Sometimes +/- does tell the tale

[GreenLantern2814]

Originally Posted by gilligans_off

Opinion? Sounded more like a complaint to me. If you look hard enough for a stat to show how bad certain players are you can find it.

Some people just want to bi*ch about the flames regardless of what they do on the ice. If your not happy with the team, stop buying tickets... if you do.

I actually didn't look all that hard at all. That was the point. I just sorted +/- from highest to lowest. And the four lowest guys on the list are the four guys that could be replaced with the smallest amount of fan outcry.

[heep223]

Originally Posted by gilligans_off

Opinion? Sounded more like a complaint to me. If you look hard enough for a stat to show how bad certain players are you can find it.

Some people just want to bi*ch about the flames regardless of what they do on the ice. If your not happy with the team, stop buying tickets... if you do.

You must see the irony in this post, given that you're the one whining and OP is not.

[kevman]

What I never understood is how +/- (goal differential while on the ice) is thrown out as being completely useless but CORSI (shot attempt differential while on the ice) is the next big thing.

Since when did shot attempts become more important than actual goals?

[Enoch Root]

Originally Posted by kevman

What I never understood is how +/- (goal differential while on the ice) is thrown out as being completely useless but CORSI (shot attempt differential while on the ice) is the next big thing.

Since when did shot attempts become more important than actual goals?

Exactly. Essentially, you have 4 variations of the same type of stat:

1) +/-, which tracks goal differential while a player is on the ice
2) shot differential, self explanantory
3) Fenwick, which tracks (shots + missed shots) differential, and
4) Corsi, which tracks (shots + missed shots + blocked shots) differential

Of those, +/- is the most important, IMO, because it focuses on goals, which - accurate or not - at least focuses on the most important events of the game. But Corsi, which is the most inclusive, and thereby the most convoluted, is considered somehow to be some insightful breakthrough. When in reality, it is simply a more inclusive +/-.

[GioforPM]

Originally Posted by GreenLantern2814

I actually didn't look all that hard at all. That was the point. I just sorted +/- from highest to lowest. And the four lowest guys on the list are the four guys that could be replaced with the smallest amount of fan outcry.

Funny though, if you'd done your exercise before last game, you'd have found Raymond at only -2.

[Fan in Exile]

Originally Posted by GioforPM

Funny though, if you'd done your exercise before last game, you'd have found Raymond at only -2.

That's the result of Hartley giving Raymond more ice time. Watching him, it's not a big surprise the Flames don't score many goals when he's on the ice (including the powerplay where he is a mainstay for some reason), nor is it surprising that he is not a defensive stalwart.

[Weitz]

Originally Posted by Enoch Root

Exactly. Essentially, you have 4 variations of the same type of stat:

1) +/-, which tracks goal differential while a player is on the ice
2) shot differential, self explanantory
3) Fenwick, which tracks (shots + missed shots) differential, and
4) Corsi, which tracks (shots + missed shots + blocked shots) differential

Of those, +/- is the most important, IMO, because it focuses on goals, which - accurate or not - at least focuses on the most important events of the game. But Corsi, which is the most inclusive, and thereby the most convoluted, is considered somehow to be some insightful breakthrough. When in reality, it is simply a more inclusive +/-.

The issue with +/- is that it is only based on goals and goals are a rare occurrence. A goal scored right when you step on the ice has a dramatic effect on your +/- where as 1 shot attempt against does not. Corsi takes the sudden and unexpected goal because someone made a bad change away from judging your play.

[Enoch Root]

Originally Posted by Weitz

The issue with +/- is that it is only based on goals and goals are a rare occurrence. A goal scored right when you step on the ice has a dramatic effect on your +/- where as 1 shot attempt against does not. Corsi takes the sudden and unexpected goal because someone made a bad change away from judging your play.

Obviously.

But goals as you're stepping on or off the ice balance out over time, just as shots from the perimeter vs quality shots balance out.

That is the nature of stats.

[Fan in Exile]

Originally Posted by Weitz

The issue with +/- is that it is only based on goals and goals are a rare occurrence. A goal scored right when you step on the ice has a dramatic effect on your +/- where as 1 shot attempt against does not. Corsi takes the sudden and unexpected goal because someone made a bad change away from judging your play.

Yes, but when it is pointed out that Corsi does not necessarily correspond to results (i.e. disproportionate or "unsustainable" PDOs), the answer is that shooting percentage come back to the average so it is actually a reliable predictable for goals.

Really, the point is that stats are indicators not predictors and I think Green Lantern's summary of +/- is an accurate interpretation of what is actually happening (or not happening) with certain players on the ice.

But if you have Corsi stats that point to a different conclusion, I'd be interested to see them. It seems to me that Corsi and +/- should roughly correspond at least in relative terms when looking at a team although strength of competition and roles can cause fluctuations.

[GioforPM]

Originally Posted by Fan in Exile

That's the result of Hartley giving Raymond more ice time. Watching him, it's not a big surprise the Flames don't score many goals when he's on the ice (including the powerplay where he is a mainstay for some reason), nor is it surprising that he is not a defensive stalwart.

