Closing the Book on 2017-18

[The Fonz]

Originally Posted by gargamel

If you take so long setting up your shots that there are several bodies between you and the net, it's going to affect your aim. That's bad hockey, not bad luck.

I agree, completely.

The stats to me all point to the Flames, under Gulutzan, as playing such a slow game that the opposition was always in proper position. Their high danger chances weren’t dangerous? Because they couldn’t catch a goaltender out of position. Their shots kept going wide of the net? Because the opposing skaters had ample time to get in position and block the lanes to the net.

They were slow. The eye test more than confirmed it, and depending on how you interpret the stats, they sure seem to confirm it as well.

[Bingo]

Originally Posted by jammies

Finding more statistics that support your confirmation bias is not being "open to new ideas". You're presenting a variation of a failed case as if it's a fresh, intriguing slant on last season, but it's not - what would be intriguing is if you found some numbers that showed the Flames were about as bad as they should have been. What would be fresh is if you said, "In retrospect, those calling for the coach to be fired mid-season weren't at all premature."



Just because an idea is new, or seemingly objective, doesn't make it correct. A theory with no explanatory power IS useless, and data that supports no theorizing is equally so. It is incumbent on the person proposing the theory to show its validity, and not the skeptic's place to let error ride unchallenged.

You replied calling counting stats from an independent objective source wrong because it didn’t fit your bias

They can’t be wrong. They are what they are.

You’re welcome to explain why they don’t have value to you hit they’re not wrong.

Many felt the Flames were a perimeter team last year. These stats prove they weren’t.

I didn’t suggest Gulutzan’s system was vindicated in any way.

So odd reply in my opinion.


Sent from my iPhone using Tapatalk

[Bingo]

Originally Posted by monkeyman

I think when you rely on luck as a variable to explain the result of an equation you have to wonder about the science.
Surely there must be an exponent in the Flames game that would explain prolonged, historic "bad luck".
If you can't accurately predict the outcome over the long-run and your only explanation is luck, doesn't that defeat the whole purpose of using stats in the first place?

I’ve mentioned the lack of finishing skills on numerous locations


Sent from my iPhone using Tapatalk

[Bingo]

Originally Posted by mikephoen

Where did you get this 85% correlation number from? That's clearly not true at all.



Last year only 9 of the top 16 teams in 5v5 shot attempt differential made the playoffs. Three of the top 4 and 4 of the top 6 missed. Two of the bottom 5 made it, and the Stanley Cup champs finished 24th.

It’s been mentioned in a few articles. I’ll try to find them. However one season doesn’t usually give you enough data to come to conclusions


Sent from my iPhone using Tapatalk

[Bingo]

Originally Posted by mikephoen

Taking a look at this site Bingo likes, I checked out the HDCF% last year. I assume this is High Danger Corsi %. This stat did better than the 5v5 Corsi in my last post, as 10 of the top 16 teams made the playoffs. That's still not a great correlation, but it's slightly better. However, consider this. The WORST team in the leage at HDCF%? The Washington Capitals. Yeah, if the Stanley Cup champ is the worst in the league at this, then clearly it's value is highly suspect.

There’s always going to be exceptions for sure. The 2015 Calgary Flames were definitely a team that had terrible underline numbers but had good bounces and made it


Sent from my iPhone using Tapatalk

[Bingo]

Originally Posted by gargamel

That's not how you saw it back in 2015.







For you to be right now, you must have been wrong back then, but both the standings and the eye test say it's the other way around.







If you take so long setting up your shots that there are several bodies between you and the net, it's going to affect your aim. That's bad hockey, not bad luck.

What am I claiming to be right about now?

Opinions change as we evolve. That shouldn’t come as a shocker. I was barely scratching the surface on underlying stats 3 1/2 years ago.


Sent from my iPhone using Tapatalk

[Bingo]

Originally Posted by Resolute 14

The 14-15 Flames are an interesting case. Personally, I would not agree that they were frequently outplayed. They just had a severe case of LOFT. What made that season work was the fact that Hartley managed to get the entire roster pulling in the same direction, such that they weren't often outplayed.

Now, the number of late comebacks *was* decidedly outside the norm, and that obviously was unlikely to be repeated. But, how much of that was luck? How much was determination? How much was the team's relentlessness wearing down better, more skilled opposition?

They spent so much time in their zone that season blocking shots. It was a very uncomfortable wait to watch your team try and win


Sent from my iPhone using Tapatalk

[Bingo]

Originally Posted by The Fonz

I agree, completely.

