During the 2012 June draft, as it came to a close, the one thing that kept daunting me was Did we chose the right center?? I know the player we took was considered as good a two way foward as any in the draft, but when Buffalo chose Grigorenko, I can only hope that we got the better of the two fowards. Yesterdays showing by Grigs, Four Points, is now making alot of Habs fans wondering..

So Grigo gets four points in a blowout regular season game against Chicoutimi and not even getting a star and we have to start doubting our draft selection. Did you see him in the four games in the Can-Russia series? He had Yakupov on his wing and, aside from a couple of good moments, did nothing in the series, was totally invisible and played at a pedestrian pace. Russia's second line did most of the damage offensively for Russia in the series. While I did expect more from Yakupov, he did display some great speed and hands. While I am not saying that Grigo will be bust or anything like that, watching him in the series convinced me even more that taking him with a high pick would represent enormous risk and clearly TT saw that. In fact, if Gally had been taken by Edmonton or Colombus, from what RDS reported, it would've been Terovinen for the Habs. But it seems that Grigo talk just would not go away any time soon.

Grigorenko will probably have an incredible season.

1. He was challenging for the #1 overall the whole season to drop to #13. He's probably mad and wants to prove people wrong

2. Québec will be the best offensive team this season. Stacked forwards and Roy preaches an offensive oriented team.

3. The difference between Gally and Grigo is that Grigorenko is an ace on the PP. Was like that all season long last year. Out of the 9 goals scored, I would not be surprised if there was 4+ goals on the PP.

Different players, different qualities. Gally will be good for a strong 5vs5 play where Grigo will probably get a ******** of points on the PP. I prefer a good 5 vs 5 team but we all saw last year how a bad PP sank our ship.

Edit: I was a huge Grigorenko supporter, and I still think he will do some great things. But, contrary to the general consensus, I think Gally has a higher potential of superstar status because of his explosive/drive to the net game. Grigorenko's game require space to be efficient and PP will provide that. But everyone knows how little room there is with 5 vs 5 hockey in the NHL.

Edit2: Where are the russian haters that said Grigorenko would bolt to the KHL for money as soon as possible? NHL lockout and still playing in the CHL.

During the 2012 June draft, as it came to a close, the one thing that kept daunting me was Did we chose the right center?? I know the player we took was considered as good a two way foward as any in the draft, but when Buffalo chose Grigorenko, I can only hope that we got the better of the two fowards. Yesterdays showing by Grigs, Four Points, is now making alot of Habs fans wondering..

I would actually pay the $179.99 as it works out to only $25 a month which is not that bad considering tickets are $20 a game.. but the fact that you can`t do monthly billing makes me not want too

We should not forget too that Grigorenko plays in a much better team than Galchenyuk this season.

Exactly, Grigo is much better surrounded than Gally, so his pts total should be higher but thats really not a concern to me.. Also, we have to lower our expectations about Gally at the begin of the season..

Grigorenko can get 200 pts in the Q and it won't change his status as a pro prospect in my view. He'll be 2a/1b center who is frustratingly ineffective in big games.

I'd be shocked if Galchenyuk scored 100 points this year.

- He will miss games due to the WJC.
- Yakupov is gone.

90-95 points is more realistic.

95 points is realistic and 100 is shocking?

Here's some statistics 101:

You indirectly listed 92.5 points as the mean most likely outcome for Galchenyuk.

The statistical noise on this kind of number? square root of 92.5, or 9.6 points.

Thus, based on your own assessment, Galchenyuk has a 1 in 6 chance of scoring more than 102 points, and a similar 1 in 6 chance of scoring less than 83 points.

You indirectly listed 92.5 points as the mean most likely outcome for Galchenyuk.

The statistical noise on this kind of number? square root of 92.5, or 9.6 points.

Thus, based on your own assessment, Galchenyuk has a 1 in 6 chance of scoring more than 102 points, and a similar 1 in 6 chance of scoring less than 83 points.

In reality, statistical noise is even higher.

i did my stats 221, in economics, thought it's been a while, i have hard time to remember what you are talking about. Seems a bit too simple to be true.

