MyLeague Regression - (Because the issue should not be forgotten)

[Brian7d]

Has anyone tested for results with PROGRESSION tuned DOWN? My only idea behind this is that if Progression is set lower, guys won't reach their potentials until later (late 20s/early 30s). If regression is coded to players age it may not matter, but perhaps slowing the rate of progression would delay the regression? Just spitballing here...

[BluFu]

Quote: Originally Posted by geisterhome

In that case we have a legit workaround? I mean I haven't tested how the regression works with all years at 1 but all I read on here sounded fine.

this is probably the best workaround.

Quote: Originally Posted by Brian7d

Has anyone tested for results with PROGRESSION tuned DOWN? My only idea behind this is that if Progression is set lower, guys won't reach their potentials until later (late 20s/early 30s). If regression is coded to players age it may not matter, but perhaps slowing the rate of progression would delay the regression? Just spitballing here...

That's the thing, it is. The game also uses years pro as a way of detecting whether players should regress (because no one really regresses after their rookie season.. except maybe Mayo). Scaling years pro back to 1 or a player's age back to 27 are the ways to work around the issue.

[bhxzgh]

I agree as well and my decision on whether or not to purchase the game may hinge on whether this issue will be addressed.

[DocHolliday]

I think it is safe to say that this is not going to be fixed this year.

I only hope this is not one of the bugs that, for some reason, lingers in the system for years to come.

[zfgzdg]

I agree as well and my decision on whether or not to purchase the game may hinge on whether this issue will be addressed.

[vardano]

Well, at least if they would come here and say in any manner.. that they do know about the issue and it will be addressed, I'd be somewhat satisfied. We all know why none of 2k team leplying here, but we better than that. Let's do this game better together.

[Guccilafluer]

Quote: Originally Posted by vardano

Well, at least if they would come here and say in any manner.. that they do know about the issue and it will be addressed, I'd be somewhat satisfied. We all know why none of 2k team leplying here, but we better than that. Let's do this game better together.

Not trying to be pessimistic here but what sports video game company has ever came out and admitted a mistake and then attempted a fix, in the same year? I feel like the answer is none, and will always be none.


Sent from my iPhone using Tapatalk

[BigT34]

I think it might be helpful if we started by thinking about what happens in the real world. I've posted before the percentage breakdown of NBA players. Basketball-reference.com has an easily sortable table that includes players age and is downloadable in excel format, so I compared this year and the 2011 season.

First, age breakdown for this year:
483 players are listed--essentially anyone who actually played a game this year. Their ages are as follows:

38: 4
37: 5
36: 4
35: 6
34: 17
33: 13
32: 19
31: 16
30: 23

In total, 22% of the league is age 30 or older, 14% is OVER 30, and 3.9% is 35 or older.

So, it seems like the issue is that people are complaining about players regressing too much, too soon. In general, the data suggest that NBA players peak around age 26, and indeed, this season, 26 year olds are the single largest group in the NBA. So I decided to compare 30 year olds to their former selves at age 26, and went back to the 2011 season to see who was in the league then and what became of them. I also looked at 28 year olds and 30 year olds. There was an unusually large cohort of 30 year olds in the NBA then, a total of 37 players.

Comparing 2011 and 2015:
-40 players aged 26
-22 still in the league now (one addition is Chris Copeland, who joined the league in 2012)
-If average PER is 15, NONE of the players out of the league now were even league average at age 26
-Most notable players out of the league by 30: Brandon Roy, Sasha Vujacic, Travis Outlaw

-31 players aged 28
-15 still in the league
-Notable players out of the league: Will Bynum (PER 15.1), Hakim Warrick (PER 16.5, Luther Head, Kwame Brown

-37 players aged 30
-18 still in the NBA at age 34
-notably no longer still in the NBA: Yao Ming, Keith Bogans, Dahntay Jones, Al Harrington, Troy Murphy

From 31 to 35, the drop is even more pronounce:
-19 players 31 years old in, only 6 still in the NBA at age 35 (Jannero Pargo was out of the league for that one season, so it's 5/19 + Pargo
-Notably out of NBA now: Lamar Odom, Tracy McGrady, Corey Maggette, Metta World Peace, Baron Davis, Rashard Lewis, Mehmet Okur. 5 of those guys were all-stars, Odom had a great career, and Artest was DPOY on top of that
-Worth nothing that many of those guys had productive seasons, including Odom (PER of 19), Davis (PER of 17), and Maggette (PER of 15.8).
-Rasual Butler is the only player left who has a PER of 15 or better
-Elton Brand is still around, but he has fallen from a PER of 18.6 to a PER of 11.6

Granted, I'm using PER as a catch-all metric, but you can use whatever measure you want.

This isn't a large enough sample size to build a real model, but the data is pretty clearly there. Injuries obviously play a huge role--but that's true in the game, too--I've seen guys regress horribly after injuries in 2K15. But this is why it's risky to give big contracts to veterans--that max deal you give to a 28 year old all-star might buy you a couple of good seasons, and then you'll find yourself spending tons of cap space on a mediocre role player.

I think the regression system in the game needs tweaking, especially because it'd be cool to see guys evolve better over time, a la Duncan/Dirk/Carter. But let's remember how truly rare those are.

This should give people doing testing at least some useful info to work with, I hope.