Morgan Rielly's the luckiest sum female dog in the league then.

Hainsey's a close 2nd with Jake Gardiner coming in at #4

GO LEAFS GO
- Garnie

Actually my argument is that PDO does not equal luck. At least not if the sample size large.

Actually my argument is that PDO does not equal luck. At least not if the sample size large.
- Goislanders

I think it would be more fair to say that PDO should regress to some mean. It's just that a team's mean isn't usually going to be exactly 100.

Obviously the quality of your goaltending makes a huge difference, and some teams are genuinely better at shooting than others based on their systems / personnel, though this is more tempered since you're averaging the skills of an entire team.

We can see, for example, that over the last three years the worst cumulative PDO teams (Carolina at 98.5, Buffalo at 98.9, Ottawa at 99.0) and best PDO teams (Washington at 102.2, Toronto at 101.7, Tampa at 101.4) are pretty much who you'd expect. Great teams with good-to-great goaltending can sustain a PDO a point or two above 100. Vice versa for crap teams.

But on a team level it can also swing way above or way below what you'd expect without necessarily leveling out by the end of the year. So if we see Tampa chugging along at 102, that's probably a fairly accurate reflection of their true talent rather than luck. But when Pittsburgh (a consistently good PDO team) crashed out at 98.2 last year, it was probably better understood as bad luck.

And if we see a team running 3+ points above or below 100, or a player who is way out of sync with the rest of his team (particularly on the save percentage end) it's a pretty good bet that it's a mirage.

PDO still works as a pretty good rough indicator of good or bad luck, as long as you're reasonably confident about what a team's baseline should be.

Go for it, I couldn't care less about the Jack Adams. Tochhett is definitely doing better this year than he did in the past, although, let's be realistic. The Coyotes have been destroyed by injury and a lot of their success is so improbable that I'd have a hard time just saying it was because the coaching was awesome.

I'm sure they're a well coached team and that that is a factor, but it's probably not as big of a factor as people think.
- James_Tanner

Why do you not care for the Jack Adams?

I agree on most of what you're saying James. It's a good point that a lot of those hard to measure things (like leadership) should, if they are of any use, result in higher numbers in things like shot attempts, high danger scoring chances, etc.

I also agree that Hainsey is not a good defenseman, at least not good enough to be on a first pairing facing top competition. If he's used as a 3rd pairing guy and on the pk, he can have some limited use and maybe can pass on some valuable advice to younger guys...

What I cannot agree on is your terrible slandering of the Foo Fighters! Just because they don't sound as hipster-like as Arcade Fire doesn't mean they aren't good or haven't produced great songs. It just means they aren't to your selective taste.

I think it would be more fair to say that PDO should regress to some mean. It's just that a team's mean isn't usually going to be exactly 100.

Obviously the quality of your goaltending makes a huge difference, and some teams are genuinely better at shooting than others based on their systems / personnel, though this is more tempered since you're averaging the skills of an entire team.

We can see, for example, that over the last three years the worst cumulative PDO teams (Carolina at 98.5, Buffalo at 98.9, Ottawa at 99.0) and best PDO teams (Washington at 102.2, Toronto at 101.7, Tampa at 101.4) are pretty much who you'd expect. Great teams with good-to-great goaltending can sustain a PDO a point or two above 100. Vice versa for crap teams.

But on a team level it can also swing way above or way below what you'd expect without necessarily leveling out by the end of the year. So if we see Tampa chugging along at 102, that's probably a fairly accurate reflection of their true talent rather than luck. But when Pittsburgh (a consistently good PDO team) crashed out at 98.2 last year, it was probably better understood as bad luck.

And if we see a team running 3+ points above or below 100, or a player who is way out of sync with the rest of his team (particularly on the save percentage end) it's a pretty good bet that it's a mirage.

PDO still works as a pretty good rough indicator of good or bad luck, as long as you're reasonably confident about what a team's baseline should be.
- Sven22

Interesting post!

Why do you not care for the Jack Adams?
- neem55

I think him and others have pointed out that it doesn't really seem to go to the best coaches. It's usually just an award for whoever coaches a team that grossly under-performed one year and then bounced back or over-performed the next year lol.

