Roster/Ratings Suggestion

[itsmb8]

One thing ive always disliked is how players are 1) rated based on position, and 2) allocated minutes based on "primary position."

The suggestions I have is that players should not have primary or secondary positions, but just positions they are able to play. For example, guys like Giannis and LeBron can play 1-5, KD and Kawhi can play 1-4, big men can usually play a 4 or 5. Instead of saying Giannis' primary position is PF and secondary is SF, say Giannis can play PG, SG, SF, PF, or even C. That would make rotations a lot more dynamic which they should when you have players that can play all 5 positions.

The second suggestion is that players should be rated as a basketball player, not as a certain position player, which would be needed if the first suggestion were to happen. Regardless, some players are underrated in the game because they have a play-style that isnt typical for that position. For example, lets take Giannis again. Lets say the Bucks were deep with big men but thin at guard. Giannis would play a guard position so Milwaukee can have their best players on the floor and get everyone they need to PT, but Giannis would be underrated at guard (probably around an 89 or 90) even though he's still the same 96 ovr player he would be at PF. His skills didnt change, only his position did. And theres no way execs would value Giannis any lower if he was a guard instead of a forward.

[Mikelopedia]

Re: Roster/Ratings Suggestion

I'm interested to see if 2k's overall rating formula has been redone or improved at all for 2k20 seeing as they have removed some ratings. I can't determine that for sure just by looking at overalls because we all know 2k manipulates them to be whatever they want.

[antdoggydogg]

Re: Roster/Ratings Suggestion

I noticed this issue during the player ratings reveal. Particularly with Giannis having a significantly higher rating than Anthony Davis. Both players are the same size, play roughly the same position. AD is a 4/5 and Giannis plays as a 3/4 and a 5 in small-ball lineups. Both are good defensively, you can even say great. The real difference between the 2 is on offense where AD is a much more skilled player, better shooter from everywhere outside the paint. I would think that side of the ball alone would give AD an edge with his overall rating. But those shooting skills don't seem to be valued highly enough at his Power Forward position. Shooting skills should be valued at both Big positions.

[Mikelopedia]

Re: Roster/Ratings Suggestion

I noticed this issue during the player ratings reveal. Particularly with Giannis having a significantly higher rating than Anthony Davis. Both players are the same size, play roughly the same position. AD is a 4/5 and Giannis plays as a 3/4 and a 5 in small-ball lineups. Both are good defensively, you can even say great. The real difference between the 2 is on offense where AD is a much more skilled player, better shooter from everywhere outside the paint. I would think that side of the ball alone would give AD an edge with his overall rating. But those shooting skills don't seem to be valued highly enough at his Power Forward position. Shooting skills should be valued at both Big positions.

PF's have been exhibit A on why the overall formula is arcaic and flat out trash. This is why LeBron and Giannis arent listed PF's in 2k because when you change their position, the overall rating takes a nose dive of 4 to 5 points.

[ronyell]

Re: Roster/Ratings Suggestion

The system should be:

ball-handler
wing
big

[ztennen11]

Re: Roster/Ratings Suggestion

They've already shown that the rating system was revamped and simplified. I forgot where I saw it but it looked improved for sure.

[J_Posse]

Re: Roster/Ratings Suggestion

They've already shown that the rating system was revamped and simplified. I forgot where I saw it but it looked improved for sure.

Until we have it in our hands and see (with our own eyes) that they were consistent, "fair" and used scaling (for non-sim stat attributes) then we can't definitively say it is "improved."

If anything they've complicated accurately rating/defining inside scoring by removing the 'Standing Layup' rating. How do you account for points in the paint - that don't end in a dunk, driving layup, hook shot, fadeaway or tip - in - without a rating to dictate efficiency?

Sent from my SM-G935V using Operation Sports mobile app

[Real2KInsider]

Re: Roster/Ratings Suggestion

Until we have it in our hands and see (with our own eyes) that they were consistent, "fair" and used scaling (for non-sim stat attributes) then we can't definitively say it is "improved."

If anything they've complicated accurately rating/defining inside scoring by removing the 'Standing Layup' rating. How do you account for points in the paint - that don't end in a dunk, driving layup, hook shot, fadeaway or tip - in - without a rating to dictate efficiency?

Not really. Uncontested those shots don't require a rating because they're automatic. Contested, they added all those badges for a reason.

They are also probably using the Close shot or Driving Layup rating, as an educated guess.

[itsmb8]

Re: Roster/Ratings Suggestion

The system should be:

ball-handler
wing
big

I agree to an extent. Most players fit that category, but then theres players like Ben Simmons that can be a ball-handler or wing. Giannis and LeBron can be a wing OR a big.

Thats why I suggested basically a system where each position is listed and you can set whether they can play that position or not. IMO it becomes much more detailed and specific that way.

[antdoggydogg]

Re: Roster/Ratings Suggestion

I actually would like to keep the positional flexibility that currently exists with lineups, but would like to see third, fourth and fifth positions available to each player. I would hate to be locked out of certain positions due to a guard, wing, big system in MyGM. What really needs to happen is creating a smarter rotation/lineup engine. The lineup engine could figure out what the optimal lineup is based off of primary position, secondary position, overall lineup shooting strength, lineup perimeter defense, lineup interior defense, and ball handling.

