2017-18 Defensemen Pair Analysis

[nfreeman]

I like Skinner too.  

[TrueBlueGED]

Just now, nfreeman said:

I like Skinner too.  

I'm shocked you like Skinner. He plays like Derek Roy! Every irritating thing about Roy's game is there with Skinner. 

[Randall Flagg]

1 minute ago, TrueBlueGED said:

I'm shocked you like Skinner. He plays like Derek Roy! Every irritating thing about Roy's game is there with Skinner. 

(I liked Derek Roy)

[TrueBlueGED]

5 minutes ago, Randall Flagg said:

And I can pretend to be smart like this because we all know that in reality, I have opinions like loving Jeff Skinner (and believing in the Hurricanes in general), betting everything I own on Vegas being a bottom 5 team even when they had a winning record in December, etc. 

As far as Skinner goes, at least you admit to being a for good edge work ?

Carolina, I think, is the perfect example of what happens to a team of they don't have a real top center. They have a lot of pieces to like, but without that, it doesn't seem to matter. Of course, league average goaltending would help too. 

[Randall Flagg]

Also, IKP, that post really was great. I just hate how our defense is built so I try not to think about it and thus clutter the thread with unrelated topics, and I apologize 

[TrueBlueGED]

1 minute ago, Randall Flagg said:

(I liked Derek Roy)

Me too. But if he weren't a Sabre, I'd have wanted him to keel over. Without question. 

[Randall Flagg]

8 minutes ago, TrueBlueGED said:

As far as Skinner goes, at least you admit to being a for good edge work ?

Carolina, I think, is the perfect example of what happens to a team of they don't have a real top center. They have a lot of pieces to like, but without that, it doesn't seem to matter. Of course, league average goaltending would help too. 

 

I don't think I've ever personally watched a hockey player get more out of their edges. He's the one that figure skated his whole life, isn't he?

[TrueBlueGED]

Just now, Randall Flagg said:

I don't think I've ever personally watched a hockey player get more out of their edges. He's the one that figure skated his whole life, isn't he?

I will not make the obvious Skinner plays like a figure skater joke. I will not make the obvious Skinner plays like a figure skater joke. I will not make the obvious.....welp. So much for that.

[nfreeman]

It wasn’t the diving that bothered me about Roy.  It was the gutlessness.  

To be fair, Skinner has never made the playoffs, so we don’t know whether he’d vanish under pressure like Roysie did.  

Also, Skinner is much faster and a much better finisher than Roy was.  And Skinner was a US Olympian.  

[WildCard]

10 minutes ago, nfreeman said:

It wasn’t the diving that bothered me about Roy.  It was the gutlessness.  

To be fair, Skinner has never made the playoffs, so we don’t know whether he’d vanish under pressure like Roysie did.  

Also, Skinner is much faster and a much better finisher than Roy was.  And Skinner was a US Olympian.  

What's the difference?

'Merica

[nfreeman]

The difference is that diving is an annoyance created in an attempt to give one’s team an advantage, while Roysie’s gutlessness manifested in his vanishing when it mattered, which very much disadvantaged his team.  

And gutless players don’t make the US (or Canadian) Olympic team. 

[French Collection]

Jeff Skinner is Canadian and has not played in the Olympics.

[nfreeman]

Yikes.  My mistake on Skinner.  

I still like him better than Roysie though.  

[5th line wingnutt]

11 hours ago, TrueBlueGED said:

The thing I always have trouble squaring is...if he does these things with the frequency he is accused, how could he possibly produce useful metrics? It's like the Thomas Vanek is lazy meme that was so popular. My hypothesis is that Beaulieu's mistakes are so comically glaring that they get seared into our minds in a fashion that does not lend itself to the human brain putting it in proper perspective relative to what he does bring to the ice.

This is not to say I think Beaulieu is good. I certainly do not. But I also do think he's a viable 3rd pair NHL defender, whereas I think the consensus is he's an AHL player.

I have never taken a hard look at advanced stats but try this on for size...

Take a random defenseman, I'll call him Nathan Beaulieu, (not his real name) and suppose that he plays a good, steady game for the most part.  He will have a really good CF%.  Now add in, on average, 2 catastrophically bad mistakes per game, resulting in 1.5 shots and .75 goals.  This lowers his CF% from really good to just plain good.  His goal differential will suck.  We use CF% because of sample size.  Maybe CF% is a good predictive stat but goal differential is a better explanatory stat.

[Randall Flagg]

2 hours ago, 5th line wingnutt said:

I have never taken a hard look at advanced stats but try this on for size...

