The Cap is going to go up, and so is player salary. And MacDonald's deal will be cheap in a few short years. Look at Giroux's deal, oh my gosh, 8.275M a year! Well Kane and Toews just re-signed for 10.5M. It's already happening, and is going to continue to happen.
- MJL

Which is why the owners would never do contracts as a percentage of the cap, no matter how much I want them to do so

I look at the Evander Kane situation similar to a disgruntled employee. I think alot of us lose sight of the fact that it is a job for these guys. So imagine going to place you didn't enjoy being at.

Your work suffers. Your attitude changes. Maybe you don't like your boss. Or the company itself isn't successful. It doesn't take away that you are a good worker but you need a change of scenery and a fresh start so you can be happy.

Same rules apply to athletes. I think Kane just needs to play somewhere he can win and be happy. I was sort of hoping the rumors about Philly being interested in him were true. I would love to see Kane at LW with G and Voracek. He would easily get 35 or more goals.
- Pelle31Forever

I agree, and could very well be the case. Maybe he's just simply not happy there, and wants out. I have no problem with that. I know what that can do for you. And money doesn't cure all ills.

Which is why the owners would never do contracts as a percentage of the cap, no matter how much I want them to do so
- nails

Well the NHL max salary is based on a percentage of the Cap.

Well the NHL max salary is based on a percentage of the Cap.
- MJL

Yup
The only thing that having a percentage contract helps is when the caps goes up a little or down.
The cap savings received from cap jumps but stable contract figures is what let's the bug teams keep getting better.
It will also likely keep the poor teams never getting better

What I learned back reading the last 4 pages

Only in hockey is 3/4 > 4/5, or maybe it's equal.
Or we can just wipe out the 4's and say that 3>5, which is finally true.
except in Hockey where sometimes, 3=5
- nails

King Arthur: Right. One... two... five!

Galahad: Three, sir.

King Arthur: Three!

Yup
The only thing that having a percentage contract helps is when the caps goes up a little or down.
The cap savings received from cap jumps but stable contract figures is what let's the bug teams keep getting better.
It will also likely keep the poor teams never getting better
- nails

Well that's why I've always said that one of the reasons given for having a Cap, is that it creates parity, is misleading. Not sure that is does. But it does help to keep player salaries down, but is not really effective at that either. Top players are going to continue to get paid.

There's some nice insults you put in there. I'll skip returning in kind, I don't need to go there. I'll stick to the topic at hand.

The same smart people who are working to refine and expand our rational understanding of the sport, is who I formed my opinion on advanced stats from. And what they've wrote about them, and what the flaws are, and how the correct context needs to be used. I've studied Corsi extensively. I didn't just wake up and decide to have a crusade.
And secondly, my biggest issues with Corsi and using it to rate individual players, is a sound opinion, that can be backed up with reference material. And I've yet to see any of the Corsi proponents of this thread, prove that the issue doesn't exist, and why it's not a flaw of Corsi. It's just tapped danced around as if it doesn't exist.

Let's look at a Hockey pioneer in Roger Neilson. Who used video to correct the flaws of the Roth/Irvin +/- system. He actually reviewed game tape, and removed the false data, and only awarded a +/- to players that earned it or deserved it. And it was a far more accurate system.
Imaging if that could be done with Corsi? Remove the false data. Then we'd have something far more viable and reliable. Corsi is a good team stat, because the false data doesn't matter in that context.
- MJL

You are incorrigible. Some of us think that his stat issues come from things we can see in his game. We, in our own viewings, are removing the false data.

You are incorrigible. Some of us think that his stat issues come from things we can see in his game. We ourselves, in our own viewings, are removing the false data.
- aflyerpower8

If you can't attack the message, attack the messenger. It's impossible to remove the false data from the stat by watching the game. Issues such as conformation bias come into play.

Here is my issue with Corsi when rating individual players.

Let’s say a player is out on the ice for 30 shots at net for in a game and 20 shots at net against. The player is a defensive defenceman and he only helped to create 5 of those shots at net, while he made mistakes that led to 15 shots at net against. His real impact on the game is -10 Corsi, but the mechanical system now being used would give him a +10 Corsi plus-minus. He got that positive mark only because he was on the ice with better players who created a positive but unearned differential for him. This same dynamic can happen game after game, month after month, giving a player an unfair and inaccurate Corsi rating over a full season.
If a player is out for many, many shots-at-net for, he’s spending a lot of time in the opposition end of the ice, which is a very good place for a player to be in a hockey game. If he’s out for a lot of shots against, he’s in his own end a lot, not such a good place to be. If his team regularly has 15 shots-at-net when he’s on the ice during a game, and gives up just 10 shots-at-net against, something is definitely going right when that player is on the ice. And because a player is out on the ice for so many more shots-at-net than goals during a full season of play, Corsi takes much of the luck out of the equation for measuring players.

