Not to worry; this won't last says Myrtle

[Arya Stark]

Originally Posted by foshizzle11

What about a stat like scoring chances? Or quality of scoring chances per game. Does that come into play and how are the flames compared to the teams who dominated them this season?

Not advanced enough for the advanced stat guys. Too hard to push your agenda this way.

[indes]

Originally Posted by Arya Stark

Not advanced enough for the advanced stat guys. Too hard to push your agenda this way.

Or even the media won't turn to extremely subjective stats to push their "agendas".

[BACKCHECK!!!]

Originally Posted by nik-

They have the best player in the world dude. And another who's at least top 5.

Yes, and the Bruins' stretch of exceedingly strong PDO is because they had Thomas and Rask pushing up the sv% side of things.

And crappy teams have low PDOs year after year because they have crappy goalies that struggle to post a .900 sv%, and few high-end goal scorers with strong shooting %.

Which is to say its not just a measure of "luck".

variations away from 100 are not purely a matter of chance, and teams routinely sustain PDOs at the top and bottom of the league for entire seasons, or even several seasons.

[_Q_]

If you take the PDO for the Oilers from the last 5 years, advanced stats say that they should be hitting a PDO of 2.0 pretty soon.

[Anduril]

fata'ed

[Iggy Snipe]

Its a game by game approach in this league. Teams make changes on a daily basis. That article is BS. Toronto sucked down the stretch because they suck. No leadership on that team.

The flames are fun to watch and are getting rewarded for it. Its nice to see. The is no reason to suggest the wheels are all of a sudden going to fall off. They have earned their spot in the standings. They will either keep earning or fall out out of the playoff picture.

The offence is being spread so widely on the team. People are looking at that like its a negative? All the good teams get scoring from all lines, and defenceman contributing. I rather have Calgarys team today,with no stars, than a team with one or two guys who carry the load (Chicago).

It is very exciting watching the Flames this year. No one ever please compare the Flames to Toronto. They have nothing in common except they both play in Canada.

[Vinny01]

Remember when the Flames opened the Brent Sutter/Jay Bouwmeester era? That team was considered to be a team that would never score goals and they dangerously spent too much money on their backend. They started that tear 25-12-4 if I am not mistaken but the wheels started to fall off in December and completely went off the rails in January which resulted in the first of 5 playoff misses.

That season it seemed like the Flames were playing pond hockey and were rewarded for it. They scored at will especially on the power play but it eventually dried up and they ended bottom 5 in goals for. That team was also badly outshot in the early going so much so I think their advanced stats have to be worse than this years Flames. Unless we have an injury to Gio or Brodie I don't see this years team falling off that badly. They could stumble and miss the playoffs but I think we are looking at drafting 9-16th this year instead of top 5. I am okay with that

[waner]

Originally Posted by Iggy Snipe

I rather have Calgarys team today,with no stars, than a team with one or two guys who carry the load (Chicago).

Who are the one or two guys?

Toews, Keith, Kane, Hossa, Sharp, Seabrook are all top level guys that would be huge improvements on any team in the league.

Saad, Shaw, Bickell, Hjarmalsson, Crawford, Oduya are all very good secondary players that would easily play for any other team in the league.

I know people are in love with the Flames but I think I will take the team with the stars that wins Cups and goes to Conference finals.

[Street Pharmacist]

Originally Posted by Vinny01

Remember when the Flames opened the Brent Sutter/Jay Bouwmeester era? That team was considered to be a team that would never score goals and they dangerously spent too much money on their backend. They started that tear 25-12-4 if I am not mistaken but the wheels started to fall off in December and completely went off the rails in January which resulted in the first of 5 playoff misses.

That season it seemed like the Flames were playing pond hockey and were rewarded for it. They scored at will especially on the power play but it eventually dried up and they ended bottom 5 in goals for. That team was also badly outshot in the early going so much so I think their advanced stats have to be worse than this years Flames. Unless we have an injury to Gio or Brodie I don't see this years team falling off that badly. They could stumble and miss the playoffs but I think we are looking at drafting 9-16th this year instead of top 5. I am okay with that

On January 12, 2010 the flames were 26-14-6. Their underlying stats suggested they are somewhat fortunate, but nowhere near as much as this team.

2014/15 Flames: PDO 102.74 Corsi % 45. 46
2009/10 Flames: PDO 101.59 Corsi% 50.40

That team actually outshot other teams overall, albeit barely

[Henry Fool]

Pointing out an usually high shooting percentage and low number of shots is fair enough. Beyond that it appears to me that a lot depends on possession numbers and what they mean for the team. It's a little unclear for me at least.

