Post-Game Talk: Canucks 3 Leafs 2

[Gallagbi]

It is flawed but all stats are right?

The old school approach is to strictly go by shots on goal but that's more flawed than trying to discern how dangerous each shot is.

The last two games were completely lopsided in the Leafs favor for shots but the Calgary game was more of a 55-45 xGF% game for the Leafs and the Canucks game a 65-35 game which is a massive difference in the grand scheme of things and matched the eye test too.

Personally I'm finding the total xGF and difference is a better measure than xGF%. Just feels like it can balance out the volume in a way % doesn't

 

[Gallagbi]

I'm not saying these stats arent useful because they definitely are.

I think some people take them too far sometimes tho and some important real time context is left out.

I'd say there's more incorrect context that has been applied throughout this thread, rather than the stats being off.

 

[AvroArrow]

I'd say there's more incorrect context that has been applied throughout this thread.

The belief the Leafs were regularly boxed out and kept from prime scoring areas is pretty wrong looking at basically any stat or total recap of the game.

Anyone who suggests the following is either blind, never watched the game or belongs in an asylum. That was as one sided of a game as you'll ever see. Pure domination, Demko played one of the best games I've ever seen from a goalie.

 

[weems]

I'd say there's more incorrect context that has been applied throughout this thread, rather than the stats being off.

Sure but that doesnt change the fact that xG in a one game setting can be extremely flawed.

Just last night they gave Jake Muzzin a 0.24xG for taking a shot literally from about the goalline.

I think is was credited as the highest or one of the highest xG events of the entire game and in reality it hit the side of the goal which is no suprise considering the angle.

 

[Gallagbi]

Sure but that doesnt change the fact that xG in a one game setting can be extremely flawed..

Do you think last night was extremely flawed on the whole? Total xGF last night was a little over 5. Given we had 50 shots and the quality of chances that I saw, it seems about right.

 

[zeke]

So Kampf has a chance from the faceoff dot where he literally only has 1 hand on his stick and shoots the puck at about 2mph but is graded as having the same expected goal # as John Tavares who is 7 feet closer to the net and actually gets off a good quality shot?

Yes you need to consistently produce chances from these areas and over time they will churn out success rate at a certain number, but in a 1 one game setting applying proper context to each chance and factoring in things like shot type, body position, angle is very important.

Kampf actually has a higher 5v5 shooting percentage (10.4%) than Tavares does (9.6%).

The bulk of the talent is in getting the chances. Very rarely you do get someone special like Matthews who consistently converts at a much higher rate, though.

Do you really think untalented players create as many chances as talented ones?

Are you suggesting on the year that the Leafs in particular are not as good as their xgf suggests?

 

[weems]

Do you think last night was extremely flawed on the whole? Total xGF last night was a little over 5. Given we had 50 shots and the quality of chances that I saw, it seems about right.

I didnt really post anything about last nights performance and earlier actually said we dominated and deserved to win.

My first post was questioning some of the high danger chances credited.

 

[weems]

Kampf actually has a higher 5v5 shooting percentage (10.4%) than Tavares does (9.6%).

The bulk of the talent is in getting the chances. Very rarely you do get someone special like Matthews who consistently converts at a much higher rate, though.

Do you really think untalented players create as many chances as talented ones?

And not one single person believes that Kampf is actually a better shooter than John Tavares.

All I did was take the single scoring chance of Kampf (a one handed weak backhander) and compare it to Tavares who was sitting right in front of the net and actually got off a quality shot. I said that I think its rather hilarious that xG credits them bascailly as the same goal probability when in reality its not even close when looking at each scoring chance and applying the context it happened under.

No I agree over a big sample that xG most likely gives a good illustration of how a team is playing but in a one game sample of even a little bigger, it can be extremely flawed and these public sites routinely place shots/chances for in the incorrect spots.

 

[therealkoho]

It's exactly like that, and the "bad" team usually has a really solid counter attack. The Canucks had no such thing, which is why this game was so horrendous. It looked like a 60 minute Leafs powerplay.

