NHL Power Rankings

[saillias]

Honest question: Have advanced stats accurately predicted Stanley cup winners in years past?

[dino7c]

Every game the Flames win he just looks like more and more of a ######...I hope he keeps it up

[ClubFlames]

The Score guy is a dirty Canucks fan so no wonder his hate for Calgary is up there. He admits himself that he's not interested in looking at team records at this point, only underlying numbers and roster quality. Finding a way to win and be 4 games above .500 doesn't mean crap anymore. Pure garbage.

[Biyaaaah]

Advanced stats are getting out of hand and ridiculous. They have their place when breaking down a team but it seems too often that they are relied upon for conclusions by journalist that are trying to make a point about a team they don't follow or watch regularly. It's not hard to cherry pick stats to suit your argument. It seems to me that the people who rely the most on these stats are always the ones with the least meaningful insights and commentary.

Don Cherry who whether you like him or not really understands the game at a much higher level than well informed fans. I find that his video breakdowns of systems and adjustments made by teams and players provide far more insight into the intricacies of the game than any corsi quoting journalist.

If you've watched the flames play this year it's no fluke. They stole a couple games, they've had a couple stolen from them as well. A few prominent coaches have come out and called the team amongst the hardest to play against and incredibly well structured. Discipline within your system and consistency win games. Bottom line this team has earned every bit of it's record thus far.

GFG

[formulate]

Originally Posted by saillias

Honest question: Have advanced stats accurately predicted Stanley cup winners in years past?

No, and that's not really the point. Guys like Yost seem to heavily rely on these statistics, which I feel really distracts from their actual usefulness.

This picture is old, but gives a snapshot of the impact of possession. It's a collection of possession numbers (for 5 years of regular seasons from 2007-08 to 2011-2012). It uses fenwick (unblocked shot attempts) at even strength in close score situations. Not a perfect metric by any means, but a reasonable substitute for possession %. Starting from the positive x axis (40% ie Buffalo bad) moving counter clockwise upwards in possession.

Basically, teams with a sub 50% possession rating made the playoffs 29% of the time. Teams over that mark made it 78% of the time

Spoiler!

Bad possession teams such as Calgary this year (and Montreal/Colorado last year, Toronto the year before, etc) CAN make the playoffs or better. It's just less likely based on historical data.

I tend to liken 'advanced' statistics to special teams percentages in terms of importance. Having a poor PK and PP makes winning games more difficult, but not impossible. And poor special teams doesn't necessarily mean a poor team. Same idea with possession metrics like corsi and fenwick.

The biggest problems with these stats is how they are being used. You have the Yost-type guys preaching nothing but, which usually ends up in shouting matches with the 'watch the games' crowd. Then you have casually interested fans who take the stats out of context which really takes away from the perceived credibility of the stats.

[Tinordi]

Originally Posted by Cali Flames Fan

Well it's a good thing I don't believe in luck. What a stupid archaic concept to try and force into an analytical presentation. Luck…gimme a break.

Good teams force the issue and capitalize on their chances. Maybe being outshot regularly isn't going to lead to as many wins as the elite teams, but there is a formula for the Flames current success, and it doesn't seem to matter if they give up a lot of shots from the perimeter and then play a possession game in the offensive zone getting only 1 or 2 great shots every couple of minutes. Since we're talking about this imaginary concept of luck, how about we go for at least a subjective analysis of "scoring chances" per game since at least that doesn't involve fairy dust and leprechauns.

Luck. So sick of that word.

This is seriously unhinged

You don't believe in luck? Do you not believe in air, or time either?

[Street Pharmacist]

"advanced stats" seem to be so misunderstood by both media and fans. They aren't better, they just tell another part of the story. Save percentage, shooting percentage, +/-, etc all tell a different part of the story.

And yes, corsi would've predicted LA to win the cup this year, but it's not accurate that often. Montreal had a terrible corsi and went to the conference finals. Generally speaking, the team with the best corsi wins. There's lots of exceptions.


By a million miles, the best predictive stat for future team success is goal differential. We seem to be just fine there

[Cali Panthers Fan]

Originally Posted by Tinordi

This is seriously unhinged

You don't believe in luck? Do you not believe in air, or time either?

People use luck to try and explain situations that they don't understand. It's akin to using God for the reason the planet is getting warmer.

The Flames are getting better results than a few advanced stats would indicate. One guy looks at that, doesn't understand how had they work and how they make key plays at key times and reduces it down to "luck". I hesitate to suggest a different word, but "fortunate" might be better. However, in my experience, those who appear "fortunate" have put in a lot of effort to set themselves up for success. Success is almost never accidental and luck doesn't exist.

[Red John]

Basement bloggers are just tripping over themselves in order to sound smart by using "advanced stats" right now that they are forgetting one critical point in any set of data: sample size.

It's not even a quarter of the way through a season. Sometimes not even 82 games is a big enough sample size to allow stats to regress to the mean (ie Colorado last year, Toronto the year before).

So to be spouting these numbers after a dozen + a few games is just dumb.

[Street Pharmacist]

Originally Posted by formulate

No, and that's not really the point. Guys like Yost seem to heavily rely on these statistics, which I feel really distracts from their actual usefulness.