Take a random defenseman, I'll call him Nathan Beaulieu, (not his real name) and suppose that he plays a good, steady game for the most part.  He will have a really good CF%.  Now add in, on average, 2 catastrophically bad mistakes per game, resulting in 1.5 shots and .75 goals.  This lowers his CF% from really good to just plain good.  His goal differential will suck.  We use CF% because of sample size.  Maybe CF% is a good predictive stat but goal differential is a better explanatory stat.

I think this literally nails it. 

[IKnowPhysics]

12 hours ago, Randall Flagg said:

I just hate how our defense is built so I try not to think about it and thus clutter the thread with unrelated topics, and I apologize 

No, you're right.  Just because a player's possession stats aren't as bad as you thought, it in no way implies that you have to like that player, especially the way this group played.

The effort here was to try to understand what Housley tried to do with player pairings, to see if there were any clues as to why the group was bad, or if there were any signs of life from the callups.

[Randall Flagg]

3 minutes ago, IKnowPhysics said:

No, you're right.  Just because a player's possession stats aren't as bad as you thought, it in no way implies that you have to like that player, especially the way this group played.

The effort here was to try to understand what Housley tried to do with player pairings, to see if there were any clues as to why the group was bad, or if there were any signs of life from the callups.

Yeah. And I like a few individual players on the defense, it just feels like we have such a weird and hard-to-put-together group on the low end of the league talent scale, even with Dahlin. 

 

[Mustache of God]

This was a great write-up, thanks for taking the time IKP.

I was surprised to see how "reliable" Baloo looks on that chart. Building on what 5th line wrote, it'd also be interesting to see the player's +/- during the playing time with each partner.

[IKnowPhysics]

2 minutes ago, Mustache of God said:

 it'd also be interesting to see the player's +/- during the playing time with each partner.

So, this could be done without a whole lot of trouble, and maybe I will.  But be warned: sample sizes for goals are a whole lot smaller than attempted shots, so the uncertainties in the goal analysis will get a lot larger, especially for pairings that only played a few games together.  It'll be hard to tell what's meaningful.

[TrueBlueGED]

2 hours ago, 5th line wingnutt said:

I have never taken a hard look at advanced stats but try this on for size...

Take a random defenseman, I'll call him Nathan Beaulieu, (not his real name) and suppose that he plays a good, steady game for the most part.  He will have a really good CF%.  Now add in, on average, 2 catastrophically bad mistakes per game, resulting in 1.5 shots and .75 goals.  This lowers his CF% from really good to just plain good.  His goal differential will suck.  We use CF% because of sample size.  Maybe CF% is a good predictive stat but goal differential is a better explanatory stat.

Goal differential is definitely better...eventually. For teams, as has been discussed, one season is normally enough for that convergence to happen. I say normally because you still have to be cognizant of outlying performances unlikely to replicate or sustain (abnormally high/low shooting and save percentages are the usual suspects). For instance, if you have a +40 goal differential but that's coupled with a 10% team shooting percentage, that's going to come crashing down to earth. In such a case, only looking at goal differential will lead you astray. 

Anyway, for individual players, you need around 3 years of goal data for it to be a better predictor than shot metrics. Because of how many fewer goal events individual players have than a team has, relying on only one season for individuals is going to leave you with a whole mess of false positives for the question at hand. 

Generalities aside, turning to Beaulieu, his both his CF% and GF% place him directly between Scandella (above) and Risto (below). That's not bad, until you consider he did that in a very sheltered role. His xGF% puts him below both of them, but above Tennyson, Falk, Antipin, and slightly above Nelson while slightly below McCabe and Gorges (who ranks #1 on the team in this metric for any defender who played at least 30 games).

Taking everything into account, I think the stats ultimately show Beaulieu is a mediocre 3rd pair defender whose possession value gets offset by blockheaded mistakes. He won't kill you in the right role, but you probably want to aim higher. 

[Thorny]

20 hours ago, Randall Flagg said:

Hey True, check this out: https://www.sabrespace.com/community/topic/25121-beaulieu-signs-for-2-years-24-aav/?do=findComment&comment=967027

and your reply.

Why are we so smart ?

 

On 7/31/2017 at 6:50 PM, Thorny said:

 

Good stats, bad eye test?

 

I think a lot will like him, if only because he's an inarguably large upgrade over Gorges or Falk.

?

[Thorny]

18 hours ago, nfreeman said:

The difference is that diving is an annoyance created in an attempt to give one’s team an advantage, while Roysie’s gutlessness manifested in his vanishing when it mattered, which very much disadvantaged his team.  

