Are the Capitals and Jets PDO merchants ?

[Susannah]

For those who understand PDO, can you answer these?

Is the team's PDO high because that team is having a good season or the other way around?

Do good teams have high PDOs and not so good teams have low PDOs?

Two teams that had great records last year that are having not so great records: Rangers and Canucks. Did people predict these two teams to crash down because they had high PDO numbers last year? Is this just a convenient and easy argument that their PDOs were high so they are not going to have good seasons.

How about the opposite side?

 

[JaegerDice]

It’s a very shallow look at a teams performance. Doesn’t take into account a lot of factors. Just a very poor advanced stat.

It's really not, though, if you understand... hockey?

In any given game, both goalies are facing shots and saving shots. The shots they save are sv%. The shots that go in are sh%. The higher the sv%, obviously, the lower the sh%, and vice versa. They are in direct opposition.

Throughout an 82-game season, where all teams play, the games where the goalies overperform and the shooters overperform balance out towards a median measured as 1.

The teams above 1 are overachieving. One or both of their sh% or sv% will regress downard.

The teams below 1 are underachieving. One or both of the sh% or sv% will regress upward.

The problem with PDO, is there is no timeline. Sometimes, these teams relying on PDO crash hilariously in the 1st round (2014 Avalanche), sometimes they crash hilariously in the finals (2021 Habs), and sometimes... they even win (2011 Bruins).

All we know is that PDO is fleeting and uncontrollable, and thus will eventually regress towards 1.0.

 

[JaegerDice]

For those who understand PDO, can you answer these?

Is the team's PDO high because that team is having a good season or the other way around?

Do good teams have high PDOs and not so good teams have low PDOs?

Two teams that had great records last year that are having not so great records: Rangers and Canucks. Did people predict these two teams to crash down because they had high PDO numbers last year? Is this just a convenient and easy argument that their PDOs were high so they are not going to have good seasons.

How about the opposite side?

A team's PDO is highly correlated to success. If a team has a high PDO it is likely that they are having a good season.

The problem is that PDO is not controllable. There's not much a team can do to impact their PDO beyond increasing shot and chance volume for and reducing volume of same against.

This is because PDO is a combination of sh% and sv%. Both of which are uncontrollable.

If either were controllable, every shooter would simply choose to score a goal on every shot, and every goalie would choose to save a goal on every shot they faced.


Good teams can have high PDOs and low PDOs. Bad teams have high PDOs and low PDOs.

Generally speaking, good teams have strong puck possession, which is a buffer against the 'luck' of PDO. If you prevent the opposing team from shooting the puck (shot suppression), you are better equipped to succeed despite a low save percentage from your goalie, because they are seeing fewer shots total.

If your team generates a lot of shots and chances, then you will likely score, even if the bounces are going against you. A team that generates 20 chances is twice as likely to score as a team that generates 10 chances, all things being equal. Obviously, not all things are equal, but a lot of factors that make things unequal can be overcome with volume over time.

 

[Three On Zero]

It's really not, though, if you understand... hockey?

In any given game, both goalies are facing shots and saving shots. The shots they save are sv%. The shots that go in are sh%. The higher the sv%, obviously, the lower the sh%, and vice versa. They are in direct opposition.

Throughout an 82-game season, where all teams play, the games where the goalies overperform and the shooters overperform balance out towards a median measured as 1.

The teams above 1 are overachieving. One or both of their sh% or sv% will regress downard.

The teams below 1 are underachieving. One or both of the sh% or sv% will regress upward.

The problem with PDO, is there is no timeline. Sometimes, these teams relying on PDO crash hilariously in the 1st round (2014 Avalanche), sometimes they crash hilariously in the finals (2021 Habs), and sometimes... they even win (2011 Bruins).

All we know is that PDO is fleeting and uncontrollable, and thus will eventually regress towards 1.0.

In theory yes, but it doesn’t factor into external issues. A team can be firing on all cylinders and a rash of injuries heavily drop their PDO. Now is that regression or just unfortunate circumstances. Players get traded mid season, they also impacts PDO

PDO best works in a perfect world where nothing changes

 

[AlexModvechkin8]

At 5-on-5, Washington is top 10 in advanced metrics like expected goals for %, high danger chances for %, scoring chances for %, etc. They’re not PDO monsters, though LT is having a terrific season in goal.