That’s good. That makes Corsi useful, in theory.

However, it still has the same major problem as the Roth-Irvin system in one other regard: since every player on the ice gets a plus or a minus after each shot-at-net, Corsi awards plus marks to players even if they did nothing to help create the shot-at net, and the system assigns minus marks to players who made no mistake on the shot-at-net against. The current system assigns these unearned marks at a high rate, likely around 40 per cent of the time.

That is my incorrigible stance against Corsi and using it to rate individual players. Now I challenge you to refute it!

I would say no state tax & that would mean they keep more of the $$$ they sign for in a deal.
- ob18

You make me sick when you speak, Maurice

If you can't attack the message, attack the messenger. It's impossible to remove the false data from the stat by watching the game. Issues such as conformation bias come into play.

I cut down your quote because it was quite long.

That is my incorrigible stance against Corsi and using it to rate individual players. Now I challenge you to refute it!
- MJL

I don't use Corsi on a game by game basis. It is large sample sizes where trends begin to occur. You take these trends and see if they have any merit by watching the games. Once I have done this, I form my opinion.

I do not praise Corsi as a perfect stat, but I do not agree with it being a useless stat like you seem to think.

You make me sick when you speak, Maurice
- KINGKENZO

Okay

I don't use Corsi on a game by game basis. It is large sample sizes where trends begin to occur. You take these trends and see if they have any merit by watching the games. Once I have done this, I form my opinion.

I do not praise Corsi as a perfect stat, but I do not agree with it being a useless stat like you seem to think.
- aflyerpower8

The trends are not accurate due to the high rate of false data involved. The larger the sample of games, the larger the amount of false data. And the percentage difference in players is not wide enough to account for that margin of error. On a game by game basis, Corsi can give insight into what happened, but not why it happened.

I don't think Corsi is a useless stat. I think Corsi is good in theory. However the data collected to come up the with numbers is flawed, due to the high amount of false data, as was explained previously with examples. I think Corsi is a much better stat when looked at team wise, because the issue of false data is then removed. I wouldn't have that opinion if I thought it was useless.
I don't want to be accused of being incorrigible again, but nothing was offered to refute the issue of the stat being flawed due to false data when looking at individual players.

Like I said previously, if a way to remove the false data could be found, I think it would be a pretty good stat to look at individual players.

The bottom line is that it is certainly an improvement over what we had in the past, and may be the best thing there is right now. But it's still highly flawed.

The shot differential trend of a good player on a bad team is not going to change over a large sample size unless the team becomes better over that time period. The very fact that a good player can have his Corsi rating affected by the team and the players around him points to the fatal flaw of Corsi. Just as a mediocre player can have his Corsi rating upgraded due to playing with good players on a good team.
It's the same issue that the +/- system had. Dallas Smith had the 4th highest +/- total in NHL history at +94. Ever heard of him? He was Bobby Orr's defence partner in 1970/71 when Orr was a NHL record +124.

The Cap is going to go up, and so is player salary. And MacDonald's deal will be cheap in a few short years. Look at Giroux's deal, oh my gosh, 8.275M a year! Well Kane and Toews just re-signed for 10.5M. It's already happening, and is going to continue to happen.
- MJL

It's 2007. I'm going to buy this house and flip it in 2 years. I can't afford it, but screw it... it's price is only going to go up and up!

It's 2007. I'm going to buy this house and flip it in 2 years. I can't afford it, but screw it... it's price is only going to go up and up!
- mmcclatchy

Bad analogy. Whether you can afford it or not, that doesn't make the price of the house a bad price.

FYI, ZS have "false data" artifacts in there, too. If a guy ices the puck 4-times on a shift, he's going to get credited with 4 d-zone starts, which may look like the coach sent him out for 4 separate shifts.

The thing is though, multiple icings is a relatively uncommon occurrence. It may skew the numbers in a one game sample, but drag the sample out over a week, month, season and it's just noise that gets filtered out. Over a larger sample, ZS does accurately describe usage quite well.

