Neon1 2k10 (360)

[bussiejr]

Re: Neon1 2k10 (360)

ok so these rosters still need a serious overhaul. ive been simming for stats and ive found that sliders arent the issue. Its a combination of rosters and sim engine. ive checked the scales used by neon, and they are on point relative to eachother, but the stats they produce are off. even in tendencies. im working on figuring out what each rating value produces so that players can be rerated for accurate stats. so far ive been working on shot ratings and tendencies. ive noticed you can create almost exactly accurate fg%'s by simply tweaking tendencies. in this roster there arent enough shots inside, dramatically reducing fg%. the shots taken inside are overall too low a % as well. so people's inside shot ratings are mostly underrated.

this is going to be a serious project im undergoing, i'll probably need about a month to finish. i plan on revisiting all tendencies and ratings including coach profiles to deliver the most accurate experience possible for both simming and gameplay. there may even be a few slider tweaks afterwards to fix any remaining gameplay issues.

neon - this is in no way trashing your work, just recognizing there is a whole lot more needed to create a more realistic product. you have to agree there are still many issues with this roster as good as it is. i just want to take the best roster available and make it even better

[apollooff320]

Re: Neon1 2k10 (360)

WTF...u ppl need to know you can't just use default game play sliders. That's why u guys are having so many issues.

[hugeheadguy]

Re: Neon1 2k10 (360)

what slider tweaks are you guys using

[bussiejr]

Re: Neon1 2k10 (360)

WTF...u ppl need to know you can't just use default game play sliders. That's why u guys are having so many issues.

hey have you seen simmed stats for the league? league averages are way off in many categories, just sim a season and look for yourself. this is not a slider issue.

[Oraeon1224]

Re: Neon1 2k10 (360)

hey have you seen simmed stats for the league? league averages are way off in many categories, just sim a season and look for yourself. this is not a slider issue.

He is correct. I posted a thread with this seasons real season statistics for all NBA teams. The FG% is the most glaring deficit (about 3% less than the NBA average). The rosters take too many shots at close range and not enough at the rim shots (layups). However, the FGA per game are also wrecked by the simulation engine (>110 attempts per game) so I don't really think sliders or rosters will fix many of the issues with the simulated statistics.

If you are going to fix the "inside shooting" be aware that close shot tendencies shoudl actually be decreased it is layups that are more in demand (I have attack the basket at 92 right now with a realistic number of layups and free throws per game).

[DavonBrown]

Re: Neon1 2k10 (360)

Everyone using default sliders with these?

[Cantgetright69]

Re: Neon1 2k10 (360)

He is correct. I posted a thread with this seasons real season statistics for all NBA teams. The FG% is the most glaring deficit (about 3% less than the NBA average). The rosters take too many shots at close range and not enough at the rim shots (layups). However, the FGA per game are also wrecked by the simulation engine (>110 attempts per game) so I don't really think sliders or rosters will fix many of the issues with the simulated statistics.

If you are going to fix the "inside shooting" be aware that close shot tendencies shoudl actually be decreased it is layups that are more in demand (I have attack the basket at 92 right now with a realistic number of layups and free throws per game).

Yeah these are some of the worst ones hes made. Last year he made some amazing ones this year its completely horrible.

[Neon1]

Re: Neon1 2k10 (360)

ok so these rosters still need a serious overhaul. ive been simming for stats and ive found that sliders arent the issue. Its a combination of rosters and sim engine. ive checked the scales used by neon, and they are on point relative to eachother, but the stats they produce are off. even in tendencies. im working on figuring out what each rating value produces so that players can be rerated for accurate stats. so far ive been working on shot ratings and tendencies. ive noticed you can create almost exactly accurate fg%'s by simply tweaking tendencies. in this roster there arent enough shots inside, dramatically reducing fg%. the shots taken inside are overall too low a % as well. so people's inside shot ratings are mostly underrated.

this is going to be a serious project im undergoing, i'll probably need about a month to finish. i plan on revisiting all tendencies and ratings including coach profiles to deliver the most accurate experience possible for both simming and gameplay. there may even be a few slider tweaks afterwards to fix any remaining gameplay issues.

neon - this is in no way trashing your work, just recognizing there is a whole lot more needed to create a more realistic product. you have to agree there are still many issues with this roster as good as it is. i just want to take the best roster available and make it even better

This is what i feel you guys are going to run into.

The stats and percentages are not going to be balanced between cpu simming and actualy sitting there and playing the game (even for watching cpu v cpu).

Like right now for instance, if you simply sim through and never play, but look at the team stats, you will say "well teams shoot too low a %".

Which is true.

Thing is, either play or watch the A.I. duke it out and those same team percentages IN-GAME are generally going to average +5% to +7% higher then what you see on the team stats screen.

They arent on the same floor.

If they were, you would see those teams shooting the low percentages IN-GAME and not just when you sim seasons.