I’m higher on the Caps D corps than most I suppose. Maybe the sum of the whole is greater than the parts but they all fit really well together. Top 2-3 might be aggressive but I don’t think they’d fall outside of the top ten. Agree to disagree tho.

Re: Strome and PLD, call it slow/fast starts and then regression to the mean both ways but Strome is still at PPG pace and PLD has been a beast at both ends of the ice even with horrible luck shooting. Neither are ever going to be mistaken for Crosby, MacK, McDavid, etc but they’re both playing well.

Man I'm not so sure about this but they have been healthy with 4 of them playing in all 46 games and then 41 and 36.

At full health though I think you would be hard pressed to put that group, in a vacuum in the top 10 among NHL defensive groups and certainly more than 2 or 3 other teams.




Strome had an insane start with 34 points in 21 games his first 2 months and now has 12 in his last 22.

PLD had a slow start and has played quite well since then but he never seems to be able to sustain it.

Dowd is incredible if he is your 4th liner but he really is the 3rd line center to Eller's 4th line.



Vegas defensive group is most certainly better 1-6


The thing is that most stats showing your above assertions will also show PDO right?

 

[thebestnic]

Thompson or Helle are gonna win the Vezina that's all it really means. PDO is a pretty boring/meaningless stat. Sure shot percentage usually regress but I'm not sure you can say the same for sv%

 

[Neil Racki]

Yes

And I am enjoying the hell out of it for as long as it will last

(But not much scoring puck luck last month or so)

 

[JaegerDice]

In theory yes, but it doesn’t factor into external issues. A team can be firing on all cylinders and a rash of injuries heavily drop their PDO. Now is that regression or just unfortunate circumstances. Players get traded mid season, they also impacts PDO

PDO best works in a perfect world where nothing changes

That's true. There's absolutely noise in the data when it comes to PDO.

 

[Three On Zero]

That's true. There's absolutely noise in the data when it comes to PDO.

I too may be a bit harsh on it, but I despise the advanced stat movement in a game that has too many factors to manipulate the data used.

 

[DitchMarner]

The Capitals are good and deep at every position and have very good coaching. I don't know why their success is so shocking to some people.

I mean, they're probably not as good as their record, but there's no reason to think they shouldn't be among the top teams in the League.

If anything, their scoring has more to do with skill and execution than with luck. Unlike some teams, they don't shoot as much as possible hoping to get bounces; they orchestrate high danger chances and have guys who can put the puck in the net. They were like that when they were a good team in the past as well. But their defense now may be better than it was in the mid 2010s and is definitely better than it was during Ovechkin's peak years.

 

[wetcoast]

At 5-on-5, Washington is top 10 in advanced metrics like expected goals for %, high danger chances for %, scoring chances for %, etc. They’re not PDO monsters, though LT is having a terrific season in goal.

I’m higher on the Caps D corps than most I suppose. Maybe the sum of the whole is greater than the parts but they all fit really well together. Top 2-3 might be aggressive but I don’t think they’d fall outside of the top ten. Agree to disagree tho.

Re: Strome and PLD, call it slow/fast starts and then regression to the mean both ways but Strome is still at PPG pace and PLD has been a beast at both ends of the ice even with horrible luck shooting. Neither are ever going to be mistaken for Crosby, MacK, McDavid, etc but they’re both playing well.

Um okay sure but do you really expect a deep run with this group this year?

Sure some guys are having "career" years but it seems like they are really relying on goaltending, a bit like last year.

Although the East and Metro division are easier paths than the West.

 

[Nogatco Rd]

Um okay sure but do you really expect a deep run with this group this year?

Sure some guys are having "career" years but it seems like they are really relying on goaltending, a bit like last year.

Although the East and Metro division are easier paths than the West.

I don't expect it, but I also wouldn't be surprised by it. Ovi still looks slow and rusty since coming back from the broken leg, but if he's able to recapture the pace (if not the sh%) he played with to start the year then they'll be a dangerous team.

