• Moderators, please send me a PM if you are unable to access mod permissions. Thanks, Habsy.

GDT #61: Attack

LeafOfFaith said:
No problem with having the stats.

Problem with letting them replace the eyes.
Click to expand...

"Your eyes are lying bastards"

-Kyle Dubas.




The most that the absolute best hockey observer can hope is to develop semi accurate impressions on what is going on out there over any sort of real sample size. Has Connor Brown been good over the last 10 games? Well define good. He hasn't been noticeable (whether for good or for bad)...but is that good, or is it bad? You can't possibly remember every shot attempt or zone entry he's been a part of over that period of time. You can't possibly believe that you can accurately tell me whether that is average, below, or above and by how much...because that would require you literally watching every ****ing game in the league for the last 3 weeks.

What the informed hockey observer should be doing is explaining the how. How does this player generate these results. Whether the results exist, or are important simply isn't up for debate. The math exists to show you the probability of a team winning a hockey game based on the shot metrics. What is left for the eyes is explaining how certain players are able to generate their results. Watch William Nylander for 2 shifts and you can describe how he's a zone exit/entry monster for example. But that he's a entry/exit monster isn't up for debate. Neither is the importance of clean entries and exits (and before you try to tell me that nobody argues that...an entire generation of Canadian hockey players were raised on up and out, and to dump and chase the puck, not to try to carry in out of the defensive zone and into the offensive zone under control).

The eyes are good enough to learn the how, but human cognition simply isn't capable of determining the importance of small events without assistance. There's too much going on, too much to remember and a bucket of human cognitive biases needing unpacking for the unguided eye test analysis to be anything but terrible.
 
I mean last night alone, his pets all came up big despite people railing on him the first half of the game.
Click to expand...

Stuff like this is why human cognition sucks at this type of analysis.

It's easy to point at the impactful events that did happen (Marleau to Ennis tic-tac-toe, Hyman blocking Marners shot and the puck falling right on his stick) and proclaim that clearly the coach was right because it worked. But how many times didn't it work? Does it work less often than other players would be able to make it work? Patrick Marleau has 112 minutes of power play time this season and has 8 points in that time...is that good? Tyler Ennis has 85 minutes of PP time this season and 3 points...is that good? Do we have players who are better choices who would have scored more often given that opportunity?
 
MindzEye said:
"Your eyes are lying bastards"

-Kyle Dubas.




The most that the absolute best hockey observer can hope is to develop semi accurate impressions on what is going on out there over any sort of real sample size. Has Connor Brown been good over the last 10 games? Well define good. He hasn't been noticeable (whether for good or for bad)...but is that good, or is it bad? You can't possibly remember every shot attempt or zone entry he's been a part of over that period of time. You can't possibly believe that you can accurately tell me whether that is average, below, or above and by how much...because that would require you literally watching every ****ing game in the league for the last 3 weeks.

What the informed hockey observer should be doing is explaining the how. How does this player generate these results. Whether the results exist, or are important simply isn't up for debate. The math exists to show you the probability of a team winning a hockey game based on the shot metrics. What is left for the eyes is explaining how certain players are able to generate their results. Watch William Nylander for 2 shifts and you can describe how he's a zone exit/entry monster for example. But that he's a entry/exit monster isn't up for debate. Neither is the importance of clean entries and exits (and before you try to tell me that nobody argues that...an entire generation of Canadian hockey players were raised on up and out, and to dump and chase the puck, not to try to carry in out of the defensive zone and into the offensive zone under control).

The eyes are good enough to learn the how, but human cognition simply isn't capable of determining the importance of small events without assistance. There's too much going on, too much to remember and a bucket of human cognitive biases needing unpacking for the unguided eye test analysis to be anything but terrible.
Click to expand...

That Dubas quote is precisely the problem.

If I see Brown playing great but oops his ****ing Corsi isn’t good, I’m not going to flip my opinion. Again, there are nine other guys on the ice. Team attempts are cute and all, but they don’t negate positive stuff you’re seeing a player do.

And yeah, his pets don’t do shit a lot of games too. But as you pointed out, same goes for the stars. Matty has looked pretty bad for stretches this year, despite his great totals. Marner was invisible for several recent games. Hyman gets shit on night and day here, and while his hands are pretty stoney, guy hustles, fights for every inch, and goes to the net. So it’s fine to downplay Marner’s shot hitting him and all, but the guy went into the dirty area and made a great play. Despite being infinitely better than him, a lot of our stars don’t want to get their hands too dirty. So on a night where Kapanen’s shot rings off the post, Hyman’s gets buried. We can’t have twelve of the same player up front. Yeah, I’d also prefer to see Hyman on the third line and another scorer in his spot, but I get it.
 