By statistical noise, do you mean the standard deviation? that would be square root of the variance, and don't see how u calculated the variance here. As to the mean, it has not much to do with the variance if i remember correctly.

2nd time he uses this example; not only it seems to be false, but he's trying to put down people with his "statistics 101", which would be quite LOL

Points can be approximated as a poisson process, so the noise goes as square root of N.

From wikipedia:

Quote:

In probability theory, a Poisson process is a stochastic process which counts the number of events[note 1] and the time that these events occur in a given time interval. The time between each pair of consecutive events has an exponential distribution with parameter λ and each of these inter-arrival times is assumed to be independent of other inter-arrival times. The process is named after the French mathematician Siméon-Denis Poisson and is a good model of radioactive decay,[1] telephone calls[2] and requests for a particular document on a web server,[3] among many other phenomena.

The Poisson process is a continuous-time process; the sum of a Bernoulli process can be thought of as its discrete-time counterpart. A Poisson process is a pure-birth process, the simplest example of a birth-death process. It is also a point process on the real half-line.

That is stats 101, as in you should learn about it in any introductory stats class. I learned about Poisson processes in Cegep. I first took a probability and stats course in CEGEP whose name I forget. I took math 323 and 324 at McGill which covered examples like this, I was later a teaching assistant for 323.

Obviously, in the real world, as I've said before, the noise will be even bigger due to factors such as injuries. However, the noise cannot be smaller.

Quote:

Originally Posted by Simerc

Yep, he just made that up. To find any kind of "statistical noise", you need a set of observations from a sample.

If you don't understand what someone else is talking about, you should ask them to clarify, it's good practice.

Last edited by DAChampion: 09-22-2012 at 06:38 PM.

That is stats 101, as in you should learn about it in any introductory stats class. I learned about Poisson processes in Cegep. I first took a probability and stats course in CEGEP whose name I forget. I took math 323 and 324 at McGill which covered examples like this, I later was a teaching assistant for 323.

Obviously, in the real world, as I've said before, the noise will be even bigger due to factors such as injuries. However, the noise cannot be smaller.

N is the mean, as in the number of points you expect a player to get.

I can try and explain it more simply than wikipedia did by parsing their explanation:

A process which counts the number of events in a given time interval. The time between each pair of consecutive events, the inter-arrival times, is assumed to be independent of other inter-arrival times.

The first sentence obviously describes points in a season. The second just means that they will come from 1 point games, 2 point games, etc.

Here's a good random number generator: http://www.stochastic-lab.com/random
You can put in "80", and "Poisson", and see that an 80 point player with no injuries might have six seasons like this:
74,91,74,70,86,69

i did my stats 221, in economics, thought it's been a while, i have hard time to remember what you are talking about. Seems a bit too simple to be true.

By statistical noise, do you mean the standard deviation? that would be square root of the variance, and don't see how u calculated the variance here. As to the mean, it has not much to do with the variance if i remember correctly.

If by "not much," you mean "can be calculated directly from," then you're CORRECT!

Yes he's actually applying a different principle. It's not Std or variance that he's using. It's called coefficient of variance (CV). Correct me if i'm wrong, my stats is fading away.

For larger sample sizes, it's actually more useful than STD which is limited by context.

CV, in poisson, is the square of the mean. Again in a poisson process, following a poisson distribution, is the rate parameter. So in this case, he's assuming a rate of 92.5 pts per season.

As an engineer, it's easier for me to explain CV as the reciprocal of SNR (signal to noise ratio) which looks at mean in relations to the STD. So if you have SNR of 1:1, for every bit of info, you have one bit of noise.

You can apply this logic to his statement.

And also about to do a small write up about the Belleville game which is what we're all in this thread for. stay tuned.

Yes he's actually applying a different principle. It's not Std or variance that he's using. It's called coefficient of variance (CV). Correct me if i'm wrong, my stats is fading away.

For larger sample sizes, it's actually more useful than STD which is limited by context.

CV, in poisson, is the square of the mean. Again in a poisson process, following a poisson distribution, is the rate parameter. So in this case, he's assuming a rate of 92.5 pts per season.

