Confirmed with Link: [NYI/PHI] Andrew MacDonald for Matt Mangene, 2015 2nd and 2014 3rd round picks

[JDinkalage Morgoone]

AMac teared apart statswise in this little write up: http://www.broadstreethockey.com/2014/3/27/5553356/a-10-game-review-of-andrew-macdonald

I have honestly thought he's been a great addition. Am I alone here?

 

[Harold Bingo]

I have honestly thought he's been a great addition. Am I alone here?

I've liked him, too. But, what the hell do I know?

 

[Jtown]

I can definitely find a several videos on just about every dman that gets beat.

There is no doubt about that, but it's more about how he gets beat.

 

[sa cyred]

I have honestly thought he's been a great addition. Am I alone here?

I thought so too but what is in the article is correct. I am not the biggest fan of BSH (due to them loving a certain player and trying to convince people who dont know statistics that what they post is always right). Their Matt Carle love fest also was crazy (come on they put him on the level of Suter and Chara for crying out loud). But in the end their analysis of MacD is correct. One can't ignore these stats. Small sample size but overall statistically he has showed to struggle defensively.

 

[Funf]

To be fair, WOWY stats are designed to mitigate these effects. Those numbers track the difference in corsi when on the ice with specific players. When you look at each player's WOWY stats, you can get a general idea of who is carrying their weight.


That being said, at 5v5, MacDonald plays the bulk of his minutes with Simmonds, Schenn, and Lecavalier amidst one of their weakest 5v5 stretches of the year.

On top of that, I still hate using corsi for players like LSchenn and MacDonald. These are players who defend by blocking shots, and corsi classifies blocking shots negatively. I'd be more interested in their fenwick stats, but I'm too lazy to look it up.


Advanced stats are useful, but they are not gospel. There is a lot they don't account for. Sure, he may be a negative corsi player, but that doesn't account for the weight of his good and bad plays. A lot of what MacDonald does well, he does better than anyone else on the roster. For example, his ability to retrieve the puck in the D zone and navigate out of trouble is excellent. That's been a phenomenal improvement on our D core, and advanced stats don't shed light on it. Though their proponents will tell you that all these skills are inherently included in their stats. I'm not convinced. Especially when you're playing with weak 5v5 forwards.


EDIT: I do agree with one thing in that article if nothing else, Gustafsson deserves to be in the lineup.

Just checked extraskater.com, his Fenwick For/60 at 5 on 5 is at 3.8 for the year. The top pairing, Norris caliber guys seem to be hanging around 8 or so. Matt Carle, who I feel like MacDonald is sort of a replacement for is at 5.5. Unfortunately, extra skater only shows his season with the Islanders and Flyers combined. I can't find numbers for his time spent in Philly, alone.

Grossmann's FF on the year is 5.4... A career high.

I'm interested in advanced statistics, and I read about them a bit. I think they're getting more useful as time goes on, but in this case, I think it shows that there's still a lot of room to grow. MacDonald's puck retrieval skills, and his first pass, might be the best on our team right now. He's not an all star, but to say he's detrimental to the team is absurd. There's a reason that players like MacDonald are getting overpaid as soon as they hit the open market.

As far as the "highlight" video goes, you can do that to every player in the league. Hockey is a game of mistakes, and every player commits them. Every single goal that's scored in the league, is a result of a mistake, somewhere by somebody. It's not a travesty that a 2nd pairing defensemen could get turned inside out by Patrick Kane or JVR - those guys are big time scorers for a reason - and it shouldn't be MacDonald's job to shut them down. Let him do what he does well, and it should pay dividends.

Edit: Just saw this comment on BSH...

Grrrreaaaat article. AMac is exactly who he was projected to be. He’s just plain bad. We had a chance to go after someone that would have been an improvement over Gus, and we got AMac instead. For every good move Homer and his staff make, there are deals like this one where you want to put a fork in your eye.

