Player Discussion: Patrik Laine IVever: a new hope? (Laine out of PAP, trade request still stands)
- Thread starter Viqsi
- Start date
- Status
I'm sorry, I don't mean to put you down.
When doing analysis on something you need as much and as complete data as possible. If you want your analysis to be as accurate as possible with the highest prediction power, you need to consider all of it. Attempting to analyze with missing/faulty data only provides bad analysis. People call it "garbage in- garbage out". If you feed bad data, you get bad results.
What you're asking of me is to ignore ALL of the good. Like totally just forget that when he is healthy he's an elite player. How would that exclusion of data provide good analysis and future prediction power? We want prediction power in this situation, because we're trying to evaluate future contributions or lack of. So we best consider everything, not just the lows.
I know enough about Laine, have seen vast majority of his NHL games, that I can't just take some of the lows and forget the highs. If your intention is to convince someone of something by cherry picking data then you should try it with someone who doesn't have a full picture of the career of said player.
I understand that is actually the reason why we clash on this. I've seen all of it, you've seen the CBJ part of it and especially the first season was a bad look for first impressions. I think you got really soured on him in the first season and then just never recovered from that once he started showing why he's worth that much. And because the seasons here have been interrupted so many times with injuries, he has not gotten that great stretch in yet where he's truly one of the best in the league, night in night out. That happens with him, even for months. It's not that he doesn't have that player in him, it's that he's been out a hell of a lot.
I don't ignore his injuries, but it is hockey and it happens. Statistically he's had more than his fair share of injuries, there's no real reason to think that they will continue to happen at the same rate. It's possible that they will, but there isn't any underlying reason like osteoporosis why he has snapped a few bones or been concussed. Maybe it's something that could be addressed in play style, he's gotten injured against the boards more but really they are just variance. Variance can go the other way, it's possible he plays 5 full seasons in a row from now on. Not likely, but possible.
Fully healthy, PP1 on left, top 6 minutes and he can be a 100 point player in a full season. He doesn't get paid like one, because he hasn't had too many full seasons in those conditions.
We only have 1 other player like that who has been able to play at that level (Gaudreau) and currently he's far from it while being healthy. So no, we don't have 3 lines of better options. We don't have any better options. Maybe you'd still prefer them, but "better" would need some quantifying. As far as production goes, they are not on Laine's level.
That's the player I have watched, so when you're telling me Texier needs to play above him in lineup, I have some questions about your methods of evaluation. When one is willing to throw out all the good, and analyze just with the bad data, it's not an honest evaluation. It's an attempt to justify a position that's been formed from feelings rather than from data. I see that stuff (lazy/faulty takes) on the main boards about Laine sometimes, just expect better from home base I guess.
Cowumbus
Registered User
Are you from Finland
Canada
It's not really important in statistical analysis where the analysts are
Cowumbus
Registered User
Remember this data is not valid regarding Laine:
2020/21
3 year weighted average 2019-22
Even when he is “good” on the stat card, his 5v5 impact is bad. 2021-24, Weird:
2020/21
3 year weighted average 2019-22
Even when he is “good” on the stat card, his 5v5 impact is bad. 2021-24, Weird:
Last edited:
So your reply is trying to make it a nationality thing first, and then just posting pictures with no context?
Did you consider anything I mentioned?
Cowumbus
Registered User
If a player has only played 4 seasons, 4 seasons is the correct amount of data.
If they have played 8, then 8.
Cowumbus
Registered User
Double-Shift Lasse
Just post better
I’m going to regret asking this because I don’t believe you to be an honest actor, but here goes…
If a player’s data over a period o years showed improvement, would you believe that the player was improving?
Yes.
I don't get why you'd question me being an honest actor. I'm not the one advocating to use cherry picked data.
Yes. Your charts show that Laine's ceiling is higher than the compared. What did you think it shows? Where's the disconnect?
Do you know, what the biggest fun is? If I don't count the fact that in that year, when he had 63%, he had a point per game, while he scored many important goals and was by far the best in the team in G/P (see other statistics), it is clearly written there in the second table he will miss six weeks with a fractured clavicle. And yet it certainly had one of the best graphs and the best PPG and GPG.
@Cowumbus made a big mistake.... . Everyone on the forum knows that he was injured every season plus there are many injuries in the team which also affects the statistics, but despite that he did very well (stats which Columbus put here).
But...we still need, for example, how he influenced on the players, who he played against, how their partners had stats etc., etc. And do you see them here? Do you see the overall analysis? I do not.
So with all these stats we can write about how Laine (and every player) is important for the team. Any statistician tells you that. I don't know about you, but when I was in high school, that's what they told us in math when we learned statistics. This is also learn in cybernetics classes, successful managers follow it. Not only certain statistics, but also as a subject influences on others as subjects on him, what is the influence of the environment minimum.
I don't know, maybe the overall stats can show that Patrik is not that important as other players or vice versa. But once again, if we want to find out the overall impact of the player, his quality, we cannot do without what I wrote above. That is why these stats very imprecise and can only serve as a guide, and even then not always. And that applies to every player.
Last edited:
I don't know how the weighing on those 3-year avg jfresh cards work, but like you can see on the card with 20-21 data only he was something like 3% and 1% for EV. For the latest with 3-year weighted that has risen to 39% and 13%, maybe that 20-21 is still dragging those numbers down quite a bit. Otherwise I find it hard to imagine how you'd look at the 22-23 season data and come up with a poor EV rating for either offense or defense, unless the model just brutally underrates anyone in a bad team. He was leading the team in a lot of stats, real and imagined, including per 60 5v5 production by some margin and most of that was supported by underlying metrics as well.