As an engineer, it's easier for me to explain CV as the reciprocal of SNR (signal to noise ratio) which looks at mean in relations to the STD. So if you have SNR of 1:1, for every bit of info, you have one bit of noise.

You can apply this logic to his statement.

And also about to do a small write up about the Belleville game which is what we're all in this thread for. stay tuned.

I'm going to watch Sarnia game too. I'm gonna try it tomorrow with the game against Kitchener and decide if I want to pay for a season. But I'm very excited to see him play. Hope he have a great year. I think he's gonna have a 80pts in 50 games with Sarnia. As for his career in Montreal, he will be a perrenial 80pts in the NHL.

The stream worked well for me yesterday, you get the pre-game with interviews like a half hour before the game and it doesn't lag, but don't expect much as far as quality goes. Gally is a joy to watch, you could become addicted to following this guy.

Before I joy puke about our 3rd overall pick, let me profess my man crush for some other players. As always, I need to put the disclaimer here that I'm not a pro scout nor do I normally scout. Simply a hockey fan that needs hockey outside my busy work schedule.

Jordan Subban - most of you know him by now. PK's younger D brother. My god what a talent. He reminds me of PK in every single way, even his brain cramps. Granted, he doesn't have the physical edge PK does (at least not in this game) but he has a really high level first pass that gets him out of trouble very quickly and at the same age as his brother, a higher understanding for the defensive side of the game. You can expect the same Subbanian rushes up the ice and trying to beat others one on one as PK. Perhaps not as hard of a shot as PK but regardless, for his age, his package screamed WOW.

Charles Sarault - there are rumours that Habs have him invited to their camp and I hope that is true. Any less would be tragic given that we have a close eye on Sarnia and ther GM/Coach is the father of our own Nathan Beaulieu. This kid can play. His hockey sense is very high. I frequently see him breaking up passes in the neutral zone and he knows where his linemates are at all times. That said, he is very very raw in his own zone. Still his passing is quite awesome and has a hard, if not always accurate wrister. We have to get him signed. If Philippe Lefevbre got a contract, I hope we give Sarault a good long look.

Connor Murphy - this kid is calm cool and collected. Pheonix picked a good one here. Many times I thought he would be in trouble with aggressive forechecking but he skated the puck up around his net and made a crisp first pass. He has a good shot from the point that is accurate and low. He won't wow you in many respects but he was solid all this game around. very noticeable player for mostly the right reasons.

And now, Gally. I'm giddy with excitement. Gally in this game did not completely live up to my high expectations but he was good nonetheless. Lets get the negatives out of the way first. For an elite playmaker, which i thought he was billed to be, he needs to use his linemates better. He is a great stickhandler but often times, I find him stickhandling out of trouble and into the corners rather than utilizing an open linemate. He started the game on LW beside Boucher and Sarault and looked decent. In the second period, he got shifted back to C on 4 on 4 situations along side Goldobin and they looked good together. He is prone to making turnovers which was very frequent earlier on in the game. Also, I'm not sure if it was a coaching decision, but he had very long shifts along with Sarault. I saw him out there for a good 2 to 2.5 min at some point. I would like to see him shorten those up. His skating is decent but he rarely beats defenders with speed. His game is more about intensity, vision and a deceptively quick shot. So on to the positives. I can really see why people have him labelled as an elite playmaker. His passing is crisp, accurate and deadly. On occasions, I see him cross the blueline and make a pass through two defenders and think, no way it hits his teammate. But as soon as I thought that, my jaw dropped and seem it connect for a good scoring opportunity. Gally has great great vision. I cannot stress that enough. He is also not afraid of traffic. I have seen him cut into the middle with the puck many times. He would occasionally lose it or get hit but it's great to see someone of his talent play a north south game. The speed with which he makes decision is already NHL level I think. He has a quick release but I would like to see him use it more often. His shot selection could also improve. What also surprised me was the feisty/chippy aspect of his game. If a scrum broke lose after the play and he was on the ice, he would be the first one involved. Its great to see him not back down. I honestly think, if he bulks up, and takes shorter shifts, we can see him in the NHL permanently as early as next year.

I'll hope to do this after every game if work allows it. Thanks for reading. Let me know what aspects of his game you'd like me to concentrate on next and I'll do my best.