This picture is old, but gives a snapshot of the impact of possession. It's a collection of possession numbers (for 5 years of regular seasons from 2007-08 to 2011-2012). It uses fenwick (unblocked shot attempts) at even strength in close score situations. Not a perfect metric by any means, but a reasonable substitute for possession %. Starting from the positive x axis (40% ie Buffalo bad) moving counter clockwise upwards in possession.

Basically, teams with a sub 50% possession rating made the playoffs 29% of the time. Teams over that mark made it 78% of the time

Spoiler!

Bad possession teams such as Calgary this year (and Montreal/Colorado last year, Toronto the year before, etc) CAN make the playoffs or better. It's just less likely based on historical data.

I tend to liken 'advanced' statistics to special teams percentages in terms of importance. Having a poor PK and PP makes winning games more difficult, but not impossible. And poor special teams doesn't necessarily mean a poor team. Same idea with possession metrics like corsi and fenwick.

The biggest problems with these stats is how they are being used. You have the Yost-type guys preaching nothing but, which usually ends up in shouting matches with the 'watch the games' crowd. Then you have casually interested fans who take the stats out of context which really takes away from the perceived credibility of the stats.

Well put. Probabilities aren't everyone's strong suit

[Tinordi]

Originally Posted by Cali Flames Fan

People use luck to try and explain situations that they don't understand. It's akin to using God for the reason the planet is getting warmer.

The Flames are getting better results than a few advanced stats would indicate. One guy looks at that, doesn't understand how had they work and how they make key plays at key times and reduces it down to "luck". I hesitate to suggest a different word, but "fortunate" might be better. However, in my experience, those who appear "fortunate" have put in a lot of effort to set themselves up for success. Success is almost never accidental and luck doesn't exist.

This could not be more untrue in any aspect in life. Luck is a huge factor in success of any kind. Mountains of research to prove it.

People try to explain what they don't understand with simple causality, not the other way around.

[GioforPM]

Originally Posted by Tinordi

This could not be more untrue in any aspect in life. Luck is a huge factor in success of any kind. Mountains of research to prove it.

People try to explain what they don't understand with simple causality, not the other way around.

I think you two are using different concepts of luck. Luck as a fortunate turn of events is fine and definitely occurs.

Luck as some intangible quality that a person or team possesses as an asset is imaginary.

[OzSome]

Power Ranking means absolutely nothing. It won't affect the NHL standing and team don't get bonus points for being the top ranked team. The only thing the Power Ranking is good for is to motivate the team to play even harder. Let them put the Calgary Flames close to the bottom as much as they want as long as the Flames keep proving them wrong.

[GettinIggyWithIt]

Originally Posted by GioforPM

I think you two are using different concepts of luck. Luck as a fortunate turn of events is fine and definitely occurs.

Luck as some intangible quality that a person or team possesses as an asset is imaginary.

Yes - thank you. Simply put - in the context of the writer's statement - it's not luck, it's great goaltending and shot blocking that has enabled the Flames to win games despite being outshot in shot attempts overall. Also I am not sure why the Flames are being penalized for blocking shots . Either way - I love that we are so low on the power rankings. All the more reason to keep smiling when we keep winning..

[TheDebaser]

Corsi is a possession stat, not a scoring stat. A low Corsi should be worrying because it's a way of showing that the opposition is the team that is carrying play. It doesn't show anything more or anything less.

[CaptainCrunch]

If the money ball concept worked in Hockey then Edmonton who bought in a basement dwelling stats nerd would be much better then they are, and guidos like Purcell and Poulliot would be all stars this year.

I still believe that there is a completely random element to the game (ie lw'er going down the right side of the ice, depth of backcheck gaps) that kills a lot of the advanced stats.

I'm not saying that there is no use for them, but I'm saying that too many people use them in arguments that there is no point in arguing.

[Enoch Root]

Originally Posted by CaptainCrunch

If the money ball concept worked in Hockey then Edmonton who bought in a basement dwelling stats nerd would be much better then they are, and guidos like Purcell and Poulliot would be all stars this year.

I still believe that there is a completely random element to the game (ie lw'er going down the right side of the ice, depth of backcheck gaps) that kills a lot of the advanced stats.

I'm not saying that there is no use for them, but I'm saying that too many people use them in arguments that there is no point in arguing.

Exactly.

Not only is there a massive random element that has to be considered, but also, many of the more 'useful' stats have very small discrepancies. In other words, people talk about this player's Corsi being 53% and that player only 49%. In order for there to be a real difference between 53% and 49% (for example), you would need a lot of data. And that's ignoring the large amount of random noise that's going on.

With all of the players, bounces, randomness and 'luck' that is a part of hockey, it requires a crazy amount of data in order to be able to have any confidence that there is any real information in the stats. Yet people throw them around like they are gospel.

Not saying they are useless, just that they are over-valued, over-used and miss-interpreted more often than not.

[Cali Panthers Fan]

This is why advanced stats are used more in baseball. With double the games they have double the amount of data coming in.

[afc wimbledon]

Originally Posted by Cali Flames Fan

This is why advanced stats are used more in baseball. With double the games they have double the amount of data coming in.

I think also,with all respect to baseball, it's a much simpler more predictable game.

[GioforPM]

Flames (at least to midway thru the third which was the last I heard) were out shooting the Canes. No one could suggest they were outplaying them.