I do agree that shooting percentage is likely to fall. We must remember though that teams react to adversity. What matters is whether or not they'll be able to adjust. You can't blame them for relying on whatever is currently making them successful. If it means fewer shots on goal right now, there's no reason why they won't start putting more pucks to the net once the easier opportunities aren't there anymore.

Goaltending doesn't need to go downwards. There might be a sense around the league that Hiller has been standing on his head the entire season which isn't true, and also Rämö has yet to reach the level he played at last season.

As a thought experiment for the statistically inclined, what stats would you use if you set out to prove that the Flames play is actually sustainable rather than the opposite? Are there such stats or does everything indicate that they will fall back?

For instance, Colorado last year has been mentioned as a team that went against the trends. In hindsight have people come up with underlying stats that tell us why they did? And furthermore, how do the current Flames compare with last season's Avs, both statistically and qualitatively? Are there similarities in their styles? Do they both, for example, score off the rush more than average (as opposed to as a result of sustained offensive zone pressure, which might have implications for possession numbers and quality of scoring chances)?

[Henry Fool]

It would be interesting to tie the stats with statements about different teams' play styles. 'Possession' just by itself is very abstract and crude, for example. But there's a reason why that type of qualitative analysis is extremely rare: it takes real expertese to make concrete statements about how a team plays. Make a generalisation about a team's style based on your observations and you risk making a fool out of yourself if you're not confident in your ability to judge that. On the other hand, there's no risk in quoting stats and saying that it means the team "drives possession" or "controls play". Safe abstractions.

[T@T]

Did Myrtle really say "Not to worry"? like as to say you can bet your life savings they'll suck like I said they would.

How about go Fv*k yourself Myrtle.... you eastern bum and creep.

[BACKCHECK!!!]

@Henry Fool

Colorado was good because Varlamov was amazing, and because their defence scored a prodigious (and completely uncharacteristic) amount of points.

Personally, I think Varlamov is one of the most talented goalies in the NHL, and so while he had his best year to date, I don't think that was particularly unsustainable. He hasn't been as good this year, but I don't think that was inevitable, and I think we'll see Vezina-contending seasons from him in the future.

However, the production from their defence was unsustainable, as IIRC it relied on career years from basically everyone, including those who had never been NHL producers before.

Regarding the Flames, if I were to pick a stat that showed the Flames actually were doing well despite poor Corsi/possession numbers, I'd want to see a strong Fenwick Close%. However, Calgary is 23rd in the NHL in Fenwick Close.

Realistically, all of the useful stats that i trust are predicting that Calgary should be a 20th-25th place team.

Sometimes teams buck the trend, sometimes their style of play means that (for example) their Corsi doesn't reflect their possession time, or their possession time doesn't reflect their ability to control play. Sometimes you can pick this out with other stats, sometimes not.

But pretty much the entire advanced stats picture says the Flames are winning more games than they deserve to win.

You can take what you want from that. I am a very big advanced stats proponent, and I honestly don't know what to make of the whole situation. By eyeball, the team looks good to me. The numbers suggest they are going to fall off hard.

If I had to lay money, I'd side with Myrtle, and bet they finish in the bottom 10. But I'm fine with just about any outcome.

[Street Pharmacist]

Originally Posted by BACKCHECK!!!

@Henry Fool

Colorado was good because Varlamov was amazing, and because their defence scored a prodigious (and completely uncharacteristic) amount of points.

Personally, I think Varlamov is one of the most talented goalies in the NHL, and so while he had his best year to date, I don't think that was particularly unsustainable. He hasn't been as good this year, but I don't think that was inevitable, and I think we'll see Vezina-contending seasons from him in the future.

However, the production from their defence was unsustainable, as IIRC it relied on career years from basically everyone, including those who had never been NHL producers before.

Regarding the Flames, if I were to pick a stat that showed the Flames actually were doing well despite poor Corsi/possession numbers, I'd want to see a strong Fenwick Close%. However, Calgary is 23rd in the NHL in Fenwick Close.

Realistically, all of the useful stats that i trust are predicting that Calgary should be a 20th-25th place team.

Sometimes teams buck the trend, sometimes their style of play means that (for example) their Corsi doesn't reflect their possession time, or their possession time doesn't reflect their ability to control play. Sometimes you can pick this out with other stats, sometimes not.

But pretty much the entire advanced stats picture says the Flames are winning more games than they deserve to win.

You can take what you want from that. I am a very big advanced stats proponent, and I honestly don't know what to make of the whole situation. By eyeball, the team looks good to me. The numbers suggest they are going to fall off hard.

If I had to lay money, I'd side with Myrtle, and bet they finish in the bottom 10. But I'm fine with just about any outcome.

