Vancouver led the league in points at Christmas, Can they continue this run into the New Year?

[TheOrangeDesk]

As a rule, ALL teams that have a hot start inevitable see a reduction in a stat like PDO as the season progresses and it has nothing to do with "regression to the mean". It has to do with the wear and tear of the season. The first month of the season a team is relatively healthy. The last month, everyone is sore. During the season, there are injuries, personnel changes, opponent adjustments, lead protection schemas/"greasy road games", trades/callups, playoff clinching, and other variables that a shit stat like PDO doesn't account for, but all contribute to a team's production efficiency.

NHL teams are not video game teams, where you get the same performance out of them night in and night out. The cumulative wear and tear of a protracted season has an attritional effect on every team, such that the team at the end, the one that hoisted the Cup, is not the most skilled or best coached team, but simply the team that outlasted the others. PDO has no model for determining which team this is.

Pdo between competing teams averages to 100 every game. All year. every team getting sore as the year goes on does not result in pdo dropping. Your theory just shows you don't understand what pdo is

 

[Bank Shot]

Overly rigid misconceptions and stat-watching aside, I am very curious about where specifically this scenario and past scenarios differ/match, and I think there can be a really interesting discussion about that.

Anyone actually watching the Canucks games can observe that they naturally play a style that would specifically result in a very high PDO. They play a highly skilled and structured puck possession game where they have the puck most of the time but they'll never waste a shot unless it's a high danger scoring chance. They'll always hold onto the puck and regroup while maintaining puck possession if they don't see the scoring chance.

You can also argue that their defensive approach affects save% in the same way, being highly successful at getting goals early and then doing a good job of turtling and preventing high danger scoring chances, but in a way that allows the opposition to take as many low% shots as they want.

My question is, did past high-PDO teams also have high PDOs for that same reason, but something about that style of play itself is truly unsustainable (and if so, what specifically ended up being the reason that made it unsustainable?), or did they play no differently from everyone else and just happened to have high shooting % as a random stroke of luck?

If it's the latter, why HAVEN'T there been more high-skill teams approaching their game this way?

Fans of teams on extreme heaters always say stuff like this. "Our style allows us to finish at a higher percentage, or our turtling allows us to maintain a high save percentage."


There can be a small amount merit to that, but save and shooting percentages tend to stick to pretty tight ranges over the years and when one or both of them are over historical ranges, well its a smart play to bet against the trend continuing.

Generally most NHL teams have pretty similar structures. No one is really doing anything revolutionary out there for the most part. I mean the majority of NHL coaches bounce around from team to team for 15-20 years.

Sometimes coaches figure out a new wrinkle, like the drop pass for PP entries that is so popular now. Other teams quickly adapt to that and then it becomes common place, so no team really carries any kind of tactical edge for any length of time.

It is human nature to downplay the chances that get saved and put special emphasis on the scoring chances that go into the back of the net. If it is your favorite team, well you are way more likely to fall into this trap of thinking your team is playing in a way that allows them to beats the odds.

I think it was maybe 2015, when the Flames led the league in third period comebacks, a fairly large proportion of their fanbase convinced themselves that Bob Hartley ran such hard practices that the Flames were better conditioned than other NHL teams and that allowed them to have more in the tank in the third. Therefore the 3rd period comebacks were a feature rather than just random variance.

Well it didnt carry over into the next season, and Hartley got canned.

 

[Sykur]

every team getting sore as the year goes on does not result in pdo dropping.

Yes it does.

Quite easily, a star player injured or reduced performance due to fatigue/soreness will absolutely cripple PDO.

And since injury/fatigure/soreness is not equitable to every player, every team, across the league, makes it the ultimate variable that PDO doesn't account for.

That's what makes it a shit stat.

 

[TheOrangeDesk]

Yes it does.

Quite easily, a star player injured or reduced performance due to fatigue/soreness will absolutely cripple PDO.

And since injury/fatigure/soreness is not equitable to every player, every team, across the league, makes it the ultimate variable that PDO doesn't account for.

That's what makes it a shit stat.

Do yourself a favor and average every nhl team pdo and see what number you get. You may be in for quite a surprise

 

[Sykur]

Do yourself a favor and average every nhl team pdo and see what number you get. You may be in for quite a surprise

We're not talking about the whole NHL, we're talking about individual NHL teams.

Try to have some focus here.

 

[Bank Shot]

Yes it does.