I think him and others have pointed out that it doesn't really seem to go to the best coaches. It's usually just an award for whoever coaches a team that grossly under-performed one year and then bounced back or over-performed the next year lol.
- Thecakeisalie

Ya that's kind of what I expected to hear. I agree to an extent, but I do think Trotz deserves it this year. Does Cooper deserve it more? maybe. I'd be fine with either. But it's not like guys that win don't deserve it. IS Trotz just the recipient of amazing goaltending? Yes, for sure. But look you don't thump the leafs like that in a statement game without having been prepared. Anyway, people love to hate on things like the awards. Should Norris go to the best offensive dman?? No, god no. Doesn't mean they weren't right there at the top. Hopefully the guys who vote change how they look at it so complete defensive liabilities can't win, but no system is flawless. Hate on I guess

I think it would be more fair to say that PDO should regress to some mean. It's just that a team's mean isn't usually going to be exactly 100.

Obviously the quality of your goaltending makes a huge difference, and some teams are genuinely better at shooting than others based on their systems / personnel, though this is more tempered since you're averaging the skills of an entire team.

We can see, for example, that over the last three years the worst cumulative PDO teams (Carolina at 98.5, Buffalo at 98.9, Ottawa at 99.0) and best PDO teams (Washington at 102.2, Toronto at 101.7, Tampa at 101.4) are pretty much who you'd expect. Great teams with good-to-great goaltending can sustain a PDO a point or two above 100. Vice versa for crap teams.

But on a team level it can also swing way above or way below what you'd expect without necessarily leveling out by the end of the year. So if we see Tampa chugging along at 102, that's probably a fairly accurate reflection of their true talent rather than luck. But when Pittsburgh (a consistently good PDO team) crashed out at 98.2 last year, it was probably better understood as bad luck.

And if we see a team running 3+ points above or below 100, or a player who is way out of sync with the rest of his team (particularly on the save percentage end) it's a pretty good bet that it's a mirage.

PDO still works as a pretty good rough indicator of good or bad luck, as long as you're reasonably confident about what a team's baseline should be.
- Sven22

I agree completely.

Regarding the systems/strategi playing a part in PDO, I think you can easily adjust for that, by calculating an expected PDO based on expected shooting% and expected save%. This way you find an xPDO if goaltending and shooting is average.

If we look at this season the best teams in terms of xPDO are: Toronto (101.4), Tampa Bay (101.0), NYI (101.0) and Ottawa (100.8)
The worst teams are: Florida (99.0), Chicago (99.1), Anaheim (99.2) and Vancouver (99.2)

Ya that's kind of what I expected to hear. I agree to an extent, but I do think Trotz deserves it this year. Does Cooper deserve it more? maybe. I'd be fine with either. But it's not like guys that win don't deserve it. IS Trotz just the recipient of amazing goaltending? Yes, for sure. But look you don't thump the leafs like that in a statement game without having been prepared. Anyway, people love to hate on things like the awards. Should Norris go to the best offensive dman?? No, god no. Doesn't mean they weren't right there at the top. Hopefully the guys who vote change how they look at it so complete defensive liabilities can't win, but no system is flawless. Hate on I guess
- neem55

I get what you're saying. We don't live in a perfect world where it's always easy to see the solution or there are lots of perfect answers lol.

How should you decide who gets the Adams? What are the perfect criteria that aren't subjective? How do you separate between which coaches are benefiting from having good players, or good goaltending, or luck, and which teams are doing well from good coaching and structure? And then how do you measure how much of a difference that coaching and structure are making compared to on other teams?

Trotz does seem like a decent candidate to me for what it's worth.

I agree completely.

Regarding the systems/strategi playing a part in PDO, I think you can easily adjust for that, by calculating an expected PDO based on expected shooting% and expected save%. This way you find an xPDO if goaltending and shooting is average.

If we look at this season the best teams in terms of xPDO are: Toronto (101.4), Tampa Bay (101.0), NYI (101.0) and Ottawa (100.8)
The worst teams are: Florida (99.0), Chicago (99.1), Anaheim (99.2) and Vancouver (99.2)
- Goislanders

The fact that Ottawa has the fourth best xPDO despite how bad they are is pretty (frank)ed lol. Is this because their goaltending is garbage? Or are they actually getting lucky on their goals despite how much they already suck?

The fact that Ottawa has the fourth best xPDO despite how bad they are is pretty (frank)ed lol. Is this because their goaltending is garbage? Or are they actually getting lucky on their goals despite how much they already suck?
- Thecakeisalie

Goaltending is awful. Their save% is more than 2% below expected, which is league worst ahead of New Jersey and Philly. Islanders is league best at 1.5% above expected.

why did you write an essay on statistics when you made zero arguments using any statistical analysis? correlation? t-tests?
how can you be so definite that it's a fluke?
all i see in this post is just a rant of your gut feeling and an inappropriate stab at people who enjoy the Foo Fighters

why did you write an essay on statistics when you made zero arguments using any statistical analysis? correlation? t-tests?
how can you be so definite that it's a fluke?
all i see in this post is just a rant of your gut feeling and an inappropriate stab at people who enjoy the Foo Fighters
- oiljapan

Well listen my man, it's not supposed to be a comprehensive guide to actual statistical analysis. It's no more in need of t-tests than if I was to just use goals and assists. We're just using basic results based stats to give an amateur enthusiast the ability to discern which players are good.