Let's take the 2020 Lakers for example. What if the computer was able to look at a lineup of...

PG - LeBron (Primary SF / Secondary PF / Tertiary PG)
SG - Danny Green (SG/SF)
SF - Kuzma (PF/SF)
PF - Anthony Davis (PF/C)
C - Dwight Howard (C)

...And determine that this lineup isn't ideal because there isn't enough shooting on the floor. Going off their shooting percentages it would see that LeBron is a below average shooter, Danny Green is above average, Kuzma, AD and Dwight are all below average too (by varying degrees). So it decides to slide Davis up one position because that is an available secondary position, keep Kuzma's below average floor spacing on the court because its better than Dwight's non existant spacing, and introduce a third shooter to the lineup, one of either Caruso, Troy Daniels, Avery Bradley or Quinn Cook...

PG - Quinn Cook
SG - Danny Green
SF - LeBron
PF - Kuzma
C - Anthony Davis

It would also need to take Coach tendencies and preferences into consideration! A lot of factors in play here. Maybe it's an old school coach who doesn't value 3 pointers and likes the old school big lineup with the smallest player being 6'6. While the previous lineup might have worked fine in the 80's, this new lineup fits better in today's league where you need to be making threes to keep up with the other team.

Something I've been thinking about a lot is advanced rotations, and how to get the CPU to generate better, smarter advanced rotations. If you've ever gone into an advanced rotations screen you have probably seen some of the silliness that goes on as the computer tries to figure out how to slot everyones minutes in. I think if the computer was given rules like...if a bench player is assigned 30 minutes, distribute those minutes using one of these options:

Option 1: Come off the bench mid 1st quarter, sub out mid 2nd quarter.
Option 2: Come off the bench mid 1st quarter, sub out early 2nd quarter, sub back in late 2nd quarter to close the half.

It would choose between those two options based on which other players are assigned minutes and what is more feasible. There is no easy way of doing this, we're talking about creating an artificial brain haha. But I think it would be most similar to what the NBA schedule makers go through. "Here are the assigned games, here are the number of days available in the season, optimize this". They get better at reducing back to backs every year as their system becomes smarter.

Introducing certain rules might help, like if you are playing 24 minutes coming off the bench, your entrance will come at the 6 minute mark and exit at the 6 minute mark of the next quarter. If it's 32 minutes, you can have the player come in at the 5 minute mark and exit at the 5 minute mark of the next quarter. Or have them close the second quarter.

[J_Posse]

Re: Roster/Ratings Suggestion

Not really. Uncontested those shots don't require a rating because they're automatic. Contested, they added all those badges for a reason.

They are also probably using the Close shot or Driving Layup rating, as an educated guess.

Neither of which covers close to the basket, stationary shots (the restricted area) which are definitely not "automatic" for shorter players. And it seems, from glancing at the video with ratings and player grades, they've compensated by inflating everyone across the board (in shooting).

Jakob Poeltl

2K19

Inside: B-
Mid: C-
3 - Pt: F

2K20

Inside: A-
Mid: B+ (What?)
3 - Pt: D-

DeJounte Murray

2K19

Inside: B
Mid: C-
3 - PT: C

2K20 (after missing entire season due to injury)

Inside: A-
Mid: B-
3 - Pt: B+ (Shot 26.5 % from three - point range in '18 and is a career 31.6% shooter in 119 gms.)



Sent from my SM-G935V using Operation Sports mobile app

[antdoggydogg]

Re: Roster/Ratings Suggestion

The letter grades might be more inaccurate because there is less distinction between the different shot types and open/contested. But the remaining attribute ratings from last year may have stayed the same. Their letter grade system has always been a little off. But it's also a matter of opinion, 36% is league average for three point shooting, does that mean a 36% shooter should be graded as a C in three point shooting? I don't know the answer to that

[Real2KInsider]

Re: Roster/Ratings Suggestion

And it seems, from glancing at the video with ratings and player grades, they've compensated by inflating everyone across the board (in shooting).

Jakob Poeltl
2K19
Inside: B-
Mid: C-
3 - Pt: F

2K20
Inside: A-
Mid: B+ (What?)
3 - Pt: D-

DeJounte Murray
2K19
Inside: B
Mid: C-
3 - PT: C

2K20 (after missing entire season due to injury)
Inside: A-
Mid: B-
3 - Pt: B+ (Shot 26.5 % from three - point range in '18 and is a career 31.6% shooter in 119 gms.)

FWIW I have in 2K19
Murray
Inside B-
Mid C
3pt B-

Poeltl
Inside B-
Mid C+
3pt F

Close Shot (48-99 for Poeltl) affects none of these letter grades in 2K19, but likely does in 2K20.

As stated, the removal of Contested/Off Dribble ratings likely boosts everyone across the board as I doubt they heavily modified the grade formula beyond that to compensate. It doesn't matter one bit if Murray has B+ 3pt if the entire league has A- or A. Grades aren't attributes. Badges (often more important than ratings) don't affect grades.