Quite easily, a star player injured or reduced performance due to fatigue/soreness will absolutely cripple PDO.

And since injury/fatigure/soreness is not equitable to every player, every team, across the league, makes it the ultimate variable that PDO doesn't account for.

That's what makes it a shit stat.

A goalie injury could cripple PDO.

A skater probably not.

Seattle didn't even have any star forwards when they led the league in 5v5 shooting percentage last season.

Standings points are also affected by injury. Are standings points a shit stat because they don't account for the ultimate variable of injuries?

 

[TheOrangeDesk]

We're not talking about the whole NHL, we're talking about individual NHL teams.

Try to have some focus here.

Honest question. Do you know what pdo is? You seem to think it's a measure of ability? It's not. Mcdavid, the best player in the league has a 5on5 pdo of 100. Dead average. Kind of blows a hole in your argument.

A goalie injury could cripple PDO.

A skater probably not.

Seattle didn't even have any star forwards when they led the league in 5v5 shooting percentage last season.

Standings points are also affected by injury. Are standings points a shit stat because they don't account for the ultimate variable of injuries?

Even so. The leafs have had arguably the worst goalie injury luck this year and their pdo is still above 100.

 

[Hammman]

Yes it does.

Quite easily, a star player injured or reduced performance due to fatigue/soreness will absolutely cripple PDO.

And since injury/fatigure/soreness is not equitable to every player, every team, across the league, makes it the ultimate variable that PDO doesn't account for.

That's what makes it a shit stat.

It is statistically impossible for everyone's PDO to go down. If one team's PDO drops, other teams will rise a commensurate amount.

If you're just suggesting that all teams with a high PDO will drop in the second half because they get worn down, that makes no sense; the teams below them in the standings are likely to be even more worn down - that's why they're lower in the standings.

This is not to say that regression over time isn't likely, but your argument makes no sense.

 

[Bank Shot]

Even so. The leafs have had arguably the worst goalie injury luck this year and their pdo is still above 100.

Well yeah. Thats PDO for you.

Generally though teams with star goaltenders tend to outperform on PDO. So when I look at Vancouver's PDO, the goaltending isn't the unsustainable part of the equation. Demko is great.

It is the shooting % which is the area I think won't sustain.

 

[Fishy McScales]

Canucks fans debunking PDO while learning about PDO is adorable.

 

[SK13]

You don't even need to reference PDO or analytics.

Meat and potatoes shooting percentage tells the tale. The Canucks have the highest shooting percentage since the NHL started tracking shot metrics in 2009. They have the largest shooting percentage since 1993, and the #43 highest shooting percentage recorded over 104 NHL years.

The next highest shooting percentage in the modern era was the 21-22 Blues, who are #156. They lost 50 goals and 29 points on the standings the next year.

 

[andora]

Has anyone broken this dumb stat down in the game states? Or found where it's readily available? Like either by period or by when they are tied winning or losing

 

[Dread Clawz]

They pass the eye test for me. They surely are an inexperienced playoff team, but otherwise they are a very strong team. Pettersson is on a mission this season and Hughes should win the Norris. Demko might win the Vezina. They really remodeled the blue line quickly, very stable D. Zadorov was a great pickup for what they got him for.

 

[Nucker101]

As someone who's into analytics, I'm fully willing to admit this run definitely has some luck factored into it, PDO luck and even injury luck as well(Soucy has really been the only big significant injury this season so far). But at the same time the PDO crowd is overlooking some things:

1. Canucks average a league high 24.5 mins per game at 5v5 with the lead, guess what score factors do to corsi/xgf? For context, the next highest team is LA and they're a distant 2nd at 20.5 mins and the league average is around 16 mins. And before you throw score adjusted numbers at me, I find those formulas to be flawed, sure they can help but only so much.

2. Using a PDO baseline of 100 for every team is just a lazy/flawed way to analyze teams, it's the classic analytical safety net argument of "well I can't predict it with the public data available to me so it must be luck!". The fact is that that some teams do have better goaltending, some teams do defend against cross ice passes and off the rush better. Some teams do a better job of screening opposing goalies on their shots, some teams do have more gifted shooters and playmakers that generate higher quality shots for their teams. The Blues had a run for about a couple of years with their on-ice sh%'s were pretty high, then you dig deeper and realize they were also one of the best teams in the league at high danger passing and cross ice passes, so rather than 100+ games of luck, it's much more likely that they passed up on low danger shots and instead preferred trying to get goalies to mvoe laterally before shooting the puck. Meanwhile teams like Carolina will shoot from anywhere. There's no way in hell I'm going into a season expecting the Canucks and Canes to finish with similar on-ice PDO with the way both teams are built.