Comparing production rates across seasons is a bit moot because the team quality varies, like this year we have something like 5 guys producing at higher pace per 60 than Laine did last season, but I wouldn't say all of those guys are playing better hockey offensively (or defensively for that matter) than Laine did 22-23. For one they are not playing in front of an AHL defense. I dunno, maybe for 22-23 I'm overrating a guy playing OK in a really bad team, but I believe it's the other way around and Laine's performance last season gets overshadowed by the fact that the team was absolute crap. He was playing 50/50 hockey in a team that played 35/65 hockey without him or something while on average playing against top-6 quality competition with middle-6 quality team mates, if I'm reading that right. In any case it was the kind of season I would've loved to see him build on and to see if it was a blip or his new base level.
It's bit ridiculous it's been like 8 months or something since we first saw him at C and we still don't really know if it's something that he could do going forward. He didn't have exactly a great start to the reg season as center but IMO he wasn't bad either and looked like it could've gone well with little more time to adjust.
I don't even understand why we are talking about him so much, he's out for a while still and we've barely seen him play any hockey at full health this season.
Comparing production rates across seasons is a bit moot because the team quality varies, like this year we have something like 5 guys producing at higher pace per 60 than Laine did last season, but I wouldn't say all of those guys are playing better hockey offensively (or defensively for that matter) than Laine did 22-23. For one they are not playing in front of an AHL defense. I dunno, maybe for 22-23 I'm overrating a guy playing OK in a really bad team, but I believe it's the other way around and Laine's performance last season gets overshadowed by the fact that the team was absolute crap. He was playing 50/50 hockey in a team that played 35/65 hockey without him or something while on average playing against top-6 quality competition with middle-6 quality team mates, if I'm reading that right. In any case it was the kind of season I would've loved to see him build on and to see if it was a blip or his new base level.
It's bit ridiculous it's been like 8 months or something since we first saw him at C and we still don't really know if it's something that he could do going forward. He didn't have exactly a great start to the reg season as center but IMO he wasn't bad either and looked like it could've gone well with little more time to adjust.
I don't even understand why we are talking about him so much, he's out for a while still and we've barely seen him play any hockey at full health this season.
Come on man, you wear it on your sleeve. It's not that hard to spot a Finn.
I live in Canada, I was born in Finland. None of that matters in statistical analysis. The numbers are the same for Zimbabweans and Yemeni. The exact same conclusions can be drawn by a Norwegian when looking at the data available.
It's an irrelevant question. Instead of responding to anything I said, he asked that. It's a tactic used by people in a losing position to deflect from the actual facts of the matter to an assumed bias of nationality. As you will now see. It's a way to "peace out" by claiming bias.
Last edited:
Double-Shift Lasse
Just post better
straw men
gaslighting
overt misrepresentation
I have no idea why I would feel you are not an honest actor in this discussion.
I would like us to just expose the fallacies in our fellow poster's positions, rather than attack their character.
CBJWerenski8
Rest in Peace Johnny
- Jun 13, 2009
- 43,511
- 26,482
Cowumbus
Registered User
Clavicle is relatively minor right? We had guys break them all the time in lacrosse and I never really felt as if it made a difference once they returned.
Judging by Laine's returns, Patrik could theoretically play against Edmonton. And even if he doesn't, most likely against Calgary. Either way, it will be interesting to see who will be a healthy-scratch upon his return.
koteka
Registered User
2 of my kids have had broken clavicles and they both healed quickly and it was like nothing ever happened after a few weeks. I know other people who have needed surgery and pins and it takes longer to recover.
i've clowned on jfresh plenty on this board, so i'll go ahead and ring this bell again: the 'analysis' part of analytics requires examination of context and understanding of how statistical principles work. simple data viz never tells the full story.
in this case, the statistical principle you're missing is outlier data and it's present in the form is his 20-21 season (the torts year). that undermines the intent of a three-year sample, because the outlier skews the dataset.
his WAR was in the 39th percentile that season. the year before (19-20) that he was ~70th%, then ~75th% in 21-22 and ~85th% in 22-23. clear which one is the outlier.
now, time to get a little bit 'inside baseball' about the twitter analytics stuff: public-sphere analytics don't do a good job of accounting for team factors. players who play limited minutes on good teams have heavily inflated 5v5 microstats, and players who play heavy minutes on bad teams have suppressed numbers. using publicly-available data to create a catch-all stat like WAR in a free-flowing sport like hockey is a fool's errand specifically because the data is too limited to account for those factors. hence dom saying he 'couldn't envision a scenario where brandon montour is successful' when buffalo was trying to trade him, but now saying he's an elite defenseman.
given that context, and accounting for the outlier data, laine comes across as a useful and effective player who is extremely good at things that are really hard to do (scoring goals, creating high-danger chances), decent at some things (5v5 offense) and bad at others (5v5 defense).
analytics isn't just "player x good, player y bad" -- the utility of analytics is figuring out how to maximize a player's talent/effectiveness into on-ice production. laine's analytical profile gives a clear roadmap of how to do that (build the pp around him, put him with two-way forwards at 5v5 who can get him the puck).
john tortorella didn't follow that roadmap (i get why, because laine was still really young and torts thought he could be more, but it didn't work). pascal vincent isn't following it. brad larsen did, and laine was basically a ppg player under him. not a mystery why!
Cowumbus
Registered User
Depends how you look at it.
Sounds like you are agreeing with what I have been saying this entire time. Although, I don’t think decent is the right word for his EV offense. Laine’s 5v5 EV strength offense and defensive percentile according to JFresh by year:
2017- 9th, 5th
2018- 6th, 7th
2019- 7th, 7th
2020- 23rd, 3rd
2021- 6th , 1st
2022- 12th, 3rd
2023- 38th, 31st
Last edited:
- Status