Agreed

how many times did the Canucks control the puck in the O zone, for any extended periods ie 20+ seconds, once maybe twice on the PP, the Leafs spanked them in possession all night long, chances, etal

Demko beat them fair and square, the Leafs didn't beat themselves nor did they allow the Nux to beat them.

As I stated earlier stuff happens sometimes

 

[Stephen]

I didnt really post anything about last nights performance and earlier actually said we dominated and deserved to win.

My first post was questioning some of the high danger chances credited.

There's so much eye test context that you can apply to individual plays that would be considered high danger chances.

Here's a classic play: the Leafs are circling the net looking for an option to walk in front but the slot is all jammed up. The kick it to the point where the two defensemen pitch it back and forth before throwing something towards the net. The goalie makes a save on the first attempt and kicks it out where a Leaf player can barely control the puck and throws another attempt towards the net, missing completely. Or there's a rebound but also 2 defensemen tying up the Leaf player and all he does is shovel the puck along the ice into the goalie's paddle down.

From an attempted shot, high danger heat map and puck possession in zone time, it checks off all the boxes. But in real time, you could break down those moments in ever greater detail. Throwing some ideas out there but 1) angle and velocity of the rebound 2) number of Leaf players in the slot, or ratio of Leaf players to defenders 3) relative body positioning, distance to the goal where the Leaf player regains control of the rebound 4) number of defensemen within the line of the rebound shot 5) the hand of the shot of the Leaf player receiving the rebound 6) the handedness of the goalie.

There are so many ways to break down that quality of scoring chance that you would have an extremely hard time trying to capture in volume. So probably why you need the stats to go with the eye test and not trust one for the other or look at the probability of things occurring.

 

[weems]

There's so much eye test context that you can apply to individual plays that would be considered high danger chances.

Here's a classic play: the Leafs are circling the net looking for an option to walk in front but the slot is all jammed up. The kick it to the point where the two defensemen pitch it back and forth before throwing something towards the net. The goalie makes a save on the first attempt and kicks it out where a Leaf player can barely control the puck and throws another attempt towards the net, missing completely. Or there's a rebound but also 2 defensemen tying up the Leaf player and all he does is shovel the puck along the ice into the goalie's paddle down.

From an attempted shot, high danger heat map and puck possession in zone time, it checks off all the boxes. But in real time, you could break down those moments in ever greater detail. Throwing some ideas out there but 1) angle and velocity of the rebound 2) number of Leaf players in the slot, or ratio of Leaf players to defenders 3) relative body positioning, distance to the goal where the Leaf player regains control of the rebound 4) number of defensemen within the line of the rebound shot 5) the hand of the shot of the Leaf player receiving the rebound 6) the handedness of the goalie.

There are so many ways to break down that quality of scoring chance that you would have an extremely hard time trying to capture in volume. So probably why you need the stats to go with the eye test and not trust one for the other or look at the probability of things occurring.

Exactly.
I'm not trying to post about advanced stats and say that they are useless because that is very far from the truth and even as a non stats person, I've been trying to learn more about them because alot of them make sense.
But there is alot of context that is left out with some of them in short sample sizes and its becoming clear that these public sites arent super accurate always with posting shot locations and I question some of the value they give to certain chances.
Eye test plus being able to properly understand these advanced stats is the best and most balanced way to get the best analysis.

 

[Stephen]

Exactly.
I'm not trying to post about advanced stats and say that they are useless because that is very far from the truth and even as a non stats person, I've been trying to learn more about them because alot of them make sense.
But there is alot of context that is left out with some of them in short sample sizes and its becoming clear that these public sites arent super accurate always with posting shot locations and I question some of the value they give to certain chances.
Eye test plus being able to properly understand these advanced stats is the best and most balanced way to get the best analysis.