Anyone who can’t see that from his play thus far is just trying to blow sunshine, and I understand the desire to look at the bright side and not be a Negadelphian, but the scary part of this, is that Berube is so smart, he’s sheltering AMac effectively to the point that he actually looks like he was a good acquisition to some people, and I applaud you for disabusing them of that notion before it continues to get out of hand.

I don't get it... Because MacD isn't playing 30 minutes a night on the top pairing, Berube is 'sheltering' him and that MacD is actually a terrible player?

Every player in the league has a role. Not everyone is a franchise defensemen, or a franchise forward. It's like criticizing a player for putting up the bulk of his points on the PP. Yeah - he might be better on the PP than at even strength, but if he can put up more points there than anybody else, then why not let him play there? Like I said - everyone has a place where they're most effective.

 

[Beef Invictus]

At least he can pass...unlike Grossmann who struggles defensively, and seems to have replaced "passing" with "blindly flailing the puck towards the neutral zone."

 

[Jtown]

This thread really shows how our current biases rarely change even though we may be faced with facts that tell us we are wrong. People are not very receptive to anything new and this thread pretty much sums this up.


Advanced stats have told us macdonald is pretty bad for some time now. We blame it on the team. Then he becomes a flyer , then we blame the bad stats on the absurdity of advanced stats. Then we are presented with clips of the players faults and we come up with ways to justify it such as " you could do that with any player" .

for those who have read "thinking fast and slow" this pretty much personifies our inability to escape our preconceived biases.

 

[Random Forest]

Honestly, the difference between Grossmann and MacDonald is everything you need to shut up someone who is making a point solely based on advanced stats. Grossmann might be a better "possession" player (lol) as far as the stats go, but based on the utility MacDonald provides with his first pass and ability to maneuver the puck out of trouble, there is absolutely no doubt who the more valuable player is to this team.

I'm not saying advanced stats aren't useful. That couldn't be further from the truth; I'm a huge supporter of them. But they can not be used exclusively to reach a conclusion (eg, GKJ after MacDonald's second game or the infamous Carle=Suter). If you're totally throwing your observations out the window, you're conclusion likely won't be any more accurate (unless you're a terrible judge of talent, of course).

MacDonald certainly has utility. Anyone who has watched him on the Flyers knows he does. It's about measuring this utility against his flaws that is the challenge. I don't think I'd re-sign him above 4.75m based on what I've seen, but it doesn't mean he isn't a useful player to have around.

 

[Random Forest]

This thread really shows how our current biases rarely change even though we may be faced with facts that tell us we are wrong. People are not very receptive to anything new and this thread pretty much sums this up.


Advanced stats have told us macdonald is pretty bad for some time now. We blame it on the team. Then he becomes a flyer , then we blame the bad stats on the absurdity of advanced stats. Then we are presented with clips of the players faults and we come up with ways to justify it such as " you could do that with any player" .

for those who have read "thinking fast and slow" this pretty much personifies our inability to escape our preconceived biases.

You're placing too much weight into advanced stats.

They are facts, yes, but they don't always prove the conclusions we think they do. Correlation/causation, my friend.


For the record, I've read Thinking Fast and Slow. Phenomenal book if anyone's interested in learning about how flawed we can be as decision makers.

 

[Protest]

I have honestly thought he's been a great addition. Am I alone here?

I don't think hes been bad at all really, but I'll have to look at the article. I can't see the charts/videos at work. Right off the bat though, I'm thinking about what his skills are, and who his common linemates are, and how that is going to affect his advanced stats numbers.

His biggest skills revolve around just getting the puck out of the zone, and I think he generally has played with our worst 5v5 line. So his skill set doesn't generate shots, it generates transitions, but his linemates are guys that hamper transition and don't generate shots. So it probably doesn't look too good for him.

I'm not a huge fan or Corsi numbers, or the ___For % numbers in general because I think they're easily skewed by systems and teammates.

One thing is for sure though, and that is 10 games tells you absolutely nothing either way.

