Speculation: LA Kings News, Rumors, Roster Thread 2022-23 Season
- Thread starter kingsboy11
- Start date
- Status
King'sPawn
Enjoy the chaos
- Jul 1, 2003
- 22,917
- 23,479
He seemed fine to Todd, after the game. It was just the timing of the hit that stopped him from being able to return.
From what I heard about the CLB negotiations with Gavrikov it was years, not so much the $$$. They aren't afraid to commit long term in LA obviously. It's not like CLB could have thrown money at him either, they are committed to 3 really highly paid guys for a long time and have a big group of RFAs coming up.
All but the bottom 5 teams in the league are close to the cap so unless you want to play for a rebuilding group teams aren't going to be able to back up the money truck. I would think a solid offer with a bit longer term would get it done. Just gotta make a bit of space by paring down the roster.
All but the bottom 5 teams in the league are close to the cap so unless you want to play for a rebuilding group teams aren't going to be able to back up the money truck. I would think a solid offer with a bit longer term would get it done. Just gotta make a bit of space by paring down the roster.
All The Kings Men
Registered User
- Apr 7, 2016
- 2,297
- 5,486
one of my issues with "advanced analytics" is that the publicly available ones (corsi, fenwick, xG etc etc) aren't even fully understood on their surface or wielded with any kind of nuance but then on TOP of that there are boatloads and boatloads of more specific in depth stats that might completely contradict the narratives constructed by those enterprising fans that love narrative construction but don't have the patience or perhaps even the awareness to realize that they're speaking with supreme confidence about things they may be totally wrong about.
I'm not going to sit here and pretend to tell you that the LA Kings defensive corps is perfect and beyond reproach but when I see all the conversation on this board about metrics and the "obvious" flaws in the Kings handling of personnel I almost never see any discussion about the defensive metrics that the Kings are almost dominant in.
Clearing rebounds (yes I know... shots need to get through to create the rebonds to be cleared)
blocking shots (yes shot attempts need to happen to block shots)
preventing passes to the slot
precenting slot shots
preventing high danger scoring chances
and the list goes on and on
I like to think that I'm open minded enough that if someone presented a patient argument as to why none of that matters or how any specific part of it is tilted or lacking context or nuance I would listen eagerly because ultimately I'm just trying to understand what I'm watching everyday... but I just don't see a lot of that.
I see a bunch of people telling me that Edler and Walker are literally unplayable. I see people telling me that playing Sean Durzi on the left side is tantamount to coaching malpractice.
Then I see people saying "well the standings say the Kings are good so you're dumb" which is not a cogent or mature argument BUT is actually momentarily compelling because... ya know... "scoreboard" and all that....
Whatever the criticisms of specific individual players may be the team results are, at least for now, very positive so I just don't understand the supreme confidence the people saying that 3/7 if not more of the LA Kings defensive corps isn't even replacement level.
Those player cards from the Athletic don't seem to speak to the collective impact.
I dunno... doesn't make sense to me but I could be wrong.
There are a ton of metrics out there, some far more useful than others. It's just as important to look at situations for each metric as well. Zone entries and denying entries is a very useful tool, but it's even more useful to look at those 5v5, 4v4, PP, etc to get a handle on how well a player does in each situation. Of course, you can just watch them and come to a similar conclusion.
The rabbit hole is you can use stats to explain things, but it's not always cause/effect and all that. But they are extremely useful when used in context with quality of competition. Edler and Walker aren't a bad pairing at all when rolled out as a 3rd pair. They have solid metrics in that situation - but roll them out against top and elite talent and their metrics dive as they get creamed. It's just one part of the equation. For example, Edler gets creamed against top competition and is fine against lower. Bjornfot is also good against lower, but elevates and plays better against top competition. Should he play over Edler? Well, you can't dismiss size, shot blocking, playstyle, etc. so that's always where the debate ends up.
I find all that most useful when paired with WOWY, because then you can see the synergy between guys. McNabb got roasted for a lot of things when he was here, but with Doughty he was fantastic. Both of their numbers shot through the roof when they were on the ice together and sure enough, Doughty won a Norris one of those years. We see them now with Roy and Gavrikov, it's just glaringly better. If it continues through a bigger sample size that's a pretty good indicator that it's worth doing everything you can to keep those two together.
I'm sure I didn't answer a whole lot there, but maybe explains why people use metrics sometimes to a fault. I'm sure a lot of the time I'm spouting off stats it's not as directly related to what I am seeing as I think it is. But hey, I'm scientifically trained so failure is always an option.