He got 45 or so more seconds than the previous game where he was even, and was about an average amount of time compared to the rest of the season. His larger minus right now, is pretty much the product of the Anaheim and Florida games, both of which were defensive gong shows for the whole team, and not the best goaltending performance. The Anaheim game had minus performances from Gio, Brodie, Gaudreau, Jooris, Hudler, etc. In fact, Raymond's line was even, so it seems to me that Raymond may have suffered a bit in that game from stepping on the ice at a bad time. I didn't notice any glaring errors leading to goals by him.

I'm not a big fan of any of the 4 players you identified. But I think it's a flawed analysis, because the plus minus is so skimpy at this point, that a single bad game skews the result.

[GreenLantern2814]

Originally Posted by GioforPM

He got 45 or so more seconds than the previous game where he was even, and was about an average amount of time compared to the rest of the season. His larger minus right now, is pretty much the product of the Anaheim and Florida games, both of which were defensive gong shows for the whole team, and not the best goaltending performance. The Anaheim game had minus performances from Gio, Brodie, Gaudreau, Jooris, Hudler, etc. In fact, Raymond's line was even, so it seems to me that Raymond may have suffered a bit in that game from stepping on the ice at a bad time. I didn't notice any glaring errors leading to goals by him.

I'm not a big fan of any of the 4 players you identified. But I think it's a flawed analysis, because the plus minus is so skimpy at this point, that a single bad game skews the result.

If it's raining outside, how many metrics do you need to tell you it's wet?

The stat is honestly beside the point - I just thought it was interesting that such an old school, much maligned statistic could so readily confirm what we all know is the case.

[hackamore]

Originally Posted by GioforPM

He got 45 or so more seconds than the previous game where he was even, and was about an average amount of time compared to the rest of the season. His larger minus right now, is pretty much the product of the Anaheim and Florida games, both of which were defensive gong shows for the whole team, and not the best goaltending performance. The Anaheim game had minus performances from Gio, Brodie, Gaudreau, Jooris, Hudler, etc. In fact, Raymond's line was even, so it seems to me that Raymond may have suffered a bit in that game from stepping on the ice at a bad time. I didn't notice any glaring errors leading to goals by him.

I'm not a big fan of any of the 4 players you identified. But I think it's a flawed analysis, because the plus minus is so skimpy at this point, that a single bad game skews the result.

Last year was only year Raymond was a minus player, and like this year that happened in a couple of games at end of year when whole team went in toilet. The year Kesler won the Selkie Raymond was most common linemate so while he befitted from Kesler's play he also helped Kesler with his defensive play. Only the Sedin's had better possession numbers. Fighting injuries and new system, but he could be very useful player if he finds his game again as he was before messing up shoulder. He will never be the shiny new toy compared to the young guys coming up.

[GioforPM]

Originally Posted by GreenLantern2814

If it's raining outside, how many metrics do you need to tell you it's wet?

The stat is honestly beside the point - I just thought it was interesting that such an old school, much maligned statistic could so readily confirm what we all know is the case.

Stopped clocks and all that. BTW, you left out Seto, Byron at -5 and Sven at -4 (on pace for a huge negative number at the time he was sent down). You also missed that Glencross, Colborne, Stajan and Jones, who get a lot of grief on this board, are all pluses.

[gilligans_off]

Originally Posted by heep223

You must see the irony in this post, given that you're the one whining and OP is not.

Save it pal. No need for you to come in and pick a fight. I'm not whining at all. I'm defending myself.

[DoubleK]

Originally Posted by GreenLantern2814

It isn't a thinly veiled anything - Smid and Engelland aren't good enough. I don't need +/- to tell me that, and Neither do you.

+/- confirms the four weakest links on the team. They're exactly who we thought they were.

So you are willing to give up a couple 2nd round pics for depth defenseman to improve this now? That's the going rate for NHL dmen.

[Psytic]

Originally Posted by DoubleK

So you are willing to give up a couple 2nd round pics for depth defenseman to improve this now? That's the going rate for NHL dmen.

I think start packing up a few vets for some picks at the deadline and trade them to move up and get a higher rated D prospect come draft day. I figure we are stuck with a crappy bottom D until oir prospects mature unless we can pull off something like the islanders did.

[heep223]

Originally Posted by gilligans_off

Save it pal. No need for you to come in and pick a fight. I'm not whining at all. I'm defending myself.

Cool story bro!

[gilligans_off]

You're a waste lol

+/- is useless

I'm done fighting with children.

[GioforPM]

Originally Posted by Psytic

I think start packing up a few vets for some picks at the deadline and trade them to move up and get a higher rated D prospect come draft day. I figure we are stuck with a crappy bottom D until oir prospects mature unless we can pull off something like the islanders did.

The defencemen in this draft are pretty close, right from 4 though 17 or so, and I think we would be wasting assets to move up (not that I think anyone's trading a higher first for vets anyway).

I think if you are trading for defence, you are better of trading for prospects rather than picks.

[Hackey]

Never been a big fan of the stat. I think more often it tells more about the team than the player. There are so many factors that play into +/- that can be out of a players control. Hockey is to much of a team game for me to make any difinitive conclusions using it.