Plus the conclusion that we just "Got Goalied" isn't a particularly helpful one if you're trying to find ways to break through in these type of games - when your season might be on the line at some point, as it has been in the past. This isn't a referendum on analytical tools.

Maybe there's a super human goalie out there, but how do you break down the systematic defense assisting him to remove layers of his armour? How do you use skill most effectively to negate the park the bus tactic? How do you isolate the goalie from the defense in front of him to create vulnerabilities? Spend some time addressing these issues so you have a Plan B when you find yourself in this style of game instead of thinking you did everything you could.

 

[zeke]

And not one single person believes that Kampf is actually a better shooter than John Tavares.

All I did was take the single scoring chance of Kampf (a one handed weak backhander) and compare it to Tavares who was sitting right in front of the net and actually got off a quality shot. I said that I think its rather hilarious that xG credits them bascailly as the same goal probability when in reality its not even close when looking at each scoring chance and applying the context it happened under.

No I agree over a big sample that xG most likely gives a good illustration of how a team is playing but in a one game sample of even a little bigger, it can be extremely flawed and these public sites routinely place shots/chances for in the incorrect spots.

I think you're looking at it all wrong.

The chance is what is important. Most chances, even the best ones, don't go in. But a similar percentage of them go in no matter who is doing the shooting.

 

[zeke]

Plus the conclusion that we just "Got Goalied" isn't a particularly helpful one if you're trying to find ways to break through in these type of games - when your season might be on the line at some point, as it has been in the past. This isn't a referendum on analytical tools.

What do you mean by "these types of games"?

The Leafs have had no problem scoring this year. They've had no problem coming back in games they're down in. They have had no problem racking up goals in games they dominate. They've had no problem scoring on good goalies.

What special "type" was this particular game other than the fact that the Leafs couldn't convert on a typical amount of their chances this time?

 

[weems]

I think you're looking at it all wrong.

The chance is what is important. Most chances, even the best ones, don't go in. But a similar percentage of them go in no matter who is doing the shooting.

Then why does Muzzin's chance get graded as 3 times the xG grade compared to Mikheyev's that happened 2 seconds earlier?

One was taken point blank right in front of the net while the other was taken from around the goalline.

I understand xG really values rebound chances but this wasnt a normal rebound chance where Muzzin would fire it instantly or from a more favourable angle.

Muzzin gets credited 0.24 xG but Mikeheyev only credited 0.08 when Mikheyev's chance is significantly more dangerous.

 

[3blizzard4]

Vancouver is the Tim Tebow of a franchise. Go down on one see and thank the Lord for the miracle that happened last night.

 

[Racer88]

Calgary was much more of a perimeter game. The Nucks gave up a significant amount of prime, slot and down low chances last night we simply didn't convert on.

The Flames played a defensively sound game against us, that was relatively even. Last night the Nucks were bailed out by goaltending, and our goaltenders struggle in both games.

I agree that both games were slightly different but my frustration is if we want to be serious contenders they have to learn how to win these games

 

[Seath The Scaleless]

I agree that both games were slightly different but my frustration is if we want to be serious contenders they have to learn how to win these games

The Leafs aren't going to win every game nor score 4+ goals. They were 10-2-1 going into Thursday night, averaging well over 4 goals a game. They were overdue for some crappy luck.

Having said that, goaltending needs to regress the other way.

 

[Morbo]

Then why does Muzzin's chance get graded as 3 times the xG grade compared to Mikheyev's that happened 2 seconds earlier?

One was taken point blank right in front of the net while the other was taken from around the goalline.

I understand xG really values rebound chances but this wasnt a normal rebound chance where Muzzin would fire it instantly or from a more favourable angle.

Muzzin gets credited 0.24 xG but Mikeheyev only credited 0.08 when Mikheyev's chance is significantly more dangerous.

the stat is largely nonsense. don't worry, you are absolutely correct to question it.

 

[Gallagbi]

I agree that both games were slightly different but my frustration is if we want to be serious contenders they have to learn how to win these games

What do you mean by these games?

