Yep, the brief time he was a Flyer after being acquired at the trade deadline, he played well. For 19 games. This is what is called a small sample size. Much more telling are his numbers with most of the Isles, who are almost all better without MacDonald than with.

To MacDonald's credit, his possession numbers were very solid as a Flyer last season, too. Especially when the Corsi QOC was lower for him.
- Jsaquella

Haha I was being sarcastic. I agree with you -- I was pointing out that very such thing. The year we traded for him, throughout the entire year, both with the Flyers and Isles every single player on the Flyers and Isles except for Adam Hall (d zone 4C specialist) and Eric Boulton (lug fighter extraordinaire) did better apart from AMac than with him. Even in his cushier minutes with the Flyers.

A person more inclined to doubt the stats or who is dismissive of advanced stats in general might say...."But of course the other Isles players did better without him, they had easier minutes apart from him!" And that would be correct. AMac had the highest Relative Quality of Competition among Isles d men that year along with his partner Travis Hamonic. That's where advanced stats get sloppy unless you add more stats in for context. A Corsi W/WO is not some sort of player deciding doctrine. It’s just one pouch in the stats utility belt, something to make you think about.

And here's the stat I find most compelling given that every player did better apart from AMac while taking into context what I just wrote above: Travis Hamonic did much better without AMac. As in 45.5 CF% with.....AMac 43.9% without Hamonic.....Hamonic 52.7% without AMac. Hamonic's QoC never went down nor did his zone starts or the like fluctuate dramatically without. Now THAT'S the stat I like the most inside of that giant number pile link.

I'll talk about AMac this year too if you want, given his change of circumstance to more sheltered minutes.

Haha I was being sarcastic.
- Mononoke

It went right over Jsaq's head:

And here's the stat I find most compelling given that every player did better apart from AMac while taking into context what I just wrote above: Travis Hamonic did much better without AMac. As in 45.5 CF% with.....AMac 43.9% without Hamonic.....Hamonic 52.7% without AMac. Hamonic's QoC never went down nor did his zone starts or the like fluctuate dramatically without. Now THAT'S the stat I like the most inside of that giant number pile link.

I'll talk about AMac this year too if you want, given his change of circumstance to more sheltered minutes.
- Mononoke

When Hamonic was playing without MacDonald, who was he playing with, and how was that different than when he was playing with MacDonald? Also, what were the game situations? Are those numbers from close situations, or all game situations 5 on 5, or what?

I'm not sure what you're watching, but MacDonald is solid with the puck in his own end. He's very adept at buying space and time, and making a good outlet pass in his own end. MacDonald was charged with 25 giveaways last season in 58 games played, which is very good. Now granted giveaways aren't a perfect stat, because it varies from building to building, but that is certainly an indication that he is good with the puck.
- MJL

Problem with giveaways besides what you mentioned is that they can be credited more for the big uh-oh mistake than anything. A lot more goes into a successful transition game than not making big gaffes in your own zone. Simple plays that go uncredited as giveaways. Now in terms of both percentage of successful exits (66.59%) and percentage of successful exits with player controlled possession (56.07), AMac was third worst on the team ahead of only Grossmann and Schultz. We know their weakness as puck movers. AMac is better than them, but better than them isn't necessarily the highest of praise. It's definitely not a ringing endorsement of his own zone puck moving.

Not sure how to feel about this new guy.

Problem with giveaways besides what you mentioned is that they can be credited more for the big uh-oh mistake than anything. A lot more goes into a successful transition game than not making big gaffes in your own zone. Simple plays that go uncredited as giveaways. Now in terms of both percentage of successful exits (66.59%) and percentage of successful exits with player controlled possession (56.07), AMac was third worst on the team ahead of only Grossmann and Schultz. We know their weakness as puck movers. AMac is better than them, but better than them isn't necessarily the highest of praise. It's definitely not a ringing endorsement of his own zone puck moving.
- Mononoke

Do you know what the problem with successful exits and player controlled possession stats is? It's impossible for that stat to pin a successful zone exit, or player controlled possession to one player. It's a very unreliable stat. Hockey is a team game, and a successful zone exit is on more than just the defenseman.