And gutless players don’t make the US (or Canadian) Olympic team. 

Eh...he did have 15 points in 18 playoff games in 06. And a better ppg than his regular season marks of 46 points in 70 games that regular season.

[nfreeman]

27 minutes ago, Thorny said:

Eh...he did have 15 points in 18 playoff games in 06. And a better ppg than his regular season marks of 46 points in 70 games that regular season.

Roy played heavily sheltered minutes in 2006, since the Sabres had so many good forwards that played ahead of him.  The next year, when it got harder in the absence of Dumont and Grier, he had 2 goals and 7 pts in 16 playoff games.  And, after Black Sunday, when it got much harder, he put up numbers in the regular season and then had zero goals and 2 assists in 6 playoff games in 2010.

Like Vanek, there's a reason Roy played with 5 NHL teams in 3 years after leaving Buffalo (and 4 Euro teams in 3 years after leaving the NHL).  Once other teams got a good look at each of them, they quickly decided that their numbers weren't a reliable indicator of quality hockey.

[Thorny]

8 minutes ago, nfreeman said:

Roy played heavily sheltered minutes in 2006, since the Sabres had so many good forwards that played ahead of him.  The next year, when it got harder in the absence of Dumont and Grier, he had 2 goals and 7 pts in 16 playoff games.  And, after Black Sunday, when it got much harder, he put up numbers in the regular season and then had zero goals and 2 assists in 6 playoff games in 2010.

Like Vanek, there's a reason Roy played with 5 NHL teams in 3 years after leaving Buffalo (and 4 Euro teams in 3 years after leaving the NHL).  Once other teams got a good look at each of them, they quickly decided that their numbers weren't a reliable indicator of quality hockey.

Ya, I notcied the declining numbers after the 06/07 season, but found that more indicative of the fact that Darcy had him in a role he was ill-suited for after the loss of the team's top 2 centers. 

The strength of your argument lies perhaps in the 06/07 playoff season, where it does seem to be that he underperformed.

Perhaps he was only ever best suited to a 3rd line scoring role. 

Edited July 9, 2018 by Thorny

[IKnowPhysics]

Here's a similar set of data as presented before, but now displaying goals for GF / goals against GA / goals for % (GF% = GF/(GF+GA)).  Refer to the ice times in the first post for a frame of reference.

GF/GA/GF%    Ristolainen	Bogosian	Scandella	Beaulieu	McCabe		Nelson		Guhle		Fedun		Tennyson
Ristolainen  R        		0/1/0.00    	26/36/41.94 	5/9/35.17	3/4/42.86	2/3/40.00	6/4/60.00	0/0/--.--	0/2/00.00
Bogosian            		R        	0/1/00.00   	1/5/16.67	3/5/37.50	0/0/--.--	0/0/--.--	0/0/--.--	0/0/--.--
Scandella                   		 	L        	2/1/66.67	1/2/33.33	7/11/38.89	1/2/33.33	0/0/--.--	0/3/00.00
Beaulieu                            				L		3/4/42.86	2/1/66.67	1/0/100.00	0/0/--.--	5/2/71.43
McCabe                                        					L		0/2/00.00	2/1/66.67	0/1/00.00	4/7/36.36
Nelson                                                						R		3/7/30.00	0/0/--.--	0/0/--.--
Guhle                                                       			 				L		0/0/--.--	0/0/--.--
Fedun                                                               			 					R		0/0/--.--
Tennyson                                                                   			 						R
	

Compare these stats with the team-wide stats:

Team Total (all situations): 198/278/41.60
Team Total (even strength 5v5): 119/176/40.34
Team Total (PK 4v5): 8/44/15.38
Team Total (PP 5v4): 44/10/81.48
	

The Buffalo Sabres were atrocious in even strength goal differential.  It was so rare to find a pairing that produced a positive goal differential.  The only pairs of defensemen included in this analysis to do it were Beaulieu and Nelson (2/1 over 58mins but had a gross CF% of 40.4), Beaulieu and Tennyson (5/2 over 63mins and did ok with a CF% of 53.7), and McCabe and Guhle (2/1 over 27mins both left shots, CF% 51.9).  That's only about 148mins of positive goal differential even strength pairs over the entire season.

Note that Scandella/Ristolainen posted marginally better GF% (41.94) than team total (40.34), despite taking most of the hard work.

The small sample sizes make it tough to draw many specific conclusions from, but when all of them are considered together, one things is absolutely clear: the Buffalo Sabres team defense sucked fat .

 

Edited July 10, 2018 by IKnowPhysics