Speculation: LA Kings News, Rumors, Roster Thread 2022-23 Season

[lumbergh]

You and I already had this convo so not gonna rehash it haha. I've come around a little on the idea that it can measure high performers thus far but I'd still argue it has zero predictive value and if anything someone on an absolute PDO bender is likely to come back to earth as it's unlikely they'll keep up a 20% shooting % or a .970 on-ice save %.

IE we all expected/expect Vilardi's shooting % to not be 35% all year; if his PDO dips because of that, is he 'not performing?'

That's all we mean by regression to the mean. I think we can generally agree that the best teams will have slightly higher than 1 PDO and the worst slightly lower but extreme outliers are a cause for investigation.

I think you might be mistaking what PDO is. It's the sum of the on ice shooting percentage and save percentage for the whole team while a player is on the ice. It's not dependent on just Vilardi's shooting percentage. It's the Kings' shooting percentage while Vilardi is on the ice. That's what makes PDO such an interesting stat. It tells you if the player is helping the team as a whole on the offensive side and defensive side. Vilardi can help by shooting and scoring, but he can also help by playing defense and gaining possession.

You also make an important point. Players have ups and downs. This is why you should only look at these advanced stats with a decent sample size. Also why I did this analysis at the half season mark.

I think you're overestimating everyone's understanding of "regression to the mean". ATKM's post tells me that not everyone means the same thing, and that's okay. Regression to the mean says that if you flip a coin 5 times and it comes up heads five times, the next five are less likely to all come up heads. In fact, the probability you will flip heads on the next flip is still 50%, even if you flipped heads 5 times in a row. It relies on the assumption that the thing you're measuring is random. Hockey is not random, and you don't know what Vilardi's "mean" shooting percentage truly is. #1 because it's not random, and #2 because it doesn't exist. You can predict Vilardi's shooting percentage will go down over the next 41 games because of "regression to the mean", but #1 that's based on the fact that his shooting percentage is higher than average and #2 maybe it won't.

I don't run through these numbers to predict what will happen in the next game or next 41 games. I do it to find the unsung heroes of the team. What I can tell you is that, even though they aren't leaders on the point totals, Alex Iafallo and Mikey Anderson are doing the things that are helping the team win and that Lemieux and Grundstrom are on the flip side.

I don't mean to get all preachy about this stuff. I'm a scientist and I have a decent understanding of statistics. Maybe I'd be better off starting a hockey blog...

 

[BringTheReign]

Just remember you’ll be replacing Kopitar’s cap hit with Matthews cap hit, so don’t count on that space.

Matthews
Byfield
Danault

Basically be a newer version of
Crosby
Malkin
J. Staal.

It sounds insane to say, but you have to think between our upswing, Kopi's $10M hit finishing, and the cap going up significantly once escrow is paid off...wouldn't that be something. I think he'll be 27 by the time that season would start though, and he's not the sturdiest player in terms of injury history.

All that being said, you have to take a swing when a player of that caliber is available. I'm guessing at that point he's going to want 7 years rather than taking a shorter deal again, but you never know. If he wants 3, 4, or 5, you still do it--it's a UFA signing so the only loss is cap space (though I know @All The Kings Men has thoughts about acquisitions being labeled as "free" in that sense versus trades to acquire ).

 

[bmr]

Well we're halfway through! 41 games in, it's pretty easy to project what a full season would look like at this pace. The Kings are on a 100-point pace right now, which will certainly get them to the playoffs. I thought I'd take a look at some of the numbers to see where some of the players stand. The numbers I pay attention to are Goals For % (GF%) and PDO (Team shooting % + save % while a player is on the ice). These numbers can tell you which players are contributing to winning ways outside of the points.

Even Strength Goals For % (at least 100 minutes)
Studs:
Alex Iafallo, 14 GF 7 GA, 67%
Mikey Anderson, 34 GF 23 GA, 60%
Gabe Vilardi, 24 GF 18 GA, 57%

Duds:
Carl Grundstrom, 16 GF 23 GA 41%
Brandon Lemieux, 6 GF 8 GA, 43%
Sean Durzi, 25 GF 31 GA, 45%

Even strength PDO (at least 100 minutes):
Studs:
Alex Iafallo, 1.039
Quinton Byfield, 1.033
Mikey Anderson 1.017

Duds:
Brendan Lemieux, 0.936
Brandt Clarke, 0.953
Carl Grundstrom, 0.964

The two standouts are Alex Iafallo and Mikey Anderson so far this season. The Kings are 11-7 in the games he has played and 11-12 in the games he missed. As hard as Mikey Anderson's hands are, he is bringing it all over the ice and definitely not dragging Doughty down.