I think once public data improves, we'll improve on xGF models and get better at projecting teams' baseline PDO. Just like how we improved on corsi with xgf in the first place, it wasn't too long ago when Corsi(or Fenwick) was king and teams/players were getting called lucky/unlucky until xgf helped clarify some of that. I fully expect we see xgf models improve in the next few years and some of these "luck" explanations will go away.

 

[jackjohnson]

They have almost SEVEN 20% shooters when most teams have 0/1/and maybe 2 teams have two.....It's been a great year for them so far. I do not see it like this next year. It's is impossible to keep doing that.

Your posts are getting so boring when you keep repeating or copy and pasting your post from previous post. Come up with some new materials

 

[Jay26]

We're still talking about PDO? Who cares?

Yeah, the scoring is going to regress to the mean a bit by the time the playoffs roll around but it really doesn't matter because the Canucks have proven by this point that they're a legitimately good team. That's all I care about. I've realized I have no interest in doing a deep dive statistical analysis of this team. They may win it all, they may fall just short, or they may flame out in round 1; either way I'm not going to be saying "Gee the guys on HF were right about the PDO" when it's all over.

 

[Regal]

You don't even need to reference PDO or analytics.

Meat and potatoes shooting percentage tells the tale. The Canucks have the highest shooting percentage since the NHL started tracking shot metrics in 2009. They have the largest shooting percentage since 1993, and the #43 highest shooting percentage recorded over 104 NHL years.

The next highest shooting percentage in the modern era was the 21-22 Blues, who are #156. They lost 50 goals and 29 points on the standings the next year.

They also lost Perron in the off season and traded RoR, Tarasenko and Barbashev in the middle of the season. They still shot 11.1% the following year, but also had fewer shots and much worse defense/goaltending (they gave up 59 more goals). The Canucks goaltending isn’t likely to regress as Demko has regularly been at this level, and if their shooting percentage was at 11% like the Blues the following year (or what the Canucks shot in 22-23), they’d still be middle of the pack in GPG and have a decently positive goal differential. Regression should still leave them a playoff team.

 

[jackjohnson]

Canucks don't play good in these early games. Hate this schedule

 

[SK13]

They also lost Perron in the off season and traded RoR, Tarasenko and Barbashev in the middle of the season. They still shot 11.1% the following year, but also had fewer shots and much worse defense/goaltending (they gave up 59 more goals). The Canucks goaltending isn’t likely to regress as Demko has regularly been at this level, and if their shooting percentage was at 11% like the Blues the following year, they’d still be middle of the pack in GPG and have a decently positive goal differential. Regression should still leave them a playoff team.

Demko is a good goaltender and there's nothing fishy about a .916 save percentage from him.

It's the offensive numbers that are as egregious as I've ever seen it in the NHL. Relative to the rest of the league, which is the point of sustainability here, you have to go back like 4 decades to find a stat as stupid as Vancouver at 13.8%.

If you regress them to 12.2%, which would STILL be unsustainable and STILL be #1 in the NHL this year, they'd lose 20 goals in 40 games played. That's like 8 points in the standings alone, and it doesn't address how GF going down often equates to GA going up. Less offensive confidence, less conservative hockey protecting leads, etc.

They should still be a playoff team? I don't think so. At least, not a safe one. They'd be in the wild card fight. This makes sense as they're at best an inch better on paper than they have been in years past and other tell-tale numbers like shot differential and scoring chance differential are signaling this is an average hockey team.

All I'm saying is enjoy this while it lasts and hope your management team understands it's not real., because there is serious work that needs to be done if they want to get back here.

 

[Burke's Evil Spirit]

In no particular order I'd say the explanations for the Canucks high PDO are as follows:

1. Choosey shooters like Kuzmenko, Pettersson and Miller. Even Boeser isn't really a pure volume guy.
2. Many good -> great 1 shot scorers
3. Leading so often in games, especially early. I think they have scored first more than any team in the league and have the best differential in periods 1 and 2 - that leads to long periods of time where score effects are in play.
4. A top 5 goalie duo
5. A defensive structure which focuses on protecting the middle of the ice and allowing perimeter shots.
6. Luck - there's no question this is a factor here.