If a defenseman makes a perfect outlet pass to a forward, and the forward mishandles the pass, leading to a turnover, then it's a failed exit. The defenseman should have that go against him statistically. How about if a defenseman is forced by two forecheckers, and has no safe outlet, and his only play is a chip off the glass out to center ice, to avoid a dangerous situation? I say the defenseman made a solid play, and now his team can regroup defensively, and they didn't give up a scoring chance against. There's no player possession there.

Hockey is a team game. All these stats can tell is what happened, now why. Want to know the best way to see how good MacDonald is with the puck? Watch the game. If you do, you'll see he's pretty solid with it, and a good outlet passer. Breakouts is a team game, like anything in hockey.

When Hamonic was playing without MacDonald, who was he playing with, and how was that different than when he was playing with MacDonald? Also, what were the game situations? Are those numbers from close situations, or all game situations 5 on 5, or what?
- MJL

How was that different than playing with MacDonald? Simply put: he wasn't playing with him haha. No, but really Hamonic apart from AMac played the majority of his min with de Haan. Probably most of those minutes are after the trade, some before. de Haan is a good possession player. I'd argue that he helped BOOST Hamonic -- I'm not defending Hamonic. He's pretty good but his Corsi didn't skyrocket just based on his own played but with the help of de Haan. But even WITHOUT de Haan, Hamonic registered a 49.1 cf%, still a significant jump from the 45% he registered with AMac, even if a bit lower than the 51.8% with de Haan. My argument is that AMac made him worse. All these are all game situations.

How was that different than playing with MacDonald? Simply put: he wasn't playing with him haha. No, but really Hamonic apart from AMac played the majority of his min with de Haan. Probably most of those minutes are after the trade, some before. de Haan is a good possession player. I'd argue that he helped BOOST Hamonic -- I'm not defending Hamonic. He's pretty good but his Corsi didn't skyrocket just based on his own played but with the help of de Haan. But even WITHOUT de Haan, Hamonic registered a 49.1 cf%, still a significant jump from the 45% he registered with AMac, even if a bit lower than the 51.8% with de Haan. My argument is that AMac made him worse. All these are all game situations.
- Mononoke

First of all, all game situations are not reliable. Doesn't take score effects into consideration. Secondly, here is the issue with WOWY data that makes it unreliable. Inevitably, when players who are regular line mates are split up, the line mates, game situations, and roles players are put into, are inevitably different. It's not an apples to apples comparison. Like any Corsi based stat, which in itself is unreliable, it can only tell you what happened. It cannot tell you why. Those stats simply are not capable of telling you that MacDonald made him worse. Not possible.

Do you know what the problem with successful exits and player controlled possession stats is? It's impossible for that stat to pin a successful zone exit, or player controlled possession to one player. It's a very unreliable stat. Hockey is a team game, and a successful zone exit is on more than just the defenseman.

If a defenseman makes a perfect outlet pass to a forward, and the forward mishandles the pass, leading to a turnover, then it's a failed exit. The defenseman should have that go against him statistically. How about if a defenseman is forced by two forecheckers, and has no safe outlet, and his only play is a chip off the glass out to center ice, to avoid a dangerous situation? I say the defenseman made a solid play, and now his team can regroup defensively, and they didn't give up a scoring chance against. There's no player possession there.

Hockey is a team game. All these stats can tell is what happened, now why. Want to know the best way to see how good MacDonald is with the puck? Watch the game. If you do, you'll see he's pretty solid with it, and a good outlet passer. Breakouts is a team game, like anything in hockey.
- MJL

Watch the game! How condescending are you, bub? I watch almost EVERY Flyers game, minus about 1-3 a year. I assume you watch almost all too. That means on that we are on an even playing field. But to criticize me for digging deeper into the stats apart from just watching the game....sheesh.

