[News] USA Olympic Roster announced

[strombad]

Originally Posted by diane_phaneuf

they are not giving their opinion



they are showing numbers, numbers don't lie, people's bias/eye tests do

They are placing value in a system with a questionable value. You don't ever wonder why basement bloggers come up with posts that just "show the numbers" and paid professionals who are actually respected often don't?

It's a tiny part of the equation. You can't just trot out advanced stats and say "this is why the executives are wrong", that's why those guys don't get paid, because they're level of knowledge on the game is low.

People like that look at stats to evaluate the player. Anyone with apt hockey knowledge looks at player first, and then the stats.

The problem with advanced stats, and the problem with people who rely solely on them (and are relegated to the hockey buzz line of work) is that they exist to support an argument, not to make an argument. You can't say Bobby Ryan is a good defensive forward because advanced stats say so, nor can you say Jack Johnson is a bad defenceman. Why? Because according to advanced stats, there are plenty of bottom feeding NHLer's who should be great, and great NHLer's who should be terrible.

The system is inherently flawed, but that will never stop basement bloggers and message board fans not much different from myself from trotting it out like it holds the answer.

In this case, numbers do lie. And people who prop up advanced stats don't seem to understand that.

[saillias]

Here's an idea for an all-star game:

NHL All-Stars vs Corsi-Fenwickers

Fans vote for the NHL all-stars on one team.

An executive committee of stat gurus is created to put together the best CORSI-Fenwick closers/or whatever on the other team.

The ultimate showdown.

No?

[theinfinitejar]

My favorite part of "advanced" stats arguments are the strawmen. Guys, not one person who believes in using possession metrics thinks best Corsi = best player. The folks who count Corsi know it's just one tool in a toolbox and context is key, but anytime an argument comes up about fancy stats the "I watch the games" crowd always feels the need to insult people's intelligence with "you must think Backlund's better than Kane" nonsense. All your're doing is proving you don't know what you're talking about.

As far as Team USA goes, I think there's validity to the approach that they took. They're not going to be the most talented team in the tournament so they need shut down lines and solid two-way defence to combat the depth of scoring that Canada can put on the ice.

Canada on the other hand just needs to not overthink it. You put the best team out there and don't worry about what the other team does, let the Hockey gods sort out the rest. The roster Canada can put together is going to be so talented you just roll four scoring lines and make the other guys try to figure out how to stop you.

[Dagger]

Originally Posted by strombad

They are placing value in a system with a questionable value. You don't ever wonder why basement bloggers come up with posts that just "show the numbers" and paid professionals who are actually respected often don't?

It's a tiny part of the equation. You can't just trot out advanced stats and say "this is why the executives are wrong", that's why those guys don't get paid, because they're level of knowledge on the game is low.

People like that look at stats to evaluate the player. Anyone with apt hockey knowledge looks at player first, and then the stats.

The problem with advanced stats, and the problem with people who rely solely on them (and are relegated to the hockey buzz line of work) is that they exist to support an argument, not to make an argument. You can't say Bobby Ryan is a good defensive forward because advanced stats say so, nor can you say Jack Johnson is a bad defenceman. Why? Because according to advanced stats, there are plenty of bottom feeding NHLer's who should be great, and great NHLer's who should be terrible.

The system is inherently flawed, but that will never stop basement bloggers and message board fans not much different from myself from trotting it out like it holds the answer.

In this case, numbers do lie. And people who prop up advanced stats don't seem to understand that.

Out of curiosity why is it that you can make a blanket statement like that without backing it up, but anyone who goes the other way around and examines the statistics and then considers circumstances/opinions/what they have seen is wrong? How do you know that the way you describe is correct?

[d_phaneuf]

Originally Posted by PeteMoss

I will close with this... Corsi and these advanced stats are clearly good for telling what happened and can predict on a team level what may happen. But until they can tell you how to fix it... they are useless for a GM beyond saying 'well we are being really lucky or unlucky right now and we might want to make a move to based on that'

I think they are a much better predictor of team success than individual

the one I have really found useful is when they show players Corsi when they play with certain guys vs when they don't

i.e everyone on the Bruins had a higher Corsi with Chara vs without Chara

it's still a long way to go, NHL stats are where baseball stats were 10-15 years ago and it likely won't grow as quick anyway because team sports are so much difficult than individual

[d_phaneuf]

Originally Posted by strombad

They are placing value in a system with a questionable value. You don't ever wonder why basement bloggers come up with posts that just "show the numbers" and paid professionals who are actually respected often don't?

