Francis: Treliving filled with disappointment, anger after team's collapse

[Bingo]

Originally Posted by Itse

Well now that you asked...

http://thesportjournal.org/article/g...ockey-success/

I just don't get why the conversation keeps getting steered back to corsi and shots when I asked about scoring chances.

Not even remotely similar.

[Igottago]

Originally Posted by Bingo

Just for the record ...

does my suggested deep dive make me a good or bad executive?

After seeing you go ham with stats all year I actually have no idea what your opinion is on what action would actually improve the Flames. Outside of more analysis.

Let's say you are GM. You have all the data you are going to have right now. I'm the owner and I want to know your plan. What do you do?

BTW I'm not trying to be combative. I'm just curious.

[Itse]

Originally Posted by Bingo

Just for the record ...

does my suggested deep dive make me a good or bad executive?

It depends.

Is it reasonable to assume there's really something in the statistics that would make Treliving look something other than a mediocre coach at best? This isn't really saying it's obviously unreasonable (although I think it is, but that's just my opinion), just how I would answer that question on a more general level.

If not, then that deep dive is more likely produce false results that someone wills themselves into believing than they are likely to produce useful information. It also wastes time in a situation where there might be a race to snatch a certain top coach.

(I'm not saying there's a race either, I don't know how this works, but I think AV most likely will take at least a couple of weeks off before even thinking about signing on to a new team.)

In this particular situation though, all that number crunching should already be done, because the questions raised were obvious long ago. Adding the last few games into the data isn't going to change the results significantly.

So if you're the kind of executive that wants a deep dive into the stats, you're IMO kind of a bad executive if you waited this long to put your staff to work. In fact you should have most of the relevant numbers in December and just updated regularly after that.

[CorsiHockeyLeague]

Originally Posted by Itse

Is it reasonable to assume there's really something in the statistics that would make Treliving look something other than a mediocre coach at best?

No, but there's something in his job title that would.

[Itse]

Originally Posted by CorsiHockeyLeague

No, but there's something in his job title that would.

Got me (Meant GG, obviously.)

[Textcritic]

Originally Posted by Igottago

After seeing you go ham with stats all year I actually have no idea what your opinion is on what action would actually improve the Flames. Outside of more analysis.

Let's say you are GM. You have all the data you are going to have right now. I'm the owner and I want to know your plan. What do you do?

I know you didn't direct this at me, but I think this is where a lot of people are talking past one another. Some of us are reticent to offer an opinion because honestly, we are still not sure what the right solution is. The problem with setting ourselves in the position of the GM and making a decision based on the paltry amount of information we have is precisely why I, for one, am not keen to offer a firm verdict—I simply do not feel that ANY ONE OF US behind our computer screens has the information necessary to make a decision.

We all have instinctive responses to why the Flames underperformed this year—some of us are happy to launch to a conclusion based on those instincts; others don't really trust them.

[Bingo]

Originally Posted by Igottago

After seeing you go ham with stats all year I actually have no idea what your opinion is on what action would actually improve the Flames. Outside of more analysis.

Let's say you are GM. You have all the data you are going to have right now. I'm the owner and I want to know your plan. What do you do?

BTW I'm not trying to be combative. I'm just curious.

Haven't I said ... I don't know a dozen times that I'd deep dive on the scoring chance counts and see if I could pinpoint why they run up totals without goals.

Get an idea of why.

If it's system I fire the coach. If it's a lack of execution then i look at the players not executing and determine how much is an off year, odd year, or a lack of skill.

[ricardodw]

Originally Posted by Bingo

First off I've talked eye test all season. I outline the marriage or complete diversion of stats to the eye test when I see a correlation or a break down. I'm not a one trick pony at all.

However getting the analytics department to dig deeper isn't to stare at numbers all day. The numbers are in.

The next level is to pick out a few games with high scoring chance counts and go back and watch the video, categorize them into times they were stoned by a goaltender, or had a recording of a high danger chance that wasn't all that dangerous.

Do that for enough games and get a percentage of actually dangerous dangerous.

Then look at Tampa Bay, Boston, Toronto ... get their percentages as well.

Then you may have a systemic issue that allows them to have a pop gun offence despite good counting stats.

I would hope that the teams have better stats input than the general public.

A shot from a spot on the ice does not make it a scoring chance.

The same shot with a screen in front or a goalie or the goalie moving from the other side of the net is far more of a scoring chance than a clear blast from the exact same spot with the goalie set.

A one timer from the same spot is a higher scoring chance than accepting a pass and then shooting.

An shot while being contacted by the opposition would be a lesser scoring chance.