Of course hockey is a team game and for all these stats things must be taken into account like another player fumbling something or causing a giant mistake that reflects on one player's stats. But statistics, given a large enough sample size, tend to erase things like that. Sample size is everything -- over a large enough one, any stat will average out to account for peaks and lows.

Now as to hockey being a team game and so Corsi and the like are flawed because they individualize what is team based (they ain't perfect, but they're still darn good), my response is: any player out on the ice has an effect on the game possession wise. You don't have to touch the puck to affect possession. Positional soundness can do that, etc.. If a player, who while not a great puck mover themselves necessarily, plays a smart but effective game and, say for a d man, helps shut down their area of the ice, THAT will affect possession indirectly. But that's also why we look to other stats for context. Corsi doesn't give context, but with other stats it can be good. Corsi on its own can be tricky and give misleading conclusions. This is why we love Coots -- his Corsi is decent but look at the context and his Corsi suddenly seems like a small miracle!

Not sure how to feel about this new guy.
- flyer_nutter

Watch the game! How condescending are you, bub? I watch almost EVERY Flyers game, minus about 1-3 a year. I assume you watch almost all too. That means on that we are on an even playing field. But to criticize me for digging deeper into the stats apart from just watching the game....sheesh.

Of course hockey is a team game and for all these stats things must be taken into account like another player fumbling something or causing a giant mistake that reflects on one player's stats. But statistics, given a large enough sample size, tend to erase things like that. Sample size is everything -- over a large enough one, any stat will average out to account for peaks and lows.

Now as to hockey being a team game and so Corsi and the like are flawed because they individualize what is team based (they ain't perfect, but they're still darn good), my response is: any player out on the ice has an effect on the game possession wise. You don't have to touch the puck to affect possession. Positional soundness can do that, etc.. If a player, who while not a great puck mover themselves necessarily, plays a smart but effective game and, say for a d man, helps shut down their area of the ice, THAT will affect possession indirectly. But that's also why we look to other stats for context. Corsi doesn't give context, but with other stats it can be good. Corsi on its own can be tricky and give misleading conclusions. This is why we love Coots -- his Corsi is decent but look at the context and his Corsi suddenly seems like a small miracle!
- Mononoke

Thought you said you lurked here for a while...

Welcome, by the way.

Watch the game! How condescending are you, bub? I watch almost EVERY Flyers game, minus about 1-3 a year. I assume you watch almost all too. That means on that we are on an even playing field. But to criticize me for digging deeper into the stats apart from just watching the game....sheesh.


- Mononoke

I wasn't criticizing you there. Just simply stating that the best way to determine how good a player is at something, is to watch the game. That's what really tells you the truth. Not applying team statistics to individual players.

Of course hockey is a team game and for all these stats things must be taken into account like another player fumbling something or causing a giant mistake that reflects on one player's stats. But statistics, given a large enough sample size, tend to erase things like that. Sample size is everything -- over a large enough one, any stat will average out to account for peaks and lows.


- Mononoke

No, they don't tend to erase things that. That's the fallacy of analytics. Hockey is too much of a team sport for these statistics to be applied to rating the play of individual players. They cannot eliminate the effects of teammates, opposition, game situations, scheduling, and tons of other factors that are involved. They're solid team statistics. They can tell you what happened in a game, but they can't tell you why

Now as to hockey being a team game and so Corsi and the like are flawed because they individualize what is team based (they ain't perfect, but they're still darn good), my response is: any player out on the ice has an effect on the game possession wise. You don't have to touch the puck to affect possession. Positional soundness can do that, etc.. If a player, who while not a great puck mover themselves necessarily, plays a smart but effective game and, say for a d man, helps shut down their area of the ice, THAT will possession indirectly. But that's also why we look to other stats for context. Corsi doesn't give context, but with other stats it can be good. Corsi on its own can be tricky and give misleading conclusions. This is why we love Coots -- his Corsi is decent but look at the context and his Corsi suddenly seems like a small miracle!
- Mononoke