On the flip side, Carl Grundstrom and Brendan Lemieux have been surprise duds. Grundstrom really looked like he might take the next step from the end of the last season, but it turns out that step is backward. Lemieux has taken some dumb penalties to take himself off the ice, but when he's on the ice he seems to get scored on a lot. Here's a guy who has great Corsi, Fenwick, xGF numbers, but ends up way underwater in actual results.

Our three big boys, Quinton Byfield, Gabe Vilardi, and Arthur Kaliyev, are generally doing well and growing into solid players, according to the same numbers. Vilardi and Kaliyev on the edge of breaking out. I'd love to see the three of them together on a line one day. Maybe it'll be two years from now, but they're gonna be awesome.

Great post. Iafallo and Anderson look like the most defensively responsible players on the ice this year just on the eye test alone.

Lemieux is only helpful if he limits the penalties he is taking and adds a scoring touch, which is what he was doing last year before getting injured. Grundstrom obviously has a scoring touch, but he has been a hot mess this year. Durzi is a high upside offensive player but just makes too many bad reads on defense. I think he needs to get included in a trade.

 

[lumbergh]

Great post. Iafallo and Anderson look like the most defensively responsible players on the ice this year just on the eye test alone.

Lemieux is only helpful if he limits the penalties he is taking and adds a scoring touch, which is what he was doing last year before getting injured. Grundstrom obviously has a scoring touch, but he has been a hot mess this year. Durzi is a high upside offensive player but just makes too many bad reads on defense. I think he needs to get included in a trade.

I just want to emphasize that what I posted are even strength numbers. Durzi is an interesting case because he is consistently under water at even strength, pointing to his defensive deficiencies. But he does make up for a lot of his deficiencies on the power play. On the whole his GF% isn't bad (49%) because he generates a ton of offense with a man advantage. He's generally a very high event player, a two-edged sword, one might say.

If you still want to trade him I totally understand.

 

[Raccoon Jesus]

I think you might be mistaking what PDO is. It's the sum of the on ice shooting percentage and save percentage for the whole team while a player is on the ice. It's not dependent on just Vilardi's shooting percentage. It's the Kings' shooting percentage while Vilardi is on the ice. That's what makes PDO such an interesting stat. It tells you if the player is helping the team as a whole on the offensive side and defensive side. Vilardi can help by shooting and scoring, but he can also help by playing defense and gaining possession.

You also make an important point. Players have ups and downs. This is why you should only look at these advanced stats with a decent sample size. Also why I did this analysis at the half season mark.

I think you're overestimating everyone's understanding of "regression to the mean". ATKM's post tells me that not everyone means the same thing, and that's okay. Regression to the mean says that if you flip a coin 5 times and it comes up heads five times, the next five are less likely to all come up heads. In fact, the probability you will flip heads on the next flip is still 50%, even if you flipped heads 5 times in a row. It relies on the assumption that the thing you're measuring is random. Hockey is not random, and you don't know what Vilardi's "mean" shooting percentage truly is. #1 because it's not random, and #2 because it doesn't exist. You can predict Vilardi's shooting percentage will go down over the next 41 games because of "regression to the mean", but #1 that's based on the fact that his shooting percentage is higher than average and #2 maybe it won't.

I don't run through these numbers to predict what will happen in the next game or next 41 games. I do it to find the unsung heroes of the team. What I can tell you is that, even though they aren't leaders on the point totals, Alex Iafallo and Mikey Anderson are doing the things that are helping the team win and that Lemieux and Grundstrom are on the flip side.

I don't mean to get all preachy about this stuff. I'm a scientist and I have a decent understanding of statistics. Maybe I'd be better off starting a hockey blog...

Totally fair and good explanation. I think to be clear I'm talking in levels of likelinhood. Your'e absolutely right, it may NOT come down. Hell, it COULD go up! But I'm not sure what the highest on-ice shooting % in a season is/was anyway. I'm probably splitting hairs anyway and trying to get ahead of it a bit. I do like the idea that those guys all do little things that have ripple effects. Hell, in conversations re: Erik Karlsson it came up that he regularly had the lowest or bottom-3 on ice save % on the team all years except 2017. Surely, that's not just 15 years of coincidence.