Statto can chime in, I think they are the most knowledgeable around here with that kind of stuff.
The rabbit hole is you can use stats to explain things, but it's not always cause/effect and all that. But they are extremely useful when used in context with quality of competition. Edler and Walker aren't a bad pairing at all when rolled out as a 3rd pair. They have solid metrics in that situation - but roll them out against top and elite talent and their metrics dive as they get creamed. It's just one part of the equation. For example, Edler gets creamed against top competition and is fine against lower. Bjornfot is also good against lower, but elevates and plays better against top competition. Should he play over Edler? Well, you can't dismiss size, shot blocking, playstyle, etc. so that's always where the debate ends up.
I find all that most useful when paired with WOWY, because then you can see the synergy between guys. McNabb got roasted for a lot of things when he was here, but with Doughty he was fantastic. Both of their numbers shot through the roof when they were on the ice together and sure enough, Doughty won a Norris one of those years. We see them now with Roy and Gavrikov, it's just glaringly better. If it continues through a bigger sample size that's a pretty good indicator that it's worth doing everything you can to keep those two together.
I'm sure I didn't answer a whole lot there, but maybe explains why people use metrics sometimes to a fault. I'm sure a lot of the time I'm spouting off stats it's not as directly related to what I am seeing as I think it is. But hey, I'm scientifically trained so failure is always an option.
Statto can chime in, I think they are the most knowledgeable around here with that kind of stuff.
Eye test seems good to me. They have been playing teams that are mostly out of it recently. Looking forward to the measuring stick that are the Avs on Thursday. I think they will be ready for it.
Also Jesse, did you see Cassidy call a time out after Quick let in the third goal against Montreal?
bouncesonly
Registered User
Is there a formal definition or explanation of "high danger"? Is it even strength only? Is it anywhere near the crease? Is it when goalie is so out of position he'd be found in a different arena?
no where did i thnk it did, just not interested in spending to the cap just for the sake of it. Especially since it would create more of a log jam.
If anything I would rather use that cap space to broker another teams deal.
It's based on zones and regions on the ice based on statistical success, and shots are assigned a "danger" number. So something from the point flung at the net might be a 1, where as something in the middle of the slot is a 3. If a shot is blocked, it takes this down one value, if it's a rebound, it increases it one value. A shot that is a 1 is low danger, while a 3 would be high danger.
Let's say Doughty winds up by the blueline in the middle and lets a slapper go. Based on the zone that's probably a 2. Not in close but in the middle of the ice where he can see most of the 6x4. Of course, he took too damn long and it gets blocked to the side by a defender, rendering it a 1. It might bounce right to someone but only rarely, so low danger.
Now instead he dekes and moves into the high slot for a wrist shot - that's a 3 because it's a prime area. If it's blocked, well it's still a 2 because the tighter angle means it's more likely to pinball around and deflect so someone. If the goalie kicks it out, it's a 4 because of the prime location of the original shot. The goalie can't just squeeze off a post like it's coming from the outside, and it's difficult to predict where DD is going to shoot. Therefore it becomes a really high danger chance because there is a high likelihood of the goalie being commited and out of position and that free puck can squirt anywhere.
As I understand it, 3 and above is considered high. For most sites you can sort the stat by different situations (5v5, 5v4, ES, goalie pulled, etc.)
bouncesonly
Registered User
Is the quality of the player factored in? I assume a breakaway is a high danger chance but what if the person on the breakaway is Wagner?
Sol
Smile
- Jun 30, 2017
- 24,503
- 20,631
It works like this. A high danger chance is quantified a spot on the ice where you score from more often hence being high danger. How that reflects on their individual stat analytics. So you can see Wanger getting high scoring chances but their analytics show they can’t convert
- May 9, 2014
- 5,699
- 8,054
I think you make a valid point about results and that essentially the group is performing at a good level. Results are always the most important stat and vetoes every other one and I agree that to form the valid picture you need all the stats (each weighted differently) plus the eye test.
I think the stuff saying Edler and Walker are unplayable is hyperbole. My personal argument has always been that had Spence and Bjornfot been afforded the same opportunity they’d be better… even more so Clarke. Obviously that’s no more than a subjective opinion and I can’t prove it. However even if we win the cup, that opinion won’t change… I’ll just think we’d have won it in more style with those guys.