This! ^^


If I was going to make a case for sustainability, the only 2 things I could say are a) the shooting % and save percentage while well above average is due to high talent levels, or b) stats show they're improving (albeit only slightly). Neither of those really hold water. Can't say Hiller is better than his career average suddenly with such a small sample size and with so many rookies a high shooting % because of talent is dubious at best. As for the improvement, it's there but it's still not suggestive of a playoff team.

Since 2005/06, teams have only finished a season with the same or worse corsi% than the current flames 14 times:

13/14 Buffalo 43.39% 30th overall
07/08 Atlanta 43.40% 28th overall
13/14 Toronto 43.63% 23rd overall
12/13 Edmonton 44.32% 24th overall
13/14 Edmonton 44.65% 28th overall
11/12 Minnesota 44.67% 24th overall
10/11 Anaheim 44.79% 9th overall
09/10 Florida 44.85% 28th overall
05/06 Pittsburgh 44.87% 29th overall
09/10 Edmonton 44.93% 30th overall
12/13 Buffalo 45.01% 23rd overall
12/13 Toronto 45.03% 9th overall
09/10 Colorado 45.18% 12th overall
10/11 Minnesota 45.70% 21st overall


That's 3 playoff teams in the bottom 14 of all the last 270 team-seasons played. More than I thought

[wwkayaker]

When did this advanced stats movement begin? I feel pretty humble having read this thread.

[Street Pharmacist]

Originally Posted by wwkayaker

When did this advanced stats movement begin? I feel pretty humble having read this thread.

Most of this has been going on in the background for years already. Flames have had a guy for a few years now and only recently talked about it

[RyZ]

Are the numbers improving as the season has progressed? Are games 12-18 yielding similar numbers as games 1-6 or is the team getting better as times has gone on?

What were the teams numbers last season in comparison?

[wwkayaker]

Originally Posted by RyZ

Are the numbers improving as the season has progressed? Are games 12-18 yielding similar numbers as games 1-6 or is the team getting better as times has gone on?

What were the teams numbers last season in comparison?

To build on this^, when looking at PDO, for example because I just read about it, what sample size would give an idea of a team's direction?

[BACKCHECK!!!]

Originally Posted by wwkayaker

To build on this^, when looking at PDO, for example because I just read about it, what sample size would give an idea of a team's direction?

PDO doesn't tell you what's going to happen, only what has happened.

That's important because its easy to see what's causing a high PDO.

Im half drunk and on my phone, so i dont have the numbers in front of me, but just as sxamples, if you have a high PDO because youve got 2009's Roberto Luongo in net posting sideways save percentages, and the Sedin twins up front scoring at will, you can say "We have a wicked high PDO because we're awesome".

And that is totally sustainable, for as long as your major contributors are at their peak.

I recall Calgary posted some crazy high PDO numbers a few years ago, but IIRC that was based on Kipper having a crazy good year, and players like Glencross and Tanguay who have had exceptionally high shooting percentages for their entire careers. In that case we didn't sustain it, but there was nothing inherently unsustainable about it.

In both those cases, the goalie SV% and player shooting % were well within what we had seen from several previous seasons. This was players who had previously posted high- end numbers all posting high-end numbers at the same time.

This year's high Flames PDO is based on Hiller (who is likely past his prime) posting numbers which are way better than he has ever sustained before. And guys like Raymond and Wideman posting shooting percentages that are way higher than they can possibly maintain, along with Giordano scoring like his hair's on fire.

None of the things that are driving the Flames' high PDO are a result of players playing at their "normal" level. Giordano is not going to finish with 90 points, and Hiller won't end the season with a .950 even-strength save%.

So there is no sample size for PDO that tells you where a team is going. Without looking at the roster, and the reasons behind a high or low PDO, it doesn't tell you anything. Like +/-, its basically meaningless on its own, it just tells you that a team is scoring more than the other team, with no indication as to why.

[BACKCHECK!!!]

Just to elaborate a tiny bit, you can't look at a PDO and say "this team is going to regress", because the good teams don't regress.

And you can't look at a low PDO, and say the team will improve, because crappy teams can sustain low PDOs.

What you CAN do is look at a team like the Flames, and say "They will regress because they suck." The PDO in this case just explains why they've been able to win some games.

Or you look at LA and say they will regress back into a playoff spot, because they are awesome. Their poor PDO only explains why they have been out of a playoff spot.

In the case of LA and Calgary, it is (probably) mostly luck. Their PDO is the opposite of what their roster should be doing. That suggests they will regress

The Pens and Oilers on the other hand have PDOs in line with their expected abilities. They will probably not regress.

But in all cases, we're just saying "the good teams will be good, and the bad teams will be bad". The PDO does not predict what will happen in any of these cases.