 

[BB79]

The Capitals are good and deep at every position and have very good coaching. I don't know why their success is so shocking to some people

Because they friggin sucked last season and Ovechkin looked completely cooked. Their rising from the dead is surprising

 

[AlexModvechkin8]

Um okay sure but do you really expect a deep run with this group this year?

Sure some guys are having "career" years but it seems like they are really relying on goaltending, a bit like last year.

Although the East and Metro division are easier paths than the West.

Maybe? I don’t know, hockey is random. I think the Caps could go to the SCF or get upset as the one seed. I think they’re good enough to make a run, and they’re going to get better with the addition of Ryan Leonard in April.

The Caps are much better than they were last year and it’s not particularly close. Last year the Caps either squeaked by for a win or got blown out, they had no margin for error because they weren’t very good. If you think the Caps of this year are the same as the Caps of last year I’m fairly confident that you didn’t watch either team very much. Washington has the second best GD this season and they’re tops in the league in goals per game.

 

[Zarzh]

Lol no that's not the belief. That's what people who don't understand what we are talking about believe we are saying.

It's just a measure. Good teams will have higher PDOs because they have better goaltenders and better shooters. Zero people are arguing that PDO over 100 means you are just lucky. However exceptionally high PDOs are not sustainable generally.

Separate it out into its individual parts. We can even just take one. Some players have better shooting percentages than others. Maybe a player is real good and can maintain a shooting percentage of 15%. Then one year he has a shooting percentage of 50% for 3 months. Of course the 50% isn't sustainable and will regress, but that doesn't mean his shooting percentage will just drop to league average at some point. It means it will likely regress to around his career shooting percentage, which is still above average.

Generally they don't, the amount of shooting talent you actually need to do so is absurdly high and even when you get that it still creates a statistically fragile team far more likely to be "upset" in the playoffs.

 

[Raccoon Jesus]

The Capitals are good and deep at every position and have very good coaching. I don't know why their success is so shocking to some people.

I mean, they're probably not as good as their record, but there's no reason to think they shouldn't be among the top teams in the League.

If anything, their scoring has more to do with skill and execution than with luck. Unlike some teams, they don't shoot as much as possible hoping to get bounces; they orchestrate high danger chances and have guys who can put the puck in the net. They were like that when they were a good team in the past as well. But their defense now may be better than it was in the mid 2010s and is definitely better than it was during Ovechkin's peak years.

I mean that's really all there is to it

Both things can be true--Washington can be a monster team and still be benefiting a bit from an elevated PDO.

It's the same thing as a healthy Vancouver although theirs last year was a bit crazier, very good team, but on a shooting luck bender.

Its nothing to be offended by unless someone's saying 'unsustainable' as in "washington isn't actually a divisional playoff team" but I don't see any of that. And I don't see, fortunately, many of their fans doing the opposite, proclaiming they've found the glitch in hockey and actually they're the exception to the rule like some other high PDO teams have done.

 

[DitchMarner]

Because they friggin sucked last season and Ovechkin looked completely cooked. Their rising from the dead is surprising

Look at the changes they've made.

They added Chychrun and Roy, who are both good defensemen. Sandin continues to mature.

They brought in PLD, who was a disaster in LA but definitely has talent and can be good when he competes and is focused. He's played well. They added some depth up front (bringing back Eller, signing Raddysh and acquiring Mangiapane).

Guys like McMichael and Protas have progressed. Wilson is better and yes, Ovechkin is a lot better.

That's a lot of change, and while they're weren't very good for a playoff team, they at least were a playoff team. It's not like they were 14th in their conference.

 

[Brookbank]

Thing is the Canucks went from 22nd overall to 6th. They went from a -22 goal differential to +56 in one season.

Last year the Jets were 4th and +60. So far this year are 2nd and +52.

Not sure you can extrapolate much comparing the Canucks to the Jets.

Last year:

2023 - Vancouver 102.9%, Winnipeg 102.7%,

 

[Mattilaus]

For those who understand PDO, can you answer these?

Is the team's PDO high because that team is having a good season or the other way around?

Do good teams have high PDOs and not so good teams have low PDOs?