The debate is annoying because so many people want to say its all luck or its all skill rather than obviously being a mix of both. How much is anyone's guess, but the longer they do this, the more it tips in the direction of this team being able to maintain an above 100 pdo. 105? Not a chance.

Really good points here. Would quibble a bit with #3 - I don't think score effects are really coupled to PDO (I do think their early scoring, which to a degree is percentage driven, contributes a bit to their pedestrian posession numbers).

 

[SK13]

Really good points here. Would quibble a bit with #3 - I don't think score effects are really coupled to PDO (I do think their early scoring, which to a degree is percentage driven, contributes a bit to their pedestrian posession numbers).

Nope. Their possession numbers are overall worse when they're tied in games. Bottom 10 in shot share and CF% when tied.

None of what was posted there accounts for the biggest gap in team shooting percentage in over 40 years.

 

[Regal]

Demko is a good goaltender and there's nothing fishy about a .916 save percentage from him.

It's the offensive numbers that are as egregious as I've ever seen it in the NHL. Relative to the rest of the league, which is the point of sustainability here, you have to go back like 4 decades to find a stat as stupid as Vancouver at 13.8%.

If you regress them to 12.2%, which would STILL be unsustainable and STILL be #1 in the NHL this year, they'd lose 20 goals in 40 games played. That's like 8 points in the standings alone, and it doesn't address how GF going down often equates to GA going up. Less offensive confidence, less conservative hockey protecting leads, etc.

They should still be a playoff team? I don't think so. At least, not a safe one. They'd be in the wild card fight. This makes sense as they're at best an inch better on paper than they have been in years past and other tell-tale numbers like shot differential and scoring chance differential are signaling this is an average hockey team.

All I'm saying is enjoy this while it lasts and hope your management team understands it's not real., because there is serious work that needs to be done if they want to get back here.

If you drop them to 11% like I said, they’d still have a +24 goal goal differential. They’ve won multiple games by large numbers, which they still would win with less puck luck. They were 13th in the league in scoring last year with a similar lineup but much worse in goal which sunk them to an 83 point team. If they had .910 goaltending last year, they’d have prevented 72 more goals and would have been a playoff team. The year before that with a worse version of Pettersson and worse depth in the lineup, they had 92 points and were 8th in the league in goals against, but 18th in the league in goals for. If that team shot 11%, they would have been a playoff team. I think they’ve shown they can be mid pack in goals and top 10 in goals against and it’s pretty unlikely with the current lineup and Demko that they wouldn’t be a playoff team. Granted that’s a mostly healthy lineup.

 

[SK13]

If you drop them to 11% like I said, they’d still have a +24 goal goal differential. They’ve won multiple games by large numbers, which they still would win with less puck luck. They were 13th in the league in scoring last year with a similar lineup but much worse in goal which sunk them to an 83 point team. If they had .910 goaltending last year, they’d have prevented 72 more goals and would have been a playoff team. The year before that with a worse version of Pettersson and worse depth in the lineup, they had 92 points and were 8th in the league in goals against, but 18th in the league in goals for. If that team shot 11%, they would have been a playoff team. I think they’ve shown they can be mid pack in goals and top 10 in goals against and it’s pretty unlikely with the current lineup and Demko that they wouldn’t be a playoff team. Granted that’s a mostly healthy lineup.

You wouldn't have a +24 goal differential outside of this thought experiment. You'd trail in more games, you'd give up more goals trying to get back into them. That's just how it works.

All we need is time to bear this out. So let's just see.

 

[Regal]

You wouldn't have a +24 goal differential outside of this thought experiment. You'd trail in more games, you'd give up more goals trying to get back into them. That's just how it works.

All we need is time to bear this out. So let's just see.

That’s not how it works. Teams usually control play more when behind. If they give up more goals because of that we’d also have to assume they’d be shooting more and scoring more as well.

Nope. Their possession numbers are overall worse when they're tied in games. Bottom 10 in shot share and CF% when tied.

None of what was posted there accounts for the biggest gap in team shooting percentage in over 40 years.

They score the first goal a ton and so their minutes while tied are low and often early. Over a bigger sample of “within 1” their numbers are much better. The score effects issue is because they haven’t played as much while behind and they protect leads late in games by turtling. Their expected goal numbers were heavily skewed by the blowouts early and have been trending back up. They’re now above 50% overall and top 10 in the league the past 30 games.

 

[Kirikanoir]

Once again the only stat that matters 29-11-3=61 points