It no longer matters. The direction has been set and I’ll support everyone until the end… whilst occasionally forgetting I’m doing that and rejoining the debate again… and again
King'sPawn
Enjoy the chaos
- Jul 1, 2003
- 22,917
- 23,479
Fishhead, Chazz, and others can explain much better than I can (as I'm still trying to build my understanding to supplement what I see versus what the stats say), but I don't think the quality of the player is accounted for. Mostly because the people prescribing the numbers doesn't know every single NHL player who steps on the ice in depth. At least as far as evaluating defense is related to this.
For example, let's say the Kings played 7th round pick Kaleb Lawrence at their next game, and he has a breakaway. They don't know how well he typically does on breakaways relative to, say, Samuel Fagemo.
Conversely, a quality of goaltending can't really be factored in to high danger vs. low danger chances. Playing against Dan Cloutier can't make every shot from behind the red line high danger, no matter how justified it is. The analytics don't say "Well, McDavid had a breakaway and missed, but it was less dangerous because he was up against Vasilevsky."
The individual's end result is still tabulated separately, as @Sol explains.
Schrute farms
LA Kings: new GM wanted -- inquire within
- Jul 7, 2020
- 2,550
- 4,621
BigKing
Blake Out of Hell III: Back in to Hell
- May 9, 2014
- 5,699
- 8,054
You make some good points.
I’ll complicate it a bit further and say that the importance of each stat will vary on the system. It will also vary in terms of how a coach want’s certain details to be focused on. For example for a certain goalie he may want shots from certain positions, so play is funnelled to certain areas. You’ll find teams take their own stats and I’d bet have their own interpretations of various stats, especially high danger chances. I would often have to tweak my definitions of various things whenever we changed coach.
I recall a great argument (when the standard dropped) with me insisting that body contact (no matter how hard) is not a hit… for a hit you need to separate man from puck or if you don’t do so you have to still be in the play. If you bounce off and the guy goes through with the puck it’s not a hit. I won the argument. However there are lots of valid variations depending on what the coach looks for. The important stats for the Kings are the important ones for TMac… end of story.
For the record I like the usage and pairing stats. By definition they are a stat that brings nuance before you start. Obviously they need sample size but I think it’s a smaller sample than many of the others because of the specific context.
In all honesty I don’t have time to delve into them (advanced analysis) much these days. The reason bring is if I start I will go down the rabbit hole for literal days. When I did some giveaway analysis earlier this season I did so for about 12 hrs.. just damn giveaways. In part to verify things but also because I see something else interesting and go off on tangents (oooo shiny). I get obsessed and it takes time I don’t have, so I end up sleep deprived. I’m a bit odd like that. Hence I try to stay away apart from the odd relapse.
- May 9, 2014
- 5,699
- 8,054
In short no, too subjective. It also doesn’t matter because you are more interested in how the chance occurred anyway (what caused the breakdown).
However if looking at the goaltender you might separate breakaways from normal high danger chances because they are contextually different in terms of the type of chance. The same way that you’d analyse an offensive giveaway differently to a defensive zone one.
I’d certainly want to split it when analysing goaltender performance on chances. Definitely not the same as a normal 5/5 chance. I’d imagine teams do that, as well as breaking up high danger chances into off the rush and from set up in the OZ. Also further split into 5/5, 5/4 etc.
See… rabbit hole… LOL.
TruKingFan
Two-Time Cup Winner
This illustrates what Jesse said in his first paragraph above. All that explanation is really just for one stat, and there are hundreds of them. High-danger is one of the more common and useful ones, so it's relatively easy to understand. But some of those obscure ones are brain-turners.
Thankfully a lot of stats are being compiled into easier-to-use versions, like expected events above average. For goalies, I'd say GSAA (goals saved above average) is the best indicator of how "good" one is right now, with a good sample size it almost always aligns with who is elite and who is struggling. It's a fantastic stat to look at when trying to figure out if it's a case where a good goalie is on a bad team or a mediocre goalie on a good team. It is essentially how many goal-stopping saves are made that the goalie had no business making. It's not always saves like those old Quick post-to-posts, telepathic saves, or when he robbed Marleau. It's tied in with shot quality so it tells you exactly who tends to let bad goals in and who saves pretty much every easy shot and then some.
Thankfully a lot of stats are being compiled into easier-to-use versions, like expected events above average. For goalies, I'd say GSAA (goals saved above average) is the best indicator of how "good" one is right now, with a good sample size it almost always aligns with who is elite and who is struggling. It's a fantastic stat to look at when trying to figure out if it's a case where a good goalie is on a bad team or a mediocre goalie on a good team. It is essentially how many goal-stopping saves are made that the goalie had no business making. It's not always saves like those old Quick post-to-posts, telepathic saves, or when he robbed Marleau. It's tied in with shot quality so it tells you exactly who tends to let bad goals in and who saves pretty much every easy shot and then some.