Two teams that had great records last year that are having not so great records: Rangers and Canucks. Did people predict these two teams to crash down because they had high PDO numbers last year? Is this just a convenient and easy argument that their PDOs were high so they are not going to have good seasons.

How about the opposite side?

It doesn't mean anything. It's just sv% combined with sh%. Good and bad teams can have good and bad PDOs. If you have a team with middling possession metrics however and an extremely high PDO, it's likely they will regress a bit at some point when their goaltender either goes cold or their players stop shooting at a super high rate.

Think of it like this. You have two teams as follows:

Team A:
take 40 shots a game
allow only 20 shots against
Their goaltender is average and usually floats around .900 sv%
Their team shooting % is around 10%

Their sh% and sv% are quite middling, but they could be a decent team because they are getting so many shots more than their opponents that they could win a lot anyways. Even with only a 10% sh% they would average around 4 goals a game which is great.

This team only has a PDO of 100

Team B:
Only take 20 shots a game
Allow 40 shots a game
Goaltender is hot and is currently sporting a .950 sv%
Their shooters are also hot and currently team sh% is 20%

We all pretty much agree, it's likely not great to continue playing this way, allowing 40 shots a game and getting only 20, but this team would theoretically also score 4 goals a game and only allow 2.

This team has a PDO of 115.

That's all it is, it's just combining two regularly used and normal stats into one number. Nobody bats an eye when you say "Yea Team B's record is good but they are clearly just riding a hot goalie right now and it isn't likely to last", but as soon as you talk about it in the same breath as sh% everyone loses their minds.

Of course there is context like high danger chances and statistically better goalies, but we aren't writing an essay here. This is just an internet forum.

 

[Breakers]

10000% yes for the capitals

 

[FlappyGiraffe]

I think in the Jets case it's Helle having an all time great season. He's nearly at a .930% (league average 0.901%) leading all goalies in games played.

PDO being a function of save % and shooting % means the Jets would rank high even with a league average offense

 

[Rec T]

I was in the same boat, but I made it easier using this as the google search "PDO hockey"

Well I did too eventually & that worked fine. My point was that we shouldn't 'have' to go digging through google to figure out what they're talking about. If the hockey version of PDO isn't the first result in a search, then it's not well known enough to be obvious & the thing to do is define it for those of us who had no clue.

After that the discussion can devolve into the usual "You don't know what you're talking about because of 'x' (advanced stat)" "Nuh huh, 'you're' the one that couldn't find your butt with both hands..." "Oh yeah, well your momma..." And life goes on.

 

[Brookbank]

Combo of sv% and sh%.
Basically its unsustainable scoring and saving.

See: Canucks this year and last year.

Last year the Canucks never played a game without all of the top 6 and top pair defensman.

This year , they've played a grand total of 2 games with all of the same pieces in place. The second game was the win vs the Oilers.

 

[qc14]

Caps are not PDO merchants in the sense that their underlying numbers are legitimately good even if they've come down from the top-3 in the league pace they were playing at early in the year. They are legitimately playing significantly better than the wild card mess in the east, for example.

They absolutely are PDO merchants to be leading the Presidents' trophy race though.

 

[Pablo El Perro]

This may not have been a serious question, but I did some quick digging for fun and I have to imagine it's Edmonton in 1983-84 with 105. 17% shooting and 88% goaltending.

The Oilers were 1st in the NHL for shooting percent every year of the decade except for the Penguins in 1989-90.

The 81-82 Isles were tied with the 83-84 Oilers, at 105.2. I'd guess it's harder to track with older teams, but the highest recorded PDO belongs to the 78-79 Isles, at 106.3. Another high PDO dynasty was the 70s Habs.

For those who understand PDO, can you answer these?

Is the team's PDO high because that team is having a good season or the other way around?

Do good teams have high PDOs and not so good teams have low PDOs?

Two teams that had great records last year that are having not so great records: Rangers and Canucks. Did people predict these two teams to crash down because they had high PDO numbers last year? Is this just a convenient and easy argument that their PDOs were high so they are not going to have good seasons.

How about the opposite side?

Historically, some of the best teams ever have had high PDOs.