[News] CBS Sports power ranking w/ intro geared at Calgary

[Wood]

I know these are just random power rankings, but how do they put San Jose ahead of us?

We are ahead of them in the standings, have more wins in less games, we literally just beat them before they wrote the article in a dominant game, and we've won 4 of 5 games against them this year

San Jose is not a better team than the Flames

They even say in San Jose's blurb that no western team will want to play them in the first round. Why would the Flames not want to play the team they've dominated this year? Not to mention they are close to being the biggest playoff chokers of all time

[RyZ]

Originally Posted by Wood

I know these are just random power rankings, but how do they put San Jose ahead of us?

We are ahead of them in the standings, have more wins in less games, we literally just beat them before they wrote the article in a dominant game, and we've won 4 of 5 games against them this year

San Jose is not a better team than the Flames

They even say in San Jose's blurb that no western team will want to play them in the first round. Why would the Flames not want to play the team they've dominated this year? Not to mention they are close to being the biggest playoff chokers of all time

They are clearly stat watchers and not game watchers.

Many power ranking lists this season are more concerned about Corsi % than points % or goal differential and it's laughable.

[polak]

Advanced stats have literally ruined "hockey talk".

100% of the time any hockey discussion about a team breaks down to corsi and PDO and one side arguing how awesome and flawless it is and the other side argues why it sucks.

Any stat that doesn't take into consideration quality over quantitiy shouldn't be relied on as the be all and end all for a discussion, but it's literally all the advanced stats guys do. They never try to explain why the Flames are defying their statistical analysis, which is the anti-thesis of statistical analysis, they just brush it off and yell "unsustainable", "they'll fall back to earth", "COLORADO" "COLORADO!!!!!"

[Street Pharmacist]

Originally Posted by Fighting Banana Slug

I don't care for the "defying logic" narrative, but the write-up wasn't that bad. Indicates we have good, not great, goaltending, an outstanding 1/2 D pair and good seasons from Monahan and Hudler.
I must admit, I didn't understand this line:
"War on Ice also has Calgary with a scoring chance differential of minus-211."
Any sense from those in the know how they came up with scoring chance differential? Is that just a measure of shots on/at net?

War on ice took a look at all types of shots that were associated with high shooting percentages (on the rush in closer zones, from the slot and rebounds). They count all attempted shots that for that description. The easy counter to that for the flames is that the Flames are usually blocking those shots and the numbers bear that out

[FBI]

We already did the cycle in the corners game a few years ago with our declining vets.
It didn't work, I'm happy to play the rush game and lose the corsi battle.

[heep223]

Originally Posted by Red

Flames are a small team and because of that they are not able to cycle or grind it on the boards. When they try it's short lived. Thus low possession.

Hartley has them playing an off the rush and stretch pass game instead. It suits them better. This is why they are such a poor possession team.

Agreed in principle.

But in regards to Corsi as a proxy for possession...how would this be reflected in a shots for:shots against differential? IE if the Sharks are cycling and grinding along the boards, it's not helping their shot differential.

[GioforPM]

Yeah, I've always thought really good cycling teams who work the puck for long periods of time to get off a couple good shots are ripped off by shots-based possession stats.

When the tech from the ASG comes in to play there will be very accurate actual stats on time of possession in the offensive zone, and Corsi may become outmoded.

[Bandwagon In Flames]

It's really become ridiculous how widely used advanced stats have become. They can take over any debate when they should only be used as supporting stats when used in the proper context.

Go from using 'advanced stats' to support claims, then say words like 'unsustainable' and 'lucky' to counter any team that doesn't follow the trend. Doesn't seem all that scientific.

This isn't baseball and the formula for a fail-proof advanced stat in Hockey has not yet been created. Teams defy the odds for different reasons every year. The reason this sport is awesome is due to it's complexity.

/rant

[djsFlames]

Beyond the game to game stats, they should put more weight on goal differential, as the teams with the best differentials always seem to gravitate to the top of the standings. You don't often see a team that is +2 finish above the team thats +24. Just as we're (+19) now pulling ahead of San Jose and Vancouver (+1, +7), who have worse differentials. Seems to be a very consistent trend in how the standings play out.

[bob-loblaw]

Meh. The Flames play a different type of game that statistics can't pick up on.

[Tinordi]

I just don't get the hate for advanced stats especially on a message board. There are hundreds of conversations going on on this board that never or rarely mention them, but when one thread opens up offering different insight a bunch of people feel obliged to pipe in how about these stats are ruining their ability to talk sports and the enjoyment that comes with it. Here's the thing, just move on if you don't like talking stats and analytics. There's no shortage of the traditional basic sports talk to be found.

[Jay Random]

Originally Posted by Tinordi

I just don't get the hate for advanced stats especially on a message board.

I think advanced statistics are terrific.

I think it would be nice if there were some for hockey. Corsi and Fenwick aren't advanced. PDO isn't even a statistic; it's a fudge.

When one team can have terrible Corsi and Fenwick numbers and be in a playoff position, and another team can have good Corsi and Fenwick and the second-worst record in the league, a rational person would conclude that Corsi and Fenwick don't tell the whole story. Clearly there is some other factor involved in winning hockey games.

Instead, the ‘advanced’ stat hounds try to dismiss those teams and claim that their success or failure is all due to luck. They can't explain the Flames or the Oilers, so they try to explain them away.

[Tinordi]

Originally Posted by Jay Random

I think advanced statistics are terrific.