 

[Gary Nylund]

The problem is that you are posting 15-2 with zero context and being as biased as possible (leaving out 2 periods right?).

15-2 is ugly. Add in all the context you want and it will still be ugly. I left out the 2 periods because I started at the point where it all started to go to shit but included everything from that point on. I have at other times posted the 18-6 number which includes those 2 periods so LOL at you calling me "as biased as possible".

You know what I consider biased? It's you posting a bunch of numbers comparing contract values which included Marner and mostly a bunch of players who are much older than Marner and signed UFA contracts after winning multiple cups like Toews, Kopitar, Doughty etc. Good on you for posting some numbers of players more comparable after I called you out on it but that doesn't change the fact that your original list seemed like a blatant attempt to make Marner's contract a lot better than it really is (so biased, supporting an agenda).

The Leafs have sucked in game 7's. That's not bias, that's a fact. If you don't like 11-2 then make it 18-6 and take away the EN goals and it will look better but the numbers still suck, the numbers are pretty much an accurate reflection of what happened in those games but if you watched those games, you shouldn't need any numbers to tell you the obvious anyway.

 

[Gallagbi]

Plus the conclusion that we just "Got Goalied" isn't a particularly helpful one if you're trying to find ways to break through in these type of games - when your season might be on the line at some point, as it has been in the past. This isn't a referendum on analytical tools.

Maybe there's a super human goalie out there, but how do you break down the systematic defense assisting him to remove layers of his armour? How do you use skill most effectively to negate the park the bus tactic? How do you isolate the goalie from the defense in front of him to create vulnerabilities? Spend some time addressing these issues so you have a Plan B when you find yourself in this style of game instead of thinking you did everything you could.

You've pointed at offensive options as the way, but they covered the areas you wanted and it didn't pan out.

The way to win a game like last night is to not allow the 3 GA. So you either limit to no real chances against, 2 goals were off poor Leafs decisions, or get the saves. Which isn't a crazy ask and still would have left Mraz as the second best goalie last night.

Basically buy time until you break through. We had chances, and in a variety of ways, so let those build and limit opportunities against. Our attack wasn't singular or limited.

 

[Gabriel426]

15-2 is ugly. Add in all the context you want and it will still be ugly. I left out the 2 periods because I started at the point where it all started to go to shit but included everything from that point on. I have at other times posted the 18-6 number which includes those 2 periods so LOL at you calling me "as biased as possible".

You know what I consider biased? It's you posting a bunch of numbers comparing contract values which included Marner and mostly a bunch of players who are much older than Marner and signed UFA contracts after winning multiple cups like Toews, Kopitar, Doughty etc. Good on you for posting some numbers of players more comparable after I called you out on it but that doesn't change the fact that your original list seemed like a blatant attempt to make Marner's contract a lot better than it really is (so biased, supporting an agenda).

The Leafs have sucked in game 7's. That's not bias, that's a fact. If you don't like 11-2 then make it 18-6 and take away the EN goals and it will look better but the numbers still suck, the numbers are pretty much an accurate reflection of what happened in those games but if you watched those games, you shouldn't need any numbers to tell you the obvious anyway.

Well, when was the last time the Leafs won a Game 7 or series deciding game. Certainly not with the current core.

 

[Gabriel426]

You've pointed at offensive options as the way, but they covered the areas you wanted and it didn't pan out.

The way to win a game like last night is to not allow the 3 GA. So you either limit to no real chances against, 2 goals were off poor Leafs decisions, or get the saves. Which isn't a crazy ask and still would have left Mraz as the second best goalie last night.

Basically buy time until you break through. We had chances, and in a variety of ways, so let those build and limit opportunities against. Our attack wasn't singular or limited.

Or just run the opposing goalie to get him off his game.

 

[keonsbitterness]

Or just run the opposing goalie to get him off his game.

Not many still do that and are able to play a regular shift. Gallagher?