Prob should make assumption (cause you know what they say) about how others evaluate players. Those types of charts are just good summations. Bows wouldn't use them if he didn't think the info in them matched what he saw on the ice.
- L_B_R

Exactly what she said.

Then go with what you’ve seen, not throw charts out there. Throwing charts out there’s makes me believe otherwise
- rwilliams88

You mean using empirical date to back up my opinion? I am wacky like that.

Maybe he just means he wants a chance to select an additional player.
- boilermaker100

That's prob the more likely scenario. More selections means higher probability of someone turning out to be an NHL player.

I will say that it's disappointing that Bowman could neither get the 25th (where he could have picked Ceulemans or Lambos* if he was adamant on a d-man) or not given up the higher 2nd (where Pastujov, Raty, Stankoven, Pinelli, Behrens, Chaika, etc might be).

* Ironically Lambos comp is Jones.

Feel a little bad about Nolan Allan - he prob shouldn't read stuff on the internet about him right now.

The only thing we can hope for that MIGHT somewhat explain his continued declining metrics from the previous season(that the data, unless I’m mistaken, didn’t/doesn’t take into account) is that he was effected by the change of routine and everything else that came along with the pandemic. Maybe…MAYBE in a normal situation he might have had different results/performances that wouldn’t have suggested the same downward trend, and helped strengthen his case at being a potential outlier. But I’m not gonna count on this being the case.

I think you're misunderstanding what I'm saying but I legit don't know how to say it in another way so that could be my fault. But yes that's common and no I'm not saying prime is 23-27. I literally said something different.
- L_B_R

Nope...not misunderstanding any of it. You could say it 100 different ways and I’ll still come to the same conclusion. A chart will not predict an elite 26 going on 27 year old defenseman’s production in the next few years. It’s actually quite impossible. You can throw past results at me all you want, but it doesn’t mean a thing. Give me actual reasoning on why this player, Seth Jones, is going to be that type of player that that chart projects and I’ll give you 50 players that this particular chart missed on. It’s nonsense, especially at his age and past production and current skill level

Nope...not misunderstanding any of it. You could say it 100 different ways and I’ll still come to the same conclusion. A chart will not predict an elite 26 going on 27 year old defenseman’s production in the next few years. It’s actually quite impossible. You can throw past results at me all you want, but it doesn’t mean a thing. Give me actual reasoning on why this player, Seth Jones, is going to be that type of player that that chart projects and I’ll give you 50 players that this particular chart missed on. It’s nonsense, especially at his age and past production and current skill level
- rwilliams88

I respect your opinion but in this scenario it makes zero sense

Nope...not misunderstanding any of it. You could say it 100 different ways and I’ll still come to the same conclusion. A chart will not predict an elite 26 going on 27 year old defenseman’s production in the next few years. It’s actually quite impossible. You can throw past results at me all you want, but it doesn’t mean a thing. Give me actual reasoning on why this player, Seth Jones, is going to be that type of player that that chart projects and I’ll give you 50 players that this particular chart missed on. It’s nonsense, especially at his age and past production and current skill level
- rwilliams88

I think you’re underestimating how young many professional athletes are at the time they truly begin to decline. Happens a lot sooner than most might think. It’s not at all unusual for a player to peak early/early 20’s and then slowly(gradually) regress into their mid-late 20’s. It’s far more uncommon for someone to peak from age 27-30, especially after they’ve already experienced a slight decline.

I think you’re underestimating how young many professional athletes are at the time they truly begin to decline. Happens a lot sooner than most might think. It’s not at all unusual for a player to peak early/early 20’s and then slowly(gradually) regress into their mid-late 20’s. It’s far more uncommon for someone to peak from age 27-30, especially after they’ve already experienced a slight decline.
- SimpleJack

New blog! Thoughts on the Blackhawks trade with the Blue Jackets to acquire Seth Jones and the impacts moving forward.

Plus, a look at the Hawks selecting Nolan Allan as their 1st round pick in the NHL Draft.

https://bit.ly/3iK7DZW

I think you’re underestimating how young many professional athletes are at the time they truly begin to decline. Happens a lot sooner than most might think. It’s not at all unusual for a player to peak early/early 20’s and then slowly(gradually) regress into their mid-late 20’s. It’s far more uncommon for someone to peak from age 27-30, especially after they’ve already experienced a slight decline.
- SimpleJack

It’s much more common for them to peak at the 27-30 ages, especially a skilled player like Seth Jones. Prove me otherwise

- L_B_R

Still not giving me facts

It’s much more common for them to peak at the 27-30 ages, especially a skilled player like Seth Jones. Prove me otherwise
- rwilliams88

You keep dismissing proof people provide but there's been a lot of research on this.

Brandner et al. (2014) demonstrate that peak age is between 27 and 28 where skaters can exhibit near-peak performance over a wide range going from about 24 to about 34. Note: Key being that "near-peak" meaning plateau with minor decline.: https://www.researchgate....ational_hockey_league_nhl


There's also the research for more recent trends that shows that peaks are trending younger with new players but plateaus (with minor decline) from 26/27 to 30/31 still exist: https://hockey-graphs.com...s-for-nhl-skaters-part-1/

You keep dismissing proof people provide but there's been a lot of research on this.

Brandner et al. (2014) demonstrate that peak age is between 27 and 28 where skaters can exhibit near-peak performance over a wide range going from about 24 to about 34. Note: Key being that "near-peak" meaning plateau with minor decline.: https://www.researchgate....ational_hockey_league_nhl


There's also the research for more recent trends that shows that peaks are trending younger with new players but plateaus (with minor decline) from 26/27 to 30/31 still exist: https://hockey-graphs.com...s-for-nhl-skaters-part-1/
- L_B_R

Let’s do this...how about I prove that people peak at these ages at a higher rate then they decline...or would you refute that as well?

Let’s do this...how about I prove that people peak at these ages at a higher rate then they decline...or would you refute that as well?
- rwilliams88

I'd ask these questions first:

- What are your qualifications for peak?
- Are those qualifications applied as a standard across all your examples?
- Are those qualification multi-faceted and objective?
- Would you provide the necessary amount to non-outlier status?

For that last point, on players from just 2008-2016, you'd need to provide about 233 examples of players just to hit 5% of a model being wrong. If you provided 50 like you suggested before, that would only be about 1%.

Narrow to defensemen, you'd need to provide about 96 for 5% and 50 would be about 2.6%. Narrow down to just top-4 by TOI, 50 would be close 4.8%.

It'd obvious go up much higher if the data included more decades like some of those models do.

Edited: my % were off slightly so corrected and added d-men only.

I'd ask these questions first:

- What are your qualifications for peak?
- Are those qualifications applied as a standard across all your examples?
- Are those qualification multi-faceted and objective?
- Would you provide the necessary amount to non-outlier status?

For that last point, on players from just 2008-2016, you'd need to provide 163 examples of players just to hit 5% of a model being wrong. If you provided 50 like you suggested before, that would only be about 1.5%.

It'd obvious go up much higher if the data included more decades like some of those models do.
- L_B_R

I would love to see this breakdown...sorry, but what you are saying is absolute nonsense.