It's a tiny part of the equation. You can't just trot out advanced stats and say "this is why the executives are wrong", that's why those guys don't get paid, because they're level of knowledge on the game is low.

People like that look at stats to evaluate the player. Anyone with apt hockey knowledge looks at player first, and then the stats.

The problem with advanced stats, and the problem with people who rely solely on them (and are relegated to the hockey buzz line of work) is that they exist to support an argument, not to make an argument. You can't say Bobby Ryan is a good defensive forward because advanced stats say so, nor can you say Jack Johnson is a bad defenceman. Why? Because according to advanced stats, there are plenty of bottom feeding NHLer's who should be great, and great NHLer's who should be terrible.

The system is inherently flawed, but that will never stop basement bloggers and message board fans not much different from myself from trotting it out like it holds the answer.

In this case, numbers do lie. And people who prop up advanced stats don't seem to understand that.

I agree with you on most of this, it wasn't a negative I was saying

ideally stats or any numbers are used as a check so to speak

you either want to look at the Chara thing like I mentioned above, and say, "I think Chara is dominant and these stats completely prove it"

Or for example

"I think player Y is bad, but the stats show he is doing __________, where is the disconnect"

the worlds are not exclusive of each other

I imagine the majority of NHL teams use some type of statistical analysis right now, they just want to keep a competitive advantage so it's not talked about

[Resolute 14]

Originally Posted by diane_phaneuf

it's still a long way to go, NHL stats are where baseball stats were 10-15 years ago and it likely won't grow as quick anyway because team sports are so much difficult than individual

Truthfully, hockey stats are probably a century behind baseball, and the gap is unlikely to ever close significantly. Baseball is a zero sum game, and it is very, very easy to measure most events. One pitch, one pitcher. One batter, one hit, four bases. Hockey is far too fluid and every player is basically his own x-factor.

[jg13]

[strombad]

Originally Posted by Dagger

Out of curiosity why is it that you can make a blanket statement like that without backing it up, but anyone who goes the other way around and examines the statistics and then considers circumstances/opinions/what they have seen is wrong? How do you know that the way you describe is correct?

Common knowledge and the human condition.

[19Yzerman19]

Or you're just trolling.

[Dagger]

Originally Posted by strombad

Common knowledge and the human condition.

Proven where?

[OffsideSpecialist]

Originally Posted by jg13

I searched Twitter for her, and it looks like she deleted her account over that.

[19Yzerman19]

Originally Posted by Red John

Thing is these events aren't actually random. A hockey game is full of decisions every single second - decisions about what angle to take to the puck when skating, decisions about whether to make a head fake, tons of tiny little things that go on. Each of these tiny decisions ultimately makes an impact on the final outcome. People who have played the game at a high level understand this.
Skating, stickhandling, passing, and tons of little decisions that happen with every player on the ice at every moment of a game. And there is no stat out there that can track those decisions.

Don't be patronizing. Everyone understands this. Nonetheless, goals are fairly random in terms of their source. Your decision about what angle to take did not cause you to score a goal that deflects into the top corner off a defenseman's skate. Your head fake did not cause the opposing goalie to let in a dipping shot from the blue line he'd stop 49 times out of 50. Your stickhandling skill did not cause the rookie d-man you're facing to cough the puck up the middle for an easy goal. These types of aberrations create noise. The noise is washed out given a larger sample of data, which is why you need a full season and why you use shots instead of goals. That is what is meant by randomness when associated with goal scoring.

That's the problem, and that's why most of the "stats crowd" are basement bloggers, or at the very least not the ones who played the game and are working in the game - because they are trying to reverse engineer the game and use the final result to try and quantify the process. Instead of watching the process unfold with their eyes, drawing conclusions based on those observations, and letting the results fall where they may. If their observations are accurate, most of the time the results will provide confirmation.

As noted, I guarantee that these guys watch more hockey than you. The notion that people who care about the stats must not actually be watching hockey is asinine. Even more so given that if they weren't watching the games, how would they be tracking the stats in the first place?