From 2008-2013 he had the lowest GA/60 for anyone that played over 2,500 minutes of 5v5 hockey for the Islanders. I feel like that may be more relevant than a 10 game sample, even though GA/60 isn't the best stat.

 

[Funf]

At least he can pass...unlike Grossmann who struggles defensively, and seems to have replaced "passing" with "blindly flailing the puck towards the neutral zone."

My favorite Grossman play was last night, I guess it was the 3rd, on the PK, Couturier grabs the puck on the half wall and reverses it down low to Grossman for an easier clear. Instead of clearing up the middle, or reversing to Coburn for an even easier clear, he just eats it in the corner, gets outnumbered by Rangers and we lose possession.

I don't think Grossman is as useless as a lot of people do. According to Pierre he showed up to the game wearing a full boot, so I'd imagine he's not 100% healthy right now either. But my god... He can't do anything with the puck. It's so frustrating.

 

[Protest]

This thread really shows how our current biases rarely change even though we may be faced with facts that tell us we are wrong. People are not very receptive to anything new and this thread pretty much sums this up.

WAR is a fact too right...and what BABIP tells you about a pitcher is a fact too, right? What about UZR? Is that a fact? These numbers are from a much more static game than hockey.

The biggest problem with the advanced statistics community is that there are a group of people who think these stats up, and hordes of other people that eat them up as fact instead of questioning them. Any new stat should be found faulty until proven otherwise.

 

[chimrichalds18]

I don't mind MacD for what he is, but in no way should we put out a defense that has MacD, Grossmann, LSchenn as a long-term plan.

 

[orangecrush8]

I don't mind MacD for what he is, but in no way should we put out a defense that has MacD, Grossmann, LSchenn as a long-term plan.

100% agree. I'm not cool with having 4 (when you include Coburn) dmen who provide little to no offensive output.

I like MacD, I just think his money could be used far more efficiently. I know Niskanen will cost more, but I'd rather give him 5.5 than MacD 4.5-5

 

[Roo Mad Bro]

Flyers are going to re-sign him which is the wrong decision.

 

[dats81]

Advanced stats can be misleading:

MacDonald played on a crappy team for all but 10 games this season while being forced into a top pairing role. Not exactly perfect conditions to succeed.

I have to admit that I don't even understand all of those fancy stats and how they correlate and stuff but as far as his short tenure with the Flyers is concerned I guess small sample size, making the transition to a new system and him not really getting involved that much in the offensive zone so far play into this and make him look worse in advanced stats.

He looks pretty solid overall to the average non-expert hockey fan though. Should they resign him? I tend to say yes but not at all costs...

 

[JDinkalage Morgoone]

I don't know how anyone can take an advanced statistic by itself as gospel when the same advanced statistic said, as mentioned here, that Carle = Suter. That's insane.

 

[Jtown]

Flyers are going to re-sign him which is the wrong decision.

I am beginning to lean that way as well.

 

[GKJ]

Advanced stats can be misleading:

MacDonald played on a crappy team for all but 10 games this season while being forced into a top pairing role. Not exactly perfect conditions to succeed.

I have to admit that I don't even understand all of those fancy stats and how they correlate and stuff but as far as his short tenure with the Flyers is concerned I guess small sample size, making the transition to a new system and him not really getting involved that much in the offensive zone so far play into this and make him look worse in advanced stats.

He looks pretty solid overall to the average non-expert hockey fan though. Should they resign him? I tend to say yes but not at all costs...

http://www.broadstreethockey.com/20...-macdonald-analysis-usage-flyers-craig-berube

If 10 games is anything, "playing on a crappy Islanders team" is only a thing for people who don't want to accept that he was one of the reasons the Islanders were a crappy team. There's plenty of evidence out there, some of it in this thread, showing that.

Re-signing MacDonald wouldn't be a terrible decision as it would be how much they'd have to pay him. They've already given up the assets, so keeping him doesn't bother me, but it's the type of contract that will bother me. No one knows what Timonen is going to do yet, and it may not be until well into the summer before anyone knows, so the Flyers are going to feel like they are in a position where they need to make sure MacDonald is still here. I'd be fine with MacDonald being kept around if it meant Grossmann and/or Luke Schenn was gone.

