Giroux vs Kuznetsov
- Thread starter Llamamoto
- Start date
KirkAlbuquerque
#WeNeverGetAGoodCoach
Giroux has a .03 ppg advantage over Kuznetsov over the last 5 years, but I don't know how anyone can watch both players and not see that Kuznetsov is more impactful. And this isn't even getting into playoffs, where Giroux has been a complete non-factor.
Points are points no matter how you look at it. Giroux is an incredible defender as well (he’s just never acknowledged for it). You can look up their metrics via NaturalStatTrick and Giroux pretty much clears Kuznetsov on almost everything. Really the only thing Kuznetsov (this year and considering this is a down year for Giroux) was more effective in was getting more offensive zone draws and PDO (luck). Really, everything else favors Giroux.
With raw stats, Kuznetsov has 1 more assist in less games played. Idk if that’s due to injury or not.
Both are incredible players, however, if I’m building a team 100 times out of 100 (if both players are available) I take Giroux
Flashy isn't the same as impactful. Kuznetsov was great in '16 and in the '18 playoffs but has been largely disappointing for his talent level outside of that. He also might be the worst defensive top 6 center in the league while Giroux is one of the better play drivers.
1) No, the system that one plays in has a massive effect on point totals. It's how Jamie Benn won an Art Ross and how Giroux put up 100 points a few years ago. The last two years under Hakstol the Flyers changed their transition game where one forward would almost cherry-pick in the NZ along the wall, the idea being to push back the D and open up space for rush chances and odd-man situations. Giroux didn't suddenly find a second wind in his career and was inspired to score 40 more points, he's a smart player and the system change gave a player like him more chances when he was allowed more room to operate. Of course, this type of system is rarely successful in the playoffs when the opposition gameplans for it, so Claude was back to being an invisible perimeter player against the Penguins.
2) Incredible defenders do not get moved to wing.
3) PDO is not luck. The true PDO for all players is not 100%.
4) I wouldn't disagree with picking Giroux for building a team. I would with picking 32 year old, past his prime Giroux.
Nathaniel Skywalker
Registered User
- Oct 18, 2013
- 13,834
- 5,404
Giroux 102 points 2nd in scoring in 17-18 snubbed for the hart. And it’s not like he isn’t good in playoffs. Giroux easy
Voight
#winning
1) Yes, points are points no matter if they’re even strength points or from the PowerPlay. I never mentioned the system archetype (I have no idea why you brought it up unless you’re trying to change the argument into an abstract to make a point. Idk), but yes you just outlined the Hakstonian perimeter play. I don’t understand why you brought it up.
2) Incredible may be a bit overzealous, however, they can be moved to wing when Sean Couturier is the center.
plus, Giroux was facing the hardest usage for any of our F when he was our 1C. Moving him to wing is a longevity move. Give him some easier usage and hopefully he can last a little longer.
3) The statistical inferencing of PDO basically summates to luck and chance. No matter the fact that the average stat doesn’t fall on the meridian (100% in this case), doesn’t mean that 100% is a true representation. Think of a bait stat like cordial trends (-1 - 0 - 1). Neutrality in this stance is 0- similar to neutrality being 100% (you’re not lucky, but you’re not unlucky). If the trend averages the population to be .23, then true neutrality doesn’t shift from 0 to .23. Just like your #1 I truly don’t understand what you’re going for here.
Nathaniel Skywalker
Registered User
- Oct 18, 2013
- 13,834
- 5,404
Pavel Buchnevich
Drury and Laviolette Must Go
Kuznetsov more talented. Giroux more consistent. But even Giroux lacked consistency last season, so I find it hard to not pick Kuznetsov.
Voight
#winning
TomasHertlsRooster
Don’t say eye test when you mean points
Kuznetsov has posted a 5-on-5 PDO above 1.015 in every season of his career and a 5-on-5 on-ice shooting percentage above 10% in every season of his career except for one. He might be lucky to lead the league in 5V5 oiSH% over the past two years, but his “true PDO” is absolutely not 100.
1) All you said was points are points nothing about ES vs PP. I had originally said Giroux had a marginally higher points per game average.
3) I'm well aware of what PDO is. What you fail to recognize is that what is true (by definition) for an entire population is not necessarily true for each individual in that population. Some players playtime tends toward higher shooting % than others, Kuznetsov may be such a player.
, yes, but you are taking away variance...I could care less what Kuznetsovs true PDO is. All I stated was just from looking at the metrics, the 2 that Kuznetsov best Giroux in were PDO and offensive zone draw %.