Chazz Reinhold
Registered User
It looks like others have answered but I'm lazy and tired and haven't finished reading their responses yet so this may be a little repetitive. You used high-danger chance but I'm going to switch to expected goals to answer just because expected goals are an attempt to quantify the value of a particular shot (which I'll touch on more below), including those that people might call "high-danger" or "Grade-A" chances.
Quality of player isn't directly factored in in the sense that McDavid will get a higher value of an "expected goal" amount than Wagner based each having his own breakaway from the blueline in on the net where each shoots from 5 feet out from the crease (for example).
It is indirectly factored in to some extent based on how the value of an expected goal is calculated (which varies from model to model for the publicly available models like Money Puck, Natural Stat Trick, Hockey Viz, etc. based on how much surrounding information they can capture--screen, tip, rebound, etc.).
The value of the expected goal for the unimpeded breakaway where the player shoots from 5 feet out from the crease is calculated by looking at every instance that the NHL gameplay logs have recorded such a shot and calculating the percentage of times that shot has historically gone in.
I'm making this up, but let's say that shot has scored 3% of the time over all recordable events dating back to the time the NHL first started publicly tracking such gameplay events. It would get an expected-goal value of .3. So in that sense quality of player is factored in such that McDavid or other great players scoring a bunch historically on that shot is captured by the number as is Wagner or other bad players shooting it into the goalie's chest every time.
But, to your point, when McDavid specifically gets such a breakaway against the Kings next time the teams play, the .3 expected goals assigned to that shot from McDavid is very likely underselling the quality of that shot.
I find the value in them in that they're telling me the quality of the shots the Kings are creating versus allowing, not necessarily whether because it was Byfield shooting instead of Kempe the expected goal number should be higher. Maybe not everyone agrees with this, but I think that many people would agree that the team that's generating the higher quality shots on a game-to-game basis is more likely to win fairly regularly over the long run than not. So, that's all a long-winded way of trying to answer your question and explain why I find those numbers to be a useful reference point.
(As an aside, and as Jesse Cohen hinted at above, the private models do a much better job of assigning a "truer" value to the shot quality, and they also capture so many more events than the NHL's public logs do. That being said, the public models have done decent historically as far as the correlation between the best teams in the league being the ones who are on the positive side of the "expected-goal battle" over the course of the season.)
Hopefully that was helpful and not too rambling.
Last edited:
If microstats like these were fully tracked and publicly available, they would be brought up more. But the NHL doesnt provide for that. All Three Zones’ data is behind a paywall, and even then the amount of games tracked is a small sample bc it’s literally one guy doing all the work.
Baseball has a much more robust analytics conversation bc there is a greater depth of publicly avail info. We even get spin rates in baseball. In fact we have every minute detail on every single ball thrown and hit.
We cant even get shot speed data anywhere, even though it is clearly being tracked. This league has many shortcomings and this is just one of them.
Last edited:
No Name The Nameless
Registered User
Mattias
The friendly cat.
- Feb 15, 2009
- 1,908
- 898
I saw one one the main boards that was
2nd Colorado 2023
2nd Montreal 2024
5th 2025.
Seemed fair to me?
King'sPawn
Enjoy the chaos
- Jul 1, 2003
- 22,917
- 23,479
I think a team who wants a cost-controlled offensive defenseman may be willing to trade a late first (like pick 28-32).
Seattle King
Registered User
- Aug 19, 2022
- 878
- 1,994
We will find out this offseason.
If Blake wants to re-sign the new guys its going to be time to face the music and deal with the surplus he has built up. We have to look at moving Arvy, Iafallo, Walker and Durzi. There are suitable replacements on controlled contracts in the pipeline. Gavrikov will be 4-5M for 6 years and Korpisalo might be in that neighborhood if he leads us deep into the playoffs.
I like the guy but there are obvious defensive lapses that really expose us to high danger from the enemy.
With our current, new top 4 structure the team just looks much more composed and under control in the small sample we have seen.
Durzi is solid on the 3rd pair and great as a PP guy but we have Spence and Clarke who both need to be sharing that spot next year.
I would think a 2nd round pick would be his value.
- Status
Ad
Latest posts
-
Salary Cap: Pens '24-'25 Salary Thread: The Crosbicles Volume XIX- Latest: KrisLetAngry
-
Rumor: 2024-2025 Trade Rumors and Free Agency Talk | The Slow Crawl to the Season- Latest: Pierce Hawthorne
-
-