And anything that helps find value in Mikey and other defensive dmen is always right by me, lol.

 

[Statto]

Ugh, this again. Listen, just because the inventor of the PDO stat says it's a measure of puck luck doesn't mean it's a measure of puck luck. PDO is indeed on ice Sv% + SH%, but good teams have good PDOs and good players contribute to good PDO. They have a positive correlation.

If you take a look over years and years of data, no "advanced stat" matches team success (as measured by point %) as well as PDO. Why? You have to put the puck in their net and keep the puck out of your net to win. That's essentially what PDO measures.

Some players do the little things that don't show up on the scoresheet but help the team in subtle ways. Maybe they play sound positional defense so that shots on their goalie come from less dangerous areas (Mikey Anderson is a good example). Maybe they are good at keeping the forecheck alive (Alex Iafallo, anyone). They don't always get an assist or a goal on those plays, but those things add up over the course of a season to contribute to a team's success. PDO is an efficient and easily understandable way of finding those unsung heroes of a team.

When I see posters use the terms "puck luck" and "regression towards the mean" I know it's a misunderstanding of the use of numbers. It relies on a faulty assumption that the game of hockey is random. That by chance, one shot out of every 10 or so will go in the net. Everyone knows it is NOT random. There are good teams and there are bad teams. There are good players and bad players. There are good shooters and bad shooters. There are good goalies and bad goalies. Again, and I will say this until I'm blue in the face, PDO is not a measure of "puck luck". It is a measure of general on ice quality when spread out over many games.

Like all stats on its own it means very little. So yes on its own it absolutely is ‘puck luck‘, certainly in small samples (although I agree that it’s probably not the best term). If you use it with other stats you can build a proper picture, because over the short term things can be very skewed by poor or great goaltending (I’m being overly simplistic) and for individual offensive skaters spells of being snake bitten or on fire skew it heavily.

I‘ve run the stats on a European Elite team and I’d never use a single stat to conclude anything. I do understand what you’re saying about the stat to support any view or analysis. I used to analyse most of this stuff, before it was ‘advanced stats’ and when looking at anything I’d always review, then present a minimum of 5 data points. I usually wanted more but most of the coaches struggled with too much info. It’s too easy to look at one stat, draw a conclusion and the have additional data contradicting it. So as you suggest where the shots are coming from, success on the forecheck etc. All feed into that bigger picture to allow a conclusion.

I don’t think we disagree massively, just slightly different perspective and context.

 

[Maynard]

You guys and your fancy stats.

I just turn on the TV and say "That guy looks like shit!"

I say this too but it’s usually during the intermissions.

 

[Zippy316]

Has there been any word about Brandt Clarke’s status following the WJC?

I see he got his max 9 games before he can burn a year and assume a conditioning stint to play in the AHL. Is the plan for him to rejoin the NHL or go back to juniors?

Main curiosity here is his brother Graeme has had quite a run the last month of so with the Devils AHL team. If Brandt rejoins the Kings, there’s a chance Graeme could make his NHL debut against his brother or at least be called up around that time. Would make for a nice story if it’s a possibility.

 

[Brodeur]

Has there been any word about Brandt Clarke’s status following the WJC?

I see he got his max 9 games before he can burn a year and assume a conditioning stint to play in the AHL. Is the plan for him to rejoin the NHL or go back to juniors?

Main curiosity here is his brother Graeme has had quite a run the last month of so with the Devils AHL team. If Brandt rejoins the Kings, there’s a chance Graeme could make his NHL debut against his brother or at least be called up around that time. Would make for a nice story if it’s a possibility.

Saw this:





Would be interesting if Graeme were called up for that game in LA next weekend, but hopefully Ondrej Palat is cleared soon. But I'd be a little surprised if LA burned the year for Brandt now that he's at the limit.





I went to a March 2020 game when the Kings called up Mikey Anderson for his first game against his brother who was still on New Jersey at the time. The teams coordinated to have them do a lap before warmups.

 

[King'sPawn]

I say this too but it’s usually during the intermissions.

I say it when I look in the mirror and now I want to cry thanks for reminding me.

 

[All The Kings Men]

Has there been any word about Brandt Clarke’s status following the WJC?

I see he got his max 9 games before he can burn a year and assume a conditioning stint to play in the AHL. Is the plan for him to rejoin the NHL or go back to juniors?