I think it would be nice if there were some for hockey. Corsi and Fenwick aren't advanced. PDO isn't even a statistic; it's a fudge.

When one team can have terrible Corsi and Fenwick numbers and be in a playoff position, and another team can have good Corsi and Fenwick and the second-worst record in the league, a rational person would conclude that Corsi and Fenwick don't tell the whole story. Clearly there is some other factor involved in winning hockey games.

Instead, the ‘advanced’ stat hounds try to dismiss those teams and claim that their success or failure is all due to luck. They can't explain the Flames or the Oilers, so they try to explain them away.

Oh god, now we're into semantics? This proves my point. Some people are trying to gain more insight into the game but some others have nothing but derision for that effort. It's like unless we developed a unified statistical theory of hockey that explains everything then nothing is of value. Instead lets all go back to the real insight offered by Don Cherry and other savants of hockey genius to give us the important and meaningful sermons.

[Red]

Originally Posted by Tinordi

Oh god, now we're into semantics? This proves my point. Some people are trying to gain more insight into the game but some others have nothing but derision for that effort. It's like unless we developed a unified statistical theory of hockey that explains everything then nothing is of value. Instead lets all go back to the real insight offered by Don Cherry and other savants of hockey genius to give us the important and meaningful sermons.

Advanced Stats people need to show us why the Oilers are so bad. We hear about how unsustainable Flames record is based on recent fortunes (unreal goaltending for most part) followed by misfortunes of Toronto and Colorado. But let's hear the other side of the coin, teams with good stats that lose more than they win.What do the stats show there?

[sa226]

I think that most people see some value in advanced stats, but they are blinded by the conclusions that most purveyors of advanced stats jump to.

The fault is on both sides. There shouldn't be an "anti" or a "pro" advanced stats argument. They should be viewed as our very own GM views them, one more tool to use. But not the be all and end all.

[Coach]

Originally Posted by Tinordi

Oh god, now we're into semantics? This proves my point. Some people are trying to gain more insight into the game but some others have nothing but derision for that effort. It's like unless we developed a unified statistical theory of hockey that explains everything then nothing is of value. Instead lets all go back to the real insight offered by Don Cherry and other savants of hockey genius to give us the important and meaningful sermons.

It's the people that use the advanced stats to make definitive statements on teams that are the problem. Nothing wrong with looking for new ways to interpret the game, but a lot of these people (like the Yahoo guy, can't remember his name) are employing it as the be all end all, and if you are an outlier in their stats, it's luck and nothing else.

It's the "advanced stats people" (not everyone who finds them intriguing, the ones that actually use them to write articles and predict success) that are seeing it as black and white. The effort is great, keep the new stats coming out. It's the interpretation is not great, and as soon as they create a new one, they use it as a measure of success, which it's just not.

No one is looking for the "unified statistical theory of hockey that explains everything", they're sick of people using PDO and Fenwick etc... as such. Which is not everyone who looks at them, but many people that are actually being paid to analyze the game are treating them that way.

[Vulcan]

Originally Posted by Tinordi

Oh god, now we're into semantics? This proves my point. Some people are trying to gain more insight into the game but some others have nothing but derision for that effort. It's like unless we developed a unified statistical theory of hockey that explains everything then nothing is of value. Instead lets all go back to the real insight offered by Don Cherry and other savants of hockey genius to give us the important and meaningful sermons.

Everyone would like more insight into the game. The trouble starts with the stats provided often don't jibe with the game results and yet people use these stats to try and predict the results or the strength of a team by these stats. It gets even worse when these people take a holier than thou approach with these stats.

[Weitz]

Originally Posted by Tinordi

I just don't get the hate for advanced stats especially on a message board. There are hundreds of conversations going on on this board that never or rarely mention them, but when one thread opens up offering different insight a bunch of people feel obliged to pipe in how about these stats are ruining their ability to talk sports and the enjoyment that comes with it. Here's the thing, just move on if you don't like talking stats and analytics. There's no shortage of the traditional basic sports talk to be found.

I think this year especially because of the flames success in spite of the advanced stats has people up in arms about it.

I wonder if you looked back last year what the thoughts were when Colorado and Toronto (before their cliff dive) were doing well in spite of the stats, what the thoughts were.

(Not saying the flames have been out played or whatever, they certainly have deserved the majority of their wins)

[heep223]

Is Corsi as a stand alone statistic highly correlated to future wins? Is it statistically relevant?

[Resolute 14]

Originally Posted by Weitz

I wonder if you looked back last year what the thoughts were when Colorado and Toronto (before their cliff dive) were doing well in spite of the stats, what the thoughts were.

People expected both teams to nosedive. Toronto did mid-season, Colorado did post-season. Though I do agree with others on the problems being more the attitudes of the people promoting them. There is something of a logical disconnect between people who called on the Avs to fall off a cliff during last year's regular season to point to a first round playoff loss and proclaim "I WAS RIGHT!".

That being said, the Flames are very clearly an exception to the general rule. Travis Yost's article for TSN yesterday noted that of all playoff teams since 2007-08, this year's Flames would have the second worst Corsi against/60 (ahead of Toronto in the lockout year) and about the 15th worst Corsi for/60. Now, it would be nice to claim that we hit upon a formula that beats Corsi, but that is what Toronto and Colorado fans said last year.

On the balance of probabilities, if we're always chasing the puck, we're less likely to succeed. But for now, and for who knows how long, the Flames are riding some magic that hockey analytics does not currently understand.