Statistics on a spectrum don’t necessarily work that way. If they did then I would be out of a job because I’d be losing my company money.
I’m not necessarily disagreeing with your sentiment, rather your analysis.
Again, when analyzing hockey metrics to such a small population drives down variance and creates a larger error. What you and the guy above are saying is correct (Kuznetsov may not be a margin PDO guy, but your analysis in using population metrics to say this is in fact statistically incorrect- you can’t necessarily say a true PDO based off of a population sample of 1 and then compare it to the sample size of the NHL.).
What about on-ice GF rate? Or are you just grouping up everything not in Giroux's favor as luck and dismissing it?
I could care less what Kuznetsovs true PDO is. All I stated was just from looking at the metrics, the 2 that Kuznetsov best Giroux in were PDO and offensive zone draw %.
Statistics on a spectrum don’t necessarily work that way. If they did then I would be out of a job because I’d be losing my company money.
I’m not necessarily disagreeing with your sentiment, rather your analysis.
Again, when analyzing hockey metrics to such a small population drives down variance and creates a larger error. What you and the guy above are saying is correct (Kuznetsov may not be a margin PDO guy, but your analysis in using population metrics to say this is in fact statistically incorrect- you can’t necessarily say a true PDO based off of a population sample of 1 and then compare it to the sample size of the NHL.).
As for the bolded, that is true you can't definitively state that, but the opposite is true as well- you can't just assume 92's true PDO is closer to 100.
1865
Alpha Couturier
Points are a wash on a down year for Giroux, and Claude takes everything else.
TomasHertlsRooster
Don’t say eye test when you mean points
I could care less what Kuznetsovs true PDO is. All I stated was just from looking at the metrics, the 2 that Kuznetsov best Giroux in were PDO and offensive zone draw %.
Statistics on a spectrum don’t necessarily work that way. If they did then I would be out of a job because I’d be losing my company money.
I’m not necessarily disagreeing with your sentiment, rather your analysis.
Again, when analyzing hockey metrics to such a small population drives down variance and creates a larger error. What you and the guy above are saying is correct (Kuznetsov may not be a margin PDO guy, but your analysis in using population metrics to say this is in fact statistically incorrect- you can’t necessarily say a true PDO based off of a population sample of 1 and then compare it to the sample size of the NHL.).
I don't get what you're trying to say here.
You said that the statistical inferencing of PDO basically summates to luck and chance. I said that isn't true at all. Now it seems that you're agreeing with me?
Bounces R Way
Registered User
Yes, the inferencing behind PDO is basically luck. It’s literally a calculation of shooting percentage and save percentage. It’s naturally assumed that the meridian for stats is 100 because that’s where the numbers naturally regress (Same thing for integers coupled with the number 0- above is positive and below is negative. Your argument saying that 100 is not the meridian doesn’t make much sense because it doesn’t matter what the true neutrality of the league is. Everyone could be above and that wouldn’t change the fact that 100 is still the number that separates luck vs not being lucky).
Since PDO is made up of save percentage (something the player doesn’t have full control over) and shooting percentage (sometimes people get lucky goals not derived off skill which can have a big impact on on this metric) it is inferencing probability... basically.
For Giroux, if he gets a lucky goal that doesn’t take much skill... shoots it off another player that has a higher probability to miss (strategically speaking - per basic risk analysis - not that smart) then his S% and PDO increased by .41% just off of 1 shot. Any metric that can change by that much by 1 shit when the sample size is upwards of 500 samples (S% and Save%) is not really inferencing something intrinsic. Since it is a stat that naturally regresses and comes to its neutrality the more that data is given (it’s a population statistic which is why I was going against comparing it between players and actually meaning it) if there is a high degree of change, the error for inferencing becomes high. That’s why a bunch of analysts look at it as luck and nothing much more.
idk, it was nice to hear both of your thoughts though. I feel like this conversation is going in loops and taking up a lot of space.
But it's not luck. So Sidney Crosby has just been extraordinarily lucky over his 15 year career vs. one of the best players ever?
And what do you mean that's where they naturally regress? You have provide evidence for that. They are obviously tied to 100 at a league level since every shot for one side is a goal against for another. You are arguing that shot quality when a player is on the ice is equivalent for everyone, which just isn't logical. Now it may not always be easy to determine how a player affects shot quality, but that doesn't mean it doesn't exist.