Main curiosity here is his brother Graeme has had quite a run the last month of so with the Devils AHL team. If Brandt rejoins the Kings, there’s a chance Graeme could make his NHL debut against his brother or at least be called up around that time. Would make for a nice story if it’s a possibility.

No.

There has been no official word from the organization on the status of Brandt Clarke following the WJC.

Anything you see or hear will be speculation.

 

[Fishhead]

No.

There has been no official word from the organization on the status of Brandt Clarke following the WJC.

Anything you see or hear will be speculation.

One thing this organization is really good at is being tight-lipped about things, almost nothing gets out. I shudder to think of the handshake you have to memorize to get into the meetings. Probably 20 different components with an ass-bump in the middle.

 

[All The Kings Men]

One thing this organization is really good at is being tight-lipped about things, almost nothing gets out. I shudder to think of the handshake you have to memorize to get into the meetings. Probably 20 different components with an ass-bump in the middle.

I don't think it's as draconian as some might imagine.

We've talked about Trevor Moore for like a week now and it's always "is he skating is he working out yes or no" but none of us has had the presence of mind of cajones to actually ask "what happened?"

We know with Kaliyev. He's on IR.

Moore? Zero idea. Just keeps being not important enough to make it into the list of things we talk about.


Then there's some days where we randomly get all sorts of insight that we probably weren't "meant" to.

 

[Zippy316]

No.

There has been no official word from the organization on the status of Brandt Clarke following the WJC.

Anything you see or hear will be speculation.

Any indication they’re willing to burn the contract year by playing him in the next game? Even just from a roster standpoint.

I assume he’s going back to juniors as that’s the most logical. Made sense to get his feet wet professionally with 9 games + conditioning stint leading perfectly into the world juniors.

Just was curious to see if there was any possibilities of the brothers playing each other.

 

[All The Kings Men]

Any indication they’re willing to burn the contract year by playing him in the next game? Even just from a roster standpoint.

I assume he’s going back to juniors as that’s the most logical. Made sense to get his feet wet professionally with 9 games + conditioning stint leading perfectly into the world juniors.

Just was curious to see if there was any possibilities of the brothers playing each other.

I honestly have no idea.

I work for the team, I occasionally talk to people involved in running the team and I'm at every morning skate, team practice and postgame press conference that happens at home (I don't travel with the team).

I have zero idea what they're going to do. I've seen and heard plenty of people speculate that he'll be headed back to Barrie because of the obvious lineup limitations. I've also heard, over time and on the record, members of the Kings front office staff lay out in no uncertain terms their concerns with his development should he spend considerable time at a lower level of competition.

 

[Brodeur]

Just was curious to see if there was any possibilities of the brothers playing each other.

Devils just activated Ondrej Palat a few minutes ago and sent down Tyce Thompson. Clarke getting called up by next weekend would seem unlikely. Alex Holtz finally had a decent game last night, but he might be fighting with Jesper Boqvist for the 12th forward spot. Barring another injury, I don't know if Clarke warrants a call up even with his strong play in Utica.

 

[Fishhead]

I don't think it's as draconian as some might imagine.

We've talked about Trevor Moore for like a week now and it's always "is he skating is he working out yes or no" but none of us has had the presence of mind of cajones to actually ask "what happened?"

We know with Kaliyev. He's on IR.

Moore? Zero idea. Just keeps being not important enough to make it into the list of things we talk about.


Then there's some days where we randomly get all sorts of insight that we probably weren't "meant" to.

I'm sure a lot of that is there is no pressure for the information to actually get out. Here, we all want to know what's going on with Clarke, Moore, etc. so we sleep better at night. But it's not like it's the entire city of Montreal. In a way I appreciate that because it means we get better, more straight-forward news from our sources. In a highly-competitive markets writers tend to try and separate themselves with things like outlandish trade predictions or "a move is imminent" statements to get people coming back. We don't get that here and it's fantastic. I know a lot of us picture the NHL draft and deadline like some crazy scene out of the stock market where everyone is yelling and wheeling and dealing, but in reality it's more like a competitive bridge game.

Still, I'm going to visualize Blake not letting Donnelly into a meeting because he forgot the ass-bump.

 

[kilowatt]

You guys and your fancy stats.

I just turn on the TV and say "That guy looks like shit!"

I say that every morning in the mirror too

Edit: that's what I get for not reading the rest of the posts, we're just a bunch of ugly dorks over here

 

[Bandit]

I say this too but it’s usually during the intermissions.