[Red John]

Originally Posted by 19Yzerman19

Don't be patronizing. Everyone understands this. Nonetheless, goals are fairly random in terms of their source. Your decision about what angle to take did not cause you to score a goal that deflects into the top corner off a defenseman's skate. Your head fake did not cause the opposing goalie to let in a dipping shot from the blue line he'd stop 49 times out of 50. Your stickhandling skill did not cause the rookie d-man you're facing to cough the puck up the middle for an easy goal. These types of aberrations create noise. The noise is washed out given a larger sample of data, which is why you need a full season and why you use shots instead of goals. That is what is meant by randomness when associated with goal scoring.

You just demonstrated that you don't understand this. The decision about what angle to take determines who gets to the puck first, perhaps. Which influences which team takes possession...which influences a breakout...or maybe a line change. Maybe a great scoring chance occurs that wouldn't have otherwise. The head fake may have made the defenseman hesitate just a split second opening up a passing lane that wasn't there otherwise.

Every single little event like this influences the outcome. People who played the game especially at the NHL level understand this. That's why its so funny when fans often say things like "if that goal had counted we would have won" because that isn't necessarily true. If that goal had counted the faceoff would have taken place at centre ice, not inside the zone. Maybe a different line matchup gets sent out. Maybe the linesman drops the puck differently. Different possession ensues leading to a completely different course that is charted as the game goes on.

As noted, I guarantee that these guys watch more hockey than you. The notion that people who care about the stats must not actually be watching hockey is asinine. Even more so given that if they weren't watching the games, how would they be tracking the stats in the first place?

This wasn't about comparing basement bloggers to other fans such as myself. It was about comparing basement bloggers to NHL scouts and executives who watch more hockey and more importantly understand what to watch for.

The scout sitting in the stands watching the game in its entirety will learn a lot more than the basement blogger watching on TV with one eye while typing scoring chances on his laptop with the other.

And the beauty is - the scout will know WHY the stats finished the way they did because he'll have the context. The blogger will just have a bunch of numbers.

[Iggy Snipe]

No Bobby Ryan? Good for Canada I guess, that guy is 3rd among American scorers in the NHL. Team USA does have a strong team, stronger than in 2010 IMO.

Keith Yandle probably should have been there too.

[SebC]

Originally Posted by jg13

I don't see what's so bad about that. I'm glad he didn't make the USA team too!

[19Yzerman19]

Originally Posted by Red John

You just demonstrated that you don't understand this. The decision about what angle to take determines who gets to the puck first, perhaps. Which influences which team takes possession...which influences a breakout...or maybe a line change. Maybe a great scoring chance occurs that wouldn't have otherwise. The head fake may have made the defenseman hesitate just a split second opening up a passing lane that wasn't there otherwise.

Wow, not only have you continued to be unjustifiably patronizing but you've become more obtuse in the process. The fact that you're casting all of this as a skill, as if guys take a particular angle to a puck so that ten moves later it'll lead to a goal is hilarious. Often, goals result from a deliberate skill play, and often, the puck bounces off a skate right to a guy in front for a tap in. This isn't chess.

There are lots of unexpected elements that lead to goals going in or not that makes a small sample size of TOI less useful as a predictive measure. Which is why when a fourth liner has a three point night, that doesn't somehow mean he's capable of it on a regular basis. Alex Steen is not a top five goal scorer in the league, Kyle Okposo is not a better scorer than Phil Kessel, and Cam Talbot isn't the best goalie in the league... any more than Mark Parrish was an 80+ point player, the 2011-12 Wild were the best team in the Western Conference or any number of other small-sample trends that regressed to reality over time.

No one is saying that there's a silver bullet stat that explains why teams win in the NHL. All that your basement bloggers on their laptops (as you again refer to them in unbelievably arrogant fashion while typing out your opinions on a hockey message board) are doing is accumulating data and testing and tweaking the methodology to try to make a better model to predict how teams will end up performing. Why that kind of curiosity, which can't possibly hurt and can only help provide more information about hockey, would offend anyone is... silly. But not surprising, I guess, sadly.

[strombad]

Originally Posted by 19Yzerman19

Or you're just trolling.

Oh you right G, don't like advanced stats, must be a troll.

[19Yzerman19]

I'm just giving you the benefit of the doubt here.

[sergei_makarov]

Clearly upset by Burke's comments yesterday, Bobby Ryan responded with 0 points and 1 SOG tonight in a 4-3 win over Winnipeg.