It's the same with any stat, including Corsi. Unless a guy is ALWAYS out with the same players (which is rare), eventually any data skew and/or noise caused by teammates gets filtered out over a large enough sample and the players own value begins to emerge.

Kevin Hayes bronance with Harry Z
Hes gonna to penguins

FYI, ZS have "false data" artifacts in there, too. If a guy ices the puck 4-times on a shift, he's going to get credited with 4 d-zone starts, which may look like the coach sent him out for 4 separate shifts.

The thing is though, multiple icings is a relatively uncommon occurrence. It may skew the numbers in a one game sample, but drag the sample out over a week, month, season and it's just noise that gets filtered out. Over a larger sample, ZS does accurately describe usage quite well.

It's the same with any stat, including Corsi. Unless a guy is ALWAYS out with the same players (which is rare), eventually any data skew and/or noise caused by teammates gets filtered out over a large enough sample and the players own value begins to emerge.
- Tomahawk

The false data never gets filtered out with Corsi, when the sample size gets larger. A decent player is going to continue to have his corsi rating inflated if he continues to play with a top player or a top line over a large sample of games. The false data is just going to continue to pile on. Same as with a good player on a bad team with weaker players.

And the example of icings is a poor example, because it is as you say a relatively uncommon occurrence. However false shot differential data is not relatively uncommon. In fact it's very common. And a large part of the data collected is false data each and every game. Game after game, after game.
If the false data involved with Corsi was relatively uncommon, it wouldn't be an issue. But it's the opposite. It's very common, skewing the numbers.

Bad analogy. Whether you can afford it or not, that doesn't make the price of the house a bad price.
- MJL

It's not a bad analogy. The cap won't keep going up like it has these past few years. There has been the NBC and Rogers contracts that have pushed the cap up dramatically. A team that's banking on the cap going up 5-7 million every year is going to have problems when it evens out.

And choosing which thing to focus on to discount it is an impeccable rhetorical device. The point was the value of the house in 2007 was inflated. The market set the price but the crash reestablished it's own value the house

Kevin Hayes bronance with Harry Z
Hes gonna to penguins
- Winning

I believe harry the z is an islander

I believe harry the z is an islander
- aightwebang17

Don't ruin this for me

It's not a bad analogy. The cap won't keep going up like it has these past few years. There has been the NBC and Rogers contracts that have pushed the cap up dramatically. A team that's banking on the cap going up 5-7 million every year is going to have problems when it evens out.

And choosing which thing to focus on to discount it is an impeccable rhetorical device. The point was the value of the house in 2007 was inflated. The market set the price but the crash reestablished it's own value the house
- mmcclatchy

Other then the past season when the Cap was artificially lowered, it has gone up every season since the cap was instituted in 2005. The effect of the new TV contracts has not fully kicked in yet.

It's pretty simple. If appraisals and the housing market sets the value of a house at 300K. If you can't afford that house, and that price, that doesn't mean that it's a bad price.

If the Salary Cap crashes, then yea, player contracts that are currently signed are going to get more expensive relatively. But there is no evidence of that happening to this point. And regardless it can only go down 4.7M at the most. And if that does happen, then then the market will adjust itself.

Okay
- ob18

Cmon man, theyre lines from Slapshot

Cmon man, theyre lines from Slapshot
- KINGKENZO

I do not have the movie committed to memory

Other then the past season when the Cap was artificially lowered, it has gone up every season since the cap was instituted in 2005. The effect of the new TV contracts has not fully kicked in yet.

It's pretty simple. If appraisals and the housing market sets the value of a house at 300K. If you can't afford that house, and that price, that doesn't mean that it's a bad price.

If the Salary Cap crashes, then yea, player contracts that are currently signed are going to get more expensive relatively. But there is no evidence of that happening to this point. And regardless it can only go down 4.7M at the most. And if that does happen, then then the market will adjust itself.
- MJL

I see truth in your argument, but can you see truth in mine? I'm not trying to prove you wrong but I do think a broader understanding can be discussed without a need for dueling wits.

I see truth in your argument, but can you see truth in mine? I'm not trying to prove you wrong but I do think a broader understanding can be discussed without a need for dueling wits.
- mmcclatchy

I can only go by the evidence at hand. And from that and all the forecasts are that the cap is going to continue to go up. And as we see every year, player salaries are going to follow. So deals signed previously are going to be cheaper relative to new deals signed. We see that already when looking at Groux's deal when compared to the new deals that Toews and Kane just signed.