A shot forced to one side of the net by defense positioning would be a lesser scoring chance.

A clear shot with the goalie set and some one driving the net would be a better scoring chance.

A quick release shot from the boards along the icing line with someone crashing the net from the others side is a good scoring chance.

A shot to the inside of the goal post is a much better scoring chance than hitting the Goalies crest.


just saying a video analyst accounting for all these variables would do a much better categorization of shot quality than simply taking the point from where a shot was taken.


I have to believe that teams already have a much better input to fancy stats than we the fans get to use. They would not be open to sharing as it would be a competitive advantage having a better stats department.

Teams would use there own stats rather than over-value the Flames players that have put up great corsi numbers using the generally available stats.

[Itse]

Originally Posted by Textcritic

We all have instinctive responses to why the Flames underperformed this year—some of us are happy to launch to a conclusion based on those instincts; others don't really trust them.

I don't think it's really fair to call them "instincts". I mean sure, we have only very partial information, but there's a lot of very useful information available to fans these days. Not as much as a GM, but it's very far from nothing.

Also, the problem with wanting more information is that you can never have all of it. At some point you just have to decide you have enough to make a call.

We just have different points where we think we have enough.

[Bingo]

Originally Posted by Itse

It depends.

Is it reasonable to assume there's really something in the statistics that would make Treliving look something other than a mediocre coach at best? This isn't really saying it's obviously unreasonable (although I think it is, but that's just my opinion), just how I would answer that question on a more general level.

If not, then that deep dive is more likely produce false results that someone wills themselves into believing than they are likely to produce useful information. It also wastes time in a situation where there might be a race to snatch a certain top coach.

(I'm not saying there's a race either, I don't know how this works, but I think AV most likely will take at least a couple of weeks off before even thinking about signing on to a new team.)

In this particular situation though, all that number crunching should already be done, because the questions raised were obvious long ago. Adding the last few games into the data isn't going to change the results significantly.

So if you're the kind of executive that wants a deep dive into the stats, you're IMO kind of a bad executive if you waited this long to put your staff to work. In fact you should have most of the relevant numbers in December and just updated regularly after that.

You guys are aware that my deep dive is into video to break down what's actually happening right?

[Itse]

Originally Posted by Bingo

You guys are aware that my deep dive is into video to break down what's actually happening right?

I'm not sure I've seen you lay it out like that? In any case, I still think that analysis should be largely done at this point, being that the signs of trouble have been around for months.

[Lanny_McDonald]

Originally Posted by Bingo

Just for the record ...

does my suggested deep dive make me a good or bad executive?

First off, I wouldn’t consider you an executive. I’d classify you as one of those data nerds. You’d be an informer and not a decision maker. A valuable member of the organization, but one that loves to get lost in the data. You’re our Tyler Delow.

The deep dive has been on-going and with the Flames’s season ending about a month ago the data should have been in a long time ago. Suggesting to keep looking at something that has been crunched all year, and has been analyzed to death, is to suggest further gridlock in an organization already jammed up. More time reviewing the same tired faulty statistics is not coming to reveal something that isn’t already known. At this point we’re doing nothing more than building a case for covering of asses.

[Textcritic]

Originally Posted by Itse

I don't think it's really fair to call them "instincts". I mean sure, we have only very partial information, but there's a lot of very useful information available to fans these days.

Also, the problem with wanting more information is that you can never have all of it. At some point you just have to decide you have enough to make a call. Not as much as a GM, but it's very far from nothing.

We just have different points where we think we have enough.

I think this is exactly right, but I also do not believe that even with the great tools available to us in the public we are ever getting enough to make any sort of call, and that is what separates us as fans from NHL GMs.

I am not suggesting that the NHL GM is infallible. What I am suggesting is that the extremely vitriolic and exasperated responses to the fact that Treliving has not made a coaching change yet are absurdly shortsighted.

[Bingo]

Originally Posted by Itse

I'm not sure I've seen you lay it out like that? In any case, I still think that analysis should be largely done at this point, being that the signs of trouble have been around for months.

Me suggesting it doesn't mean they haven't already done it.

That ultimately will be the answer on the coach though.

And yes I've laid it out numerous times.

[Bingo]

Originally Posted by New Era

First off, I wouldn’t consider you an executive. I’d classify you as one of those data nerds. You’d be an informer and not a decision maker. A valuable member of the organization, but one that loves to get lost in the data. You’re our Tyler Delow.

The deep dive has been on-going and with the Flames’s season ending about a month ago the data should have been in a long time ago. Suggesting to keep looking at something that has been crunched all year, and has been analyzed to death, is to suggest further gridlock in an organization already jammed up. More time reviewing the same tired faulty statistics is not coming to reveal something that isn’t already known. At this point we’re doing nothing more than building a case for covering of asses.