That’s when I turn OFF the tv.

 

[Seattle King]

I'm sure a lot of that is there is no pressure for the information to actually get out. Here, we all want to know what's going on with Clarke, Moore, etc. so we sleep better at night. But it's not like it's the entire city of Montreal. In a way I appreciate that because it means we get better, more straight-forward news from our sources. In a highly-competitive markets writers tend to try and separate themselves with things like outlandish trade predictions or "a move is imminent" statements to get people coming back. We don't get that here and it's fantastic. I know a lot of us picture the NHL draft and deadline like some crazy scene out of the stock market where everyone is yelling and wheeling and dealing, but in reality it's more like a competitive bridge game.

Still, I'm going to visualize Blake not letting Donnelly into a meeting because he forgot the ass-bump.

Probably more likely to be excluded because he forgot to arrange for the ass and the bumps.

 

[Bandit]

One thing this organization is really good at is being tight-lipped about things, almost nothing gets out. I shudder to think of the handshake you have to memorize to get into the meetings. Probably 20 different components with an ass-bump in the middle.

That’s because nobody covers the team that isn’t licking their boots. Easy to be secretive when nobody’s watching.

Nothing like dicking your best defensive prospect in 15 years around from LA to Newfoundland.

 

[bmr]

I just want to emphasize that what I posted are even strength numbers. Durzi is an interesting case because he is consistently under water at even strength, pointing to his defensive deficiencies. But he does make up for a lot of his deficiencies on the power play. On the whole his GF% isn't bad (49%) because he generates a ton of offense with a man advantage. He's generally a very high event player, a two-edged sword, one might say.

If you still want to trade him I totally understand.

I'm just a lot higher on Spence. I think he's the smarter player, more defensively responsible and still extremely capable on the PP. IMHO, it's Doughty/Spence/Clarke on the right side. Anderson, Bjornfot and what i'm hoping to be a large, mobile D to fill out the left.

 

[crassbonanza]

Look, xGF% is getting better and better as a stat, but can anyone explain to me how it's calculated? If you told me Clarke has an xGF% of 62%, I would ask, what does that mean? How did he help the team? If his expected goals for percentage is 62%, why is his actual goals for percentage 47%? The team should be outscoring its opponents when he's on the ice, but it's actually the reverse. Notice how Brodzinski is expected to get outscored, but actually has 5 GF 2 GA? It's because his PDO is off the charts high. (Side lesson: Always be careful with small sample sizes)

If we knew how xGF% was being calculated, you'd probably be shocked at how complex the model has become over the years. It turns out that hockey is quite difficult to quantify. Tell me after you've read this blog if you have any idea what to do with xGF%.

Expected Goals (xG) Models Explained

<em>(Editors Note: <strong><a href="https://twitter.com/YuriGelfat">Yu...

jetsnation.ca

Use xGF% all you want. I respect it. I just don't get a lot out of it, especially when it's used on a single game on a single player. It's a mystery stat that doesn't tell me much. Over a half a season it might tell me that one team is better than another, but I can just read the standings.

I have issues with xGF%, I don't think it is a perfect stat at all. Combined with CF%, SF% and GF% it can provide an additional piece of the puzzle. However I do not think PDO is a very useful stat for evaluation. It seems that you are arguing that PDO is a representation of skill, which it could be in many cases, but the original usage of PDO was to represent luck. That the on ice results differed from the expected on ice results and that had to do with variance. Having a high PDO ws thought to be a marker that a player or team were primed for a decline. The fact that there could be two completely opposite explanations for what a high PDO represents, kind of shows that it is not a very useful. If you compare Byfield to Bjornfot you will see the issue. Both Bjornfot and Byfield have the exact same GF% at 50, however Bjornfot has significantly higher possession numbers, leading to Bjornfot having a .955 PDO and Byfield having a 1.026 PDO. They have the same on ice results, expect one was driving possession more, you are crediting the one with worse possession as being the better player.

 

[Cianide]

Walker? Sean Walker?

Technically, if Sean Walker was worth his weight in gold, his contract would be about $5.75 million. Thankfully that's not the case, the guy just isn't very good.

What game are you watching?

 

[kilowatt]

What game are you watching?

What game are you watching? I'm not trying to be antagonistic here, but he's been our least effective defenseman and arguably our least effective player this year (either him or Grundstrom). Beyond just the eye test, he's got the worst statistics among regular defensemen with the least ice time.