LeafOfFaith said:
That Dubas quote is precisely the problem.

If I see Brown playing great but oops his ****ing Corsi isn’t good, I’m not going to flip my opinion. Again, there are nine other guys on the ice. Team attempts are cute and all, but they don’t negate positive stuff you’re seeing a player do.
Click to expand...

The result don't lie. Your eyes do. You have no way to know if what you're valuing is actually important, or sustainable without better analysis.

And yeah, his pets don’t do shit a lot of games too. But as you pointed out, same goes for the stars. Matty has looked pretty bad for stretches this year, despite his great totals. Marner was invisible for several recent games. Hyman gets shit on night and day here, and while his hands are pretty stoney, guy hustles, fights for every inch, and goes to the net. So it’s fine to downplay Marner’s shot hitting him and all, but the guy went into the dirty area and made a great play. Despite being infinitely better than him, a lot of our stars don’t want to get their hands too dirty. So on a night where Kapanen’s shot rings off the post, Hyman’s gets buried. We can’t have twelve of the same player up front. Yeah, I’d also prefer to see Hyman on the third line and another scorer in his spot, but I get it.
Click to expand...

This paragraph is part of the problem I'm trying to describe. When Matthews "looks pretty bad for stretches", he still generates good shot numbers (when he's actually bad, he doesn't). The difference between him looking good and bad is often a couple pucks sneaking by the goalie. If he's generating the same number and quality of shots, it's the same guy, just perceived differently. The difference between effective and not effective isn't moving your legs really fast and trying to look more dangerous. It's generating more and better shots on net, regardless of how it looks.

So while Hyman looks like he's working his bag off...is it actually effective? Because that's all that matters, is whether it's effective. Not how we feel about how it looks out there. Like I said in my original post, literally zero real industries out there think the way about human performance that a lot of the NHL still does. Everything can be measured, everything measured can be analyzed to find better levels of efficiency. This old school thinking is how average CEO's end up being 6'3 when there should be no correlation between height and running a company. Because humans are ****ing dumb and loaded with biases.
 
Whatever, we can go around in circles on this as we usually do, so why bother. I acknowledge that there’s something to be said for your position, but I’m never going to roll over on your stance on the numbers replacing the eyes because we’re all biased know-nothings.

Your advanced stats told you Leivo is a rock star. He’s played 18-20 minutes the last five games and registered zero points. His Vancouver totals are pretty damn close to his Leaf totals, except he’s playing 20 minutes instead of 5 minutes per game and getting PP time. So the numbers lied about him or the eyes did?
 
LeafOfFaith said:
Whatever, we can go around in circles on this as we usually do, so why bother. I acknowledge that there’s something to be said for your position, but I’m never going to roll over on your stance on the numbers replacing the eyes because we’re all biased know-nothings.

Your advanced stats told you Leivo is a rock star. He’s played 18-20 minutes the last five games and registered zero points. His Vancouver totals are pretty damn close to his Leaf totals, except he’s playing 20 minutes instead of 5 minutes per game and getting PP time. So the numbers lied about him or the eyes did?
Click to expand...

A few things...

- "Middle 6 forward" is hardly a "rock star". I've consistently called him a middle 6 player and he's proven that he is a middle 6 forward in Vancouver, not remotely the non/fringe NHL scrub you claimed he was.

- Leivo isn't playing 20 minutes a game, he's played under 15 minutes a game as a Canuck and has 12 points in 30 games after his 6 points in 27 for the Leafs. How you come up with his totals being close to his Leaf totals is beyond me. Though it would explain why you have such a distrust for numbers.

- He's 2nd on the Canucks in goals per 60 (only behind Pettersson) and 4th in Points per 60. His low raw point total is largely a function of have 1 secondary assist as a Canuck.

- Pace out his 30 games as a Canuck (including this 5 game pointless stretch) and he's on pace for 19 goals and 32 points in 3rd line usage. ~20 goals and stud possession numbers out of a 3rd liner? We could use that.
 
He’s played around 20 minutes the past five games and is zeroes across the board.

In any event, what was his average with the Leafs? 7 minutes? He’s played two to three times more, with PP time, and looks with Pettersen and Boeser. And has 12 points in 30 games. Pretty sure Tyler Ennis can do that.

I claimed he was a good fourth liner. Still looks to be the case.
 
LeafOfFaith said:
He’s played around 20 minutes the past five games and is zeroes across the board.
Click to expand...

Oh noes, 5 whole games.