Talking about the timing is irrelevant.

It might already be done. So timing is moot. The key is the fact that it has to be done in order to really get a handle on what this team is. If it comes back that the system if flawed because their chances are actually chances at a rate well behind top ten teams than you know the system is the issue and you act accordingly.

And you you have me all wrong. I make decisions instantly every day with the numbers and analysis automated and crunched to make me informed and prepared to do so.

[Igottago]

Originally Posted by Bingo

Haven't I said ... I don't know a dozen times that I'd deep dive on the scoring chance counts and see if I could pinpoint why they run up totals without goals.

Get an idea of why.

If it's system I fire the coach. If it's a lack of execution then i look at the players not executing and determine how much is an off year, odd year, or a lack of skill.

So, more analysis lol.
Meanwhile AV gets hired by another team.

You're fired.

Again. Joking. I guess I just don't see why its so hard to take a stab at what you think might be the way to go. That's the fun part of a hockey message board. None of us is getting hired or fired by what we post here.

[CorsiHockeyLeague]

Originally Posted by Bingo

Carolina had terrible goalending. They're an easy solve.
The Blackhawks lost their goaltender for most of the season.
The Habs had a down year from Carey Price.

The Flames stand alone as the hardest club to figure out.

We've been talking about this stuff for years, if people still don't understand it it's because they're determined not to.

Possession correlates to success. Scoring chances correlate to success. There are a number of caveats that can change this:

• Goaltending performance - The biggest factor in team success isn't accounted for by these numbers.
• Shooting talent - Kucherov needs fewer shot attempts than Stajan to score.
• Special teams - Possession and scoring chance stats are even strength stats.
• Score effects - unless you control for them, out-corsiing someone when you're down 5-1 does not indicate a quality performance from your side.
• Puck luck - It's obvious that "the bounces" are a thing; I have no idea why people don't want to admit this is at least one of a number of factors.

And other intangible things. People are emotional creatures, so sure, leadership matters. But the thing that matters most to success in today's NHL is skill. Generating possession, generating scoring chances, these are skills that lead to goals, and goals lead to wins. That does not imply that they are the only things that lead to wins, or that you're guaranteed a win if you out-chance the other team. But if you do it consistently, you're going to put yourself in a better position than if you don't.

Just by way of example, NHL average shooting percentage in all situations was 9.3% this year. If you outshoot the other team consistently, and both sides score on about 9.3% of their shots, you're going to end up winning more often than not. If not because you're consistently scoring on less than 9% of your shots, the issue may be a lack of finishing talent - see Canadiens, Montreal. If you're not because the other team is consistently scoring on more than 9.3% of their shots, the issue may be goaltending - see Hurricanes, Carolina. That's not to say other factors aren't involved (power play proficiency, for example), but this is information you can use to start to diagnose what happened. And none of this is rocket science.

[Table 5]

For me, there's enough data, combined with plenty damning eye-test evidence, to at least say that Gulutzan needs to go. The last month of hockey has only hurt his chances.

The only thing I'm analyzing now is how to move forward. Who do I hire next and what player do I want to trade out (or trade in).

[Lanny_McDonald]

Originally Posted by Bingo

And you you have me all wrong. I make decisions instantly every day with the numbers and analysis automated and crunched to make me informed and prepared to do so.

So no deep dives required. Good to know.

[EldrickOnIce]

Originally Posted by Fuzz

Technically, yes that sounds right. The reality is(as the Flames have demonstrated) there is more going on. Otherwise they would have done better this season. And I think it has a lot more to do with uncountable things than the countable ones. Some would argue their is a stat that is going to show why the Flames did bad. My whole point is there isn't one, and wasting time looknig for that number isn't really productive.

Now, if they want to pour over the tapes and find all the times the team collapsed due to the wrong players on the ice, or not calling a time out when it could have helped the players regroup, or players looking unmotivated, or a system that breaks down or has been figured out by the opposing team, then ya. Do that. Because that's where I think the team failed, and you won't find Corsi stats that explain it.

I don't think anyone is saying the solution is in the numbers. But
-are consistently out playing opponents.
-top players are thriving, or have, in the system

So the simple answer, on the surface, appears to be simply a lack of talent.
That said, I will be happy to see Gulutzan go, but I will be happier to see the addition of some real talent.

I could see this team winning with a couple more talented forwards. Maybe even under Gulutzan, I dunno. I can't see any coach winning with this bottom 6 talent.