 

[flyershockey]

I wouldn't mind them resigning him as long as they somehow move on Grossmann. They're going to have to open up some spots long term for the young guys sooner or later. Ideally, you get him to sign for two years. Realistically, he'll never accept that. That's the problem.

 

[Protest]

When talking about advanced stats, I always enjoy this exercise.

Player A vs Player B

Player A:
CF Rel: 7.4%
CF%: 58.1%
GF% Rel: 11.7%
GF%: 67.7%
Qoc: 28.7%
QoT: 29.3%

Player B
CF Rel: 6.4%
CF%: 52.6%
GF% Rel: 6.4%
GF%: 55.5%
Qoc: 30.0%
Qot: 30.4%

So both good players. Player A clearly has better numbers, but Player B plays against harder competion. Though Player B has better teammates that offset that somewhat.

Player A: Vladimir Tarasenko
Player B: Sidney Crosby

You can use these stats to say a lot of things, but it doesn't mean it's accurate.

 

[FLYguy3911]

I trust advanced stats in baseball, starting to trust them in basketball, but don't quite trust them in hockey. They are good stats to know, they have a purpose, but I think hockey is still a game that comes down to the eye test much like football.

 

[Beef Invictus]

When talking about advanced stats, I always enjoy this exercise.

Player A vs Player B

Player A:
CF Rel: 7.4%
CF%: 58.1%
GF% Rel: 11.7%
GF%: 67.7%
Qoc: 28.7%
QoT: 29.3%

Player B
CF Rel: 6.4%
CF%: 52.6%
GF% Rel: 6.4%
GF%: 55.5%
Qoc: 30.0%
Qot: 30.4%

So both good players. Player A clearly has better numbers, but Player B plays against harder competion. Though Player B has better teammates that offset that somewhat.

Player A: Vladimir Tarasenko
Player B: Sidney Crosby

You can use these stats to say a lot of things, but it doesn't mean it's accurate.

All you're really showing is that if you cherry pick some stats and leave others out (like say, actual production), you can make something look better or worse than it is. And on top of that, those stats lack all context. Nobody argues that stats without context are useful.

 

[Striiker]

I'm not part of the advanced stats crowd, I don't believe that those numbers tell the whole story so I'm not sure why so many people do.. they've been proven to be flawed so often and yet people will still trust that over a simple eye test that immediately proves them wrong.

As far as Amac goes, I'm on the fence. He's been doing very well as far as moving the puck out of the zone goes. I'd probably say he's been the best on the team in that aspect of the game. But I'm not sure if that's enough to resign him for 4.5m+ for long term, especially since w have a couple of young Dmen who are coming up. On one side he makes LSchenn play better, but so did Gus (who's also good at moving the puck and provides more offense) and Gus will definitely be cheaper.

 

[Random Forest]

All you're really showing is that if you cherry pick some stats and leave others out (like say, actual production), you can make something look better or worse than it is. And on top of that, those stats lack all context. Nobody argues that stats without context are useful.

Right, but that's kind of the point. Advanced stats are excellent for indicating things and pointing you in the right direction, but the problem is that some people think they give you everything you need to form an accurate conclusion. Well, that's not exactly true. We don't know how to quantify everything yet. Using advanced stats exclusively is problematic and leads you to drawing false conclusions (eg, that Grossmann has been better than MacDonald since the trade deadline-- a claim we know is false).

I'm as big an advocate of advanced stats as their is at this point, but I don't respect when people use them exclusively to make comparisons between players.

On a macro scale (team to team), shot deficit/corsi is the best indicator of future performance we have, and it's pretty good, but it is still lacking, and it is especially lacking on a micro (player to player) scale. One can be better than another from a possession stand point, and still be objectively less useful to the team. To ignore all that MacDonald does well because he has bad corsi figures is narrow minded and gives more credence to advanced stats than they deserve at this point.