In any event, what was his average with the Leafs? 7 minutes? He’s played two to three times more, with PP time, and looks with Pettersen and Boeser. And has 12 points in 30 games. Pretty sure Tyler Ennis can do that.
Click to expand...

See, it's not that you hate stats. You just hate good stats. All the old shitty stats that support your argument you're all in for.

No, he hasn't played 2-3x more. He's played 4 minutes more. This is why per minute statistics are so valuable

1.71 points per 60 minutes of 5 on 5 ice time played. To deal with your shitty jab at him...yeah, Tyler Ennis can do that. But Tyler Ennis does that while being a massive possession liability. 44.5% of shot attempts taken when Tyler Ennis is on the ice, go towards the right net. 54.% of shots that occur when Leivo is on the ice, go towards the right net. Feel free to twist yourself into a pretzel explaining this, it amuses me and I'm pretty sure, most of the people who read it.

1.71 fwiw, is better than Hyman, Brown, Marleau. Kadri has 1.72 this season. Which was always my point. He deserved to play further up the lineup.


I claimed he was a good fourth liner. Still looks to be the case.
Click to expand...

Good 4th liners don't dominate puck possession and score at a 20 goal pace in less than 15 minutes a night.
 
My apologies, he's played over 40% more per game on average in Vancouver than Toronto, and obviously getting more minutes lately. You're right, it's not 2x on average, but it is 2x over at least the last five games. And he hasn't been buried on the 4th line. He's playing with their better players and getting PP time.

So whoop de doo, he's on pace for 32 points. What's next, "but he needs first line minutes in order to properly..."? He's been getting them lately and doing nothing.

You're more than welcome to have him and his dominant corsi play on your second line. While he flubs, and misses the net, and doesn't do much of anything else. I'm not missing him.
 
LeafOfFaith said:
My apologies, he's played over 40% more per game on average in Vancouver than Toronto, and obviously getting more minutes lately. You're right, it's not 2x on average, but it is 2x over at least the last five games.
Click to expand...

You have a hard time accepting when you make clear factual errors (which you do at a fairly embarrassing rate when you're pressed on specifics about some of your pet arguments) and try to explain it away, often in a smarmy fashion. It's a bad look man, smarten up.

And he hasn't been buried on the 4th line. He's playing with their better players and getting PP time. So whoop de doo, he's on pace for 32 points.
Click to expand...

Great, what if I told you that when you remove his 4th line usage in Vancouver, he's on a 40 point pace in just over 15 minutes a game?

What's next, "but he needs first line minutes in order to properly..."? He's been getting them lately and doing nothing.
Click to expand...

You're having a hard time grasping basic facts. He's producing well. 20 goal pace, good primary point numbers for his usage. Wave around his last 5 games all you want, it's just stupid to focus on small samples. We have over 100 career games on the guys resume and the trend is clear. When he gets to play on the top 3 lines, he produces solid middle 6 offensive numbers while providing very, very good possession results. I don't need to explain anything away, he's done well in Vancouver. I'm supposed to be more impressed if he passed to a guy who passed to another guy who scored a goal a couple more times? Primary points man, they're important. Far more important than secondary assists.

You're more than welcome to have him and his dominant corsi play on your second line. While he flubs, and misses the net, and doesn't do much of anything else. I'm not missing him.
Click to expand...

It's weird that you're going to talk about him flubbing chances when we're playing Hyman and Marleau in the type of roster spot I wanted Leivo in. 925K for a 20 goal pace in 3rd line usage and great possession numbers. Not ****ing bad. Oh, and he jumped in to fight Kesler last week after Pettersson hit him and Kesler went after Pettersson. Gud pro type stuff.
 
realized i never checked the TOI this game.


Rielly 24:45
Hainsey 21:16
Zaitsev 20:34
Muzzin 18:43
Gardiner 18:39
Dermott 12:49

ugh.
 
uncus said:
Is it because of babcock or is it the defense coach
Click to expand...

I don't think the highest paid coach in the league is being told how to distribute blueline minutes by his defence coach. That would actually be worse imo than Babcock doing it himself.
 
zeke said:
realized i never checked the TOI this game.


Rielly 24:45
Hainsey 21:16
Zaitsev 20:34
Muzzin 18:43
Gardiner 18:39
Dermott 12:49

ugh.
Click to expand...
This is exactly why it's so crucial that we get rid of Zaitsev today.

Bad enough having one of Hainsey or Zaitsev getting top-3 minutes on our blueline...but both of them? That'll be a killer against a good team in the playoffs.
 
You must log in or register to reply here.
Back
Top