The stats to me all point to the Flames, under Gulutzan, as playing such a slow game that the opposition was always in proper position. Their high danger chances weren’t dangerous? Because they couldn’t catch a goaltender out of position. Their shots kept going wide of the net? Because the opposing skaters had ample time to get in position and block the lanes to the net.

They were slow. The eye test more than confirmed it, and depending on how you interpret the stats, they sure seem to confirm it as well.

And that’s a very good theory. I have no problem with that compared to “ the stats are wrong “


Sent from my iPhone using Tapatalk

[SebC]

Originally Posted by GioforPM

GG coached a system that the majority of the coaches in this league, including the highly thought of ones like Babcock and Cooper, employ.

Enlighten me, please. What is this system? How does one identify it vs. the alternatives? Is Peters also in this majority? If not, what does he do differently?

=====================

IMO, the answer put forth by The Fonz is the correct one.

[jammies]

Originally Posted by Bingo

You replied calling counting stats from an independent objective source wrong because it didn’t fit your bias

My bias that the Flames were a non-playoff team who were no better than their record, as opposed to your bias that they were a top team in the league who merely lacked scoring finish? Or is it the bias I have against analysis that doesn't successfully analyze - as in, what are those stats illuminating for us and what actions should they elicit from the players and coaches?

You're like the manager of a sales team bragging about how his people are the tops in the business at timely expense report submission. Ok, sure, but no matter how many independent reports show that this correlates with moving product, it matters not at all if the team doesn't actually sell anything.

PS: Calling stats compiled by eye and interpreted by fallible human beings "objective" is droll.

[Bingo]

Originally Posted by jammies

My bias that the Flames were a non-playoff team who were no better than their record, as opposed to your bias that they were a top team in the league who merely lacked scoring finish? Or is it the bias I have against analysis that doesn't successfully analyze - as in, what are those stats illuminating for us and what actions should they elicit from the players and coaches?

You're like the manager of a sales team bragging about how his people are the tops in the business at timely expense report submission. Ok, sure, but no matter how many independent reports show that this correlates with moving product, it matters not at all if the team doesn't actually sell anything.

PS: Calling stats compiled by eye and interpreted by fallible human beings "objective" is droll.

Where did I call them a top team in the league though? If you're going to hang a bias on someone I think you need to use something I actually said.

There was a long running discussion about systems over the summer and just before it with the principal argument that they were a team that shot from everywhere and high high corsi but no real scoring chances.

New information came out that said they had a higher split with shots deemed high and medium danger than they did low.

This was my summation ... period.

All in all ... a team with bad finishers and or bad luck, but they were out playing the opposition.

So not sure why you took it so personally. I didn't throw Gultuzan a life line. I didn't say they were a top team.

You just went there on your own.

So no I'm not pumping up my sales staff, in fact I even suggested on numerous times they had trouble finishing (would equally terrible in sales).

And they're objective because they're not Saddledome staff, or Flames fans compiling the stats. They look at every team with the same amount of tabulation and do it league wide.

ob·jec·tive
/əbˈjektiv/Submit
adjective
1.
(of a person or their judgment) not influenced by personal feelings or opinions in considering and representing facts.

Doesn't mention fallible or infallible. They're human but they're not biased.

[GullFoss]

Originally Posted by Bingo

The Flames put themselves in the right spot often and didn't score. Treliving decided they didn't have the skill to finish or he wouldn't have reworked his forward roster.

But the perimeter team suggestions aren't really holding water any more.

I don't think the arguement I ever made was perimeter. It was speed of transitions and number of transitions. Gully hockey didn't create enough scoring off transitions because:

1) flames played slow in order to keep possession on the transition. They would give up the potential of a 3-on-2 or 2-on-1 because when they got the puck back they were too slow to transition.

2) too passive a forecheck and backcheck to create enough transitions.

Odd man rushes generate more goals than anything else 5-on-5. Speed and number of transitions drive the ability to generate odd man rushes. The flames didn't get enough. Prove me wrong and this is settled. Until then, it's not.

Look at the goals the flames have scored this year. How many are driven the a hard forecheck or hard back check that leads to a quick transition? A very large number. And the flames didn't have enough of them last year.

Two shots from the same high danger area can be completely different scoring chance probabilities depending on the play that generated the chance. If the system is biased towards generating low probability scoring chances in high danger areas, the data isnt comparable across teams and is thus flawed. Because the underlying assumptions will always be that the distribution within a single category is randomly distributed. And that's not necessarily the case.