Besides, based on what I can do with just five results-stats, I could be one of the top five NHL GMs, 25 of which are uniformly awful at their jobs.

I'm going to be very honest also: I don't have any idea what a t-test is.

Goaltending is awful. Their save% is more than 2% below expected, which is league worst ahead of New Jersey and Philly. Islanders is league best at 1.5% above expected.
- Goislanders

OK. Yeah, goaltending can sometimes be the reason teams play far below or above expectations and Craig Anderson has been pretty hit or miss his whole career lol. He'll have amazing seasons and then terrible ones, although he might just be too old at this point. I don't think he's got any amazing seasons left in the tank.
Admittedly, they've probably got more problems than just bad goaltending...

I'm listening to the Foo Fighters right now! What a great band!

Here is the bottom line. You can't use the analytics available to us as fans to say this player is a good player, or this player is a bad player. They only tell you what happens on the ice when a player is on the ice or off the ice. Nothing more, nothing less. Use them in that manner and they're useful. Use them as Tanner does and they're not. They're team numbers, not individual player numbers. So a player can't be a good leader if his team doesn't win? When Morgan Reilly is separated from Hainsey, who is he playing with and in what game situations? I bet Hainsey plays more when the Leafs have the lead than when they're behind. WOWY's are a really poor stat. Any shot based metric is suspect for individual players.

When we look at Ron Hainsey, we see that he plays for one of the best teams in the NHL and that relative to his team, he is very ineffective.

key statement. i think analytics are for the most part useless when isolated and not compared to a player's peers/teammates.

I'm listening to the Foo Fighters right now! What a great band!

- Thecakeisalie

Doing my part for mankind:

I'm listening to the Foo Fighters right now! What a great band!

- Thecakeisalie

And you cheer for the Leaves, just a perfect storm. Nickleback too?

Well listen my man, it's not supposed to be a comprehensive guide to actual statistical analysis. It's no more in need of t-tests than if I was to just use goals and assists. We're just using basic results based stats to give an amateur enthusiast the ability to discern which players are good.

Besides, based on what I can do with just five results-stats, I could be one of the top five NHL GMs, 25 of which are uniformly awful at their jobs.

I'm going to be very honest also: I don't have any idea what a t-test is.
- James_Tanner

No, you wouldn't be better then 25 other GM's but I bet they could write as well as you can or better.

No, you wouldn't be better then 25 other GM's but I bet they could write as well as you can or better.
- Garnie

See I don't just disagree with you, I find it insane that you wouldn't agree with me. Being an NHL GM would be a cinch. There is so many guys in charge who value the wrong things and consistently make bad trades.

With no experience at all, I could outperform Chuck Fletcher just by waking up in the morning. The Oilers are considering hiring Dave Nonis - you think I couldn't GM circles around that moron? Or even do what Bob Nicholson does, even if I had a concussion?

Come on.

See I don't just disagree with you, I find it insane that you wouldn't agree with me. Being an NHL GM would be a cinch. There is so many guys in charge who value the wrong things and consistently make bad trades.

With no experience at all, I could outperform Chuck Fletcher just by waking up in the morning. The Oilers are considering hiring Dave Nonis - you think I couldn't GM circles around that moron? Or even do what Bob Nicholson does, even if I had a concussion?

Come on.
- James_Tanner

I've seen some of your thoughts on running teams over the years. I'd be a hard pass, but thanks for applying.

See I don't just disagree with you, I find it insane that you wouldn't agree with me. Being an NHL GM would be a cinch. There is so many guys in charge who value the wrong things and consistently make bad trades.

With no experience at all, I could outperform Chuck Fletcher just by waking up in the morning. The Oilers are considering hiring Dave Nonis - you think I couldn't GM circles around that moron? Or even do what Bob Nicholson does, even if I had a concussion?

Come on.
- James_Tanner

The flaw in this opinion you have is that you have invented your own standard of how you judge GM. It's purely based on faulty analytics and an inability to use them properly. All GM's make mistakes and some are much better than others. I truly wish you could have the experience of being around an NHL team and what that is truly about. If you would open your mind, you might learn something.

Why do you not care for the Jack Adams?
- neem55

Prefer the Jack Daniels