The issue with Corsi is the same issue that it tries to correct, using shot data over goal data with +/-, and that's false data. A player can play his position perfectly, and come off the ice with a negative corsi for a shift, because his teammate screwed up. As much as 40% of corsi data is false data. That's a lot. Now if they could find a way to improve the data collection, Corsi could in theory, be a good stat for looking at individual players. It's a solid team stat, and for giving insight into what happened in a game, but unreliable in telling you why it happened. Again for that, actually best to watch the game. To be clear, that isn't in any way meant towards you as an individual.

Thought you said you lurked here for a while...

Welcome, by the way.
- wolfhounds

Thanks haha. I'm much more laid back usually and I don't feel like writing novel length stat posts every post (ugh), but I've come armed to the teeth for this debate. I'm well aware of his, shall we say, style of posting?

Id love to know what shelleys corsi was. Bet it doesn't tell the whole story

Thanks haha. I'm much more laid back usually and I don't feel like writing novel length stat posts every post (ugh), but I've come armed to the teeth for this debate. I'm well aware of his, shall we say, style of posting?
- Mononoke

You just signed up yesterday, how aware of the posters on this site can you be?

Thanks haha. I'm much more laid back usually and I don't feel like writing novel length stat posts every post (ugh), but I've come armed to the teeth for this debate. I'm well aware of his, shall we say, style of posting?
- Mononoke

Good luck with that.

You just signed up yesterday, how aware of the posters on this site can you be?
- MJL

Just signed up yesterday, but I've been reading Bill's blog religiously most of this past season and now into the offseason. I also page through the forum a lot. I don't have to post to read.....so I'm very well aware! Haha I know you guys a lot better than you know me

Just signed up yesterday, but I've been reading Bill's blog religiously most of this past season and now into the offseason. I also page through the forum a lot. I don't have to post to read.....so I'm very well aware! Haha I know you guys a lot better than you know me
- Mononoke

Well, I'd be happy to continue a conversation on this with you, but if there is an agenda behind it, more than just about talking hockey, I'm not interested.

Just signed up yesterday, but I've been reading Bill's blog religiously most of this past season and now into the offseason. I also page through the forum a lot. I don't have to post to read.....so I'm very well aware! Haha I know you guys a lot better than you know me
- Mononoke

Tell me about me

Thanks haha. I'm much more laid back usually and I don't feel like writing novel length stat posts every post (ugh), but I've come armed to the teeth for this debate. I'm well aware of his, shall we say, style of posting?
- Mononoke

exlund, hello.

You just signed up yesterday, how aware of the posters on this site can you be?
- MJL

He said he's been lurking, and that AMac post was a showdown challenge for you. Draw!

No, they don't tend to erase things that. That's the fallacy of analytics. Hockey is too much of a team sport for these statistics to be applied to rating the play of individual players. They cannot eliminate the effects of teammates, opposition, game situations, scheduling, and tons of other factors that are involved. They're solid team statistics. They can tell you what happened in a game, but they can't tell you why
- MJL

How then do you rate individual players, specifically defenseman?

Tell me about me
- PhillySportsGuy

I like your posts. You have a real level head and are willing to consider advanced stats. Now will you go to the prom with me? :p

I like your posts. You have a real level head and are willing to consider advanced stats. Now will you go to the prom with me? :p
- Mononoke

I have a room reserved already friend

Just signed up yesterday, but I've been reading Bill's blog religiously most of this past season and now into the offseason. I also page through the forum a lot. I don't have to post to read.....so I'm very well aware! Haha I know you guys a lot better than you know me
- Mononoke

Well, we know you're a fan of anime as well as hockey...so that's something.