2) Ryan Murray
3) Seth Jones

POLL 2 - NJ
1) Ben Lovejoy
2) Damon Severson
3) John Moore

POLL 3 - DET
1) Danny DeKeyser
2) Mike Green
3) Niklas Kronwall

I'll just quote this so people can find the new poll easier since our discussions have taken up quite a lot of space

Peter_Doherty wrote:Not sure where you are getting your information

Here: http://stats.hockeyanalysis.com/ratings ... rtdir=DESC

nino33 wrote:

CJ wrote:You're a bit off-limit now when talking about players who haven't been mentioned here. But I get it. A little discussion is always good.

Glad you feel that way CJ I know I certainly do (best opportunity to clarify/learn things!)

This forum is too quiet for my taste (as I visit this site many times a day). So discussion is more than welcome here!

Very surprised there was 10+ unread posts here when I logged in.

nino33 wrote:

Peter_Doherty wrote:If Ristolainen was so good, why doesn't the numbers support that? Any numbers? It's not just his raw numbers that aren't good, his Relative numbers aren't good either. Not a single analytics person is saying that Ristolainen is a top pairing D-man, they all use different models and NOT ONE model has Ristolainen as good.

Bah! raw numbers! analytics! Wait until the TBL Rosters are released and CJ will tell us how good he is

Hahahaaa....

For the upcoming TBL Rosters. I've done all D-men already. Now it's just "fine-tuning". A reason why I started this thread/poll.

You talking about iCorsi/60? Do you know what that stat means?

It's his individual Corsi For per 60 minutes. As in, how many shot attempts does he personally have per 60 minutes. Tanev doesn't drive offense and he's definitely not a goalscorer so this doesn't surprise me at all, it does however not mean much either since iCorsi isn't really a relevant stat for a defensive shutdown defenseman.

Look at this to see how incredible Tanev is defensively, http://stats.hockeyanalysis.com/ratings ... ortdir=ASC

This shows his FA60 RelTM (How many unblocked shot attempts he allows in comparison to his teammates), -10.53 is just insane, he's over -2.5 ahead of #2 on the list.

Peter_Doherty wrote:You talking about iCorsi/60? Do you know what that stat means?

It's his individual Corsi For per 60 minutes. As in, how many shot attempts does he personally have per 60 minutes. Tanev doesn't drive offense and he's definitely not a goalscorer so this doesn't surprise me at all, it does however not mean much either since iCorsi isn't really a relevant stat for a defensive shutdown defenseman.

I thought you talked about something similar. Well yes I know he's a shutdown dude who doesn't shoot hence the bad iCorsi/60.

There's so much advanced stats one can look at. It's a bit confusing who is referring to what.

EDIT:
So I believe you talked about this then:
http://stats.hockeyanalysis.com/ratings ... rtdir=DESC
So if you'd be a not so good player and played on a Top line with two star players and great D-men you'd have a lot better Corsi stats than when on your 4th line?

CJ wrote:

Peter_Doherty wrote:You talking about iCorsi/60? Do you know what that stat means?

It's his individual Corsi For per 60 minutes. As in, how many shot attempts does he personally have per 60 minutes. Tanev doesn't drive offense and he's definitely not a goalscorer so this doesn't surprise me at all, it does however not mean much either since iCorsi isn't really a relevant stat for a defensive shutdown defenseman.

I thought you talked about something similar. Well yes I know he's a shutdown dude who doesn't shoot hence the bad iCorsi/60.

There's so much advanced stats one can look at. It's a bit confusing who is referring to what.

Aight, i haven't used iCorsi at all, it's a very specific stat that is hard to find a good context to use. I get it though, it's easy to get lost in the world of analytics, especially if you are new to it. All the stats have weird and similar names. That's why i try to stick to CF% or the different variations of CF% like Rel.CF% to keep it as simple as possible.

Peter_Doherty wrote: Aight, i haven't used iCorsi at all, it's a very specific stat that is hard to find a good context to use. I get it though, it's easy to get lost in the world of analytics, especially if you are new to it. All the stats have weird and similar names. That's why i try to stick to CF% or the different variations of CF% like Rel.CF% to keep it as simple as possible.

I edited my post. Added a question there.

Yes, but i mostly use Rel.CF% or CF% RelTM which accounts for the team or teammates to give a more accurate picture. CF% is very team dependant, a bad player who is on a good team/good line will probably still have a good CF%.

Using raw CF% without context is not useless but it's not very good either.

To answer your question; Yes, you would have a much better CF% if you played with good (as in players who drive shot attempts) players then if you played on the 4th line.

Better answer, use Rel stats instead

C'mon people. We need more votes for POLL DEFENSEMEN #2.

I don't know enough about those players to dare cast a vote here, sorry guys

Alright, I'll give it a shot

2) DEFENSEMEN #2

POLL 1 - CBJ
1) Seth Jones
2) Ryan Murray
3) Jack Johnson

POLL 2 - NJ
1) Damon Severson
2) Ben Lovejoy
3) John Moore

POLL 3 - DET
1) Danny DeKeyser
2) Mike Green
3) Niklas Kronwall

Torn on Lovejoy/Moore and DDK/Green, could go either way

Bump so more people can vote

1 Jones
2 Murray
3 Johnson

1 Severson
2 Lovejoy
3 Moore

1 Green
2 DeKeyser
3 Kronwall

Results for Poll #2.

2) DEFENSEMEN #2

POLL 1 - CBJ
1) Seth Jones 27
2) Ryan Murray 17
3) Jack Johnson 10

POLL 2 - NJ
1) Damon Severson 24
2) Ben Lovejoy 14
3) John Moore 10

POLL 3 - DET
1) Mike Green 24
2) Danny DeKeyser 20
3) Niklas Kronwall 10

_________________________________________________________________

NEXT POLL!

3) FORWARDS #1

POLL 1 - BOS
1) Brad Marchand
2) David Krejci
3) David Backes

POLL 2 - CAR
1) Elias Lindholm
2) Lee Stempniak
3) Victor Rask

POLL 3 - COL
1) Gabriel Landeskog
2) Matt Duchene
3) Nathan MacKinnon

Ooooh, forwards. A lot easier to rank than defensemen for sure

POLL 1 - BOS
1) David Krejci
2) Brad Marchand
3) David Backes

Krejčí vs Marchand basically comes down to playing C and being a consistently much better PP producer (Marchand had more PP goals and points than the previous 3 seasons combined last year somehow).

POLL 2 - CAR
1) Victor Rask
2) Lee Stempniak
3) Elias Lindholm

Rask is good, Stempniak is alright, Lindholm is still trying to find his footing

POLL 3 - COL
1) Nathan MacKinnon
2) Gabriel Landeskog
3) Matt Duchene

MacK is a beast, not a fan of Duchene.