One of them seems it will always be out of kilter with the , edited the other. So when faced with the decision, i edited the game with the people that were actually going to play the game in mind.


Then keep in mind, we also had to restart over from scratch when the roster got currupted, so this current roster is pretty much thrown together as best as the limited time allowed Serra and myself to do. Some edits like rebounding,steals etc had to be left out. We didnt want to continue wasting our entire gaming season pulling our hair out over 3 and 4 percentage points, or LBJ's scoring being a point and a half too low.

If we kept editing and editing, by the time we "finished" messing around with this 2K11 would already be on the shelves.

If any slider guru's want to develop sliders or rules settings to improve gameplay, then hey its all good, im all for it. I am not going to be spending anymore free time doing major edits to this thing, so people just have to do what they have to do, even if its band-aid fixes (aka sliders) or cover ups.

If you need advice or suggestions ill try my best to help.

[bussiejr]

Re: Neon1 2k10 (360)

This is what i feel you guys are going to run into.

The stats and percentages are not going to be balanced between cpu simming and actualy sitting there and playing the game (even for watching cpu v cpu).

Like right now for instance, if you simply sim through and never play, but look at the team stats, you will say "well teams shoot too low a %".

Which is true.

Thing is, either play or watch the A.I. duke it out and those same team percentages IN-GAME are generally going to average +5% to +7% higher then what you see on the team stats screen.

They arent on the same floor.

If they were, you would see those teams shooting the low percentages IN-GAME and not just when you sim seasons.


One of them seems it will always be out of kilter with the , edited the other. So when faced with the decision, i edited the game with the people that were actually going to play the game in mind.


Then keep in mind, we also had to restart over from scratch when the roster got currupted, so this current roster is pretty much thrown together as best as the limited time allowed Serra and myself to do. Some edits like rebounding,steals etc had to be left out. We didnt want to continue wasting our entire gaming season pulling our hair out over 3 and 4 percentage points, or LBJ's scoring being a point and a half too low.

If we kept editing and editing, by the time we "finished" messing around with this 2K11 would already be on the shelves.

If any slider guru's want to develop sliders or rules settings to improve gameplay, then hey its all good, im all for it. I am not going to be spending anymore free time doing major edits to this thing, so people just have to do what they have to do, even if its band-aid fixes (aka sliders) or cover ups.

If you need advice or suggestions ill try my best to help.

well my aim is not to keep editing until i feel it's right......that would take forever lol

i'd like to just figure out how each rating and tendency changes statistical output, and recreating rating and tendency scales from there(and then having someone who is very fast input the new ratings). for instance, testing out 3 pointers i've found that the scale from 25-99 is around .170 - .470 as far as 3pt% goes. Mid range? 35-99, .320 - .490. now you may have already figured these scales, as fg%'s are easy to decipher. so i'm going to try to do this for everything, but of course that will only correct sim issues. after that is done, slider tweaks most likely should be able to fix any other gameplay issues. from all the playing around the last few days i'm confident that this can be achieved, sadly i just don't have much time to do so

a few things i have found so far......

the 3pt scale is pretty linear until around 90. i might need to do further testing in the 85-99 range to be sure
25-.170
30-.190
35-.210
40-.230
45-.250
50-.270
55-.290
60-.310
65-.330
70-.350
75-.370
80-.390
85-.410
90-.430
95-.455
99-.465

i also noticed 3 pt tendency is off. if i set a guy to do 35 inside, 25 close, 25 mid, 15 3pt, he should shoot around 15% of his shots from 3pt land. players tend to shoot more than their tendency indicates which definitely affects fg% overall. one particular test i ran was gerald wallace who was set to 15, and shot 18% instead over 5 full season tests. so i wonder what other ranges are affected? i'm guessing inside shots, but i need more testing to figure this out.

so you see once i (or we for anyone who wants to help) get all these ratings and tendencies figured out, roster editing shouldn't take more than a week. i think all in all this process shouldn't take more than a month

[BodyBump]

Re: Neon1 2k10 (360)

Any update for the new roster?

[Neon1]

Re: Neon1 2k10 (360)

Any update for the new roster?

Dont think anything has happened for an update....

[bussiejr]

Re: Neon1 2k10 (360)

hey neon, been working on the sim fixes for your roster, but it doesn't look like it can be done without lowering tempo all the way to 0 for most teams. and on top of that, the teams that shoot alot of 3's need their tendencies toned down to prevent them from having too many fga's.

how it works, tempo seems to control the amount of shots per game that are not 3's. so if you reduce every player on a team's 3 point tendency to 0 and the team's tempo to 0 they will shoot around 5500 total shots(team total). If you max out the players tendency for 3's, they will end up shooting a rediculous amount of 3's but still shoot around 5500 non-3 point attempts. in real life, teams averaged around 5200 total non-3 pointers for the season. so teams that had less than 5200 inside, close, and mid range shots for the season would need their individual player 3pt tendencies scaled back to make up for the way too many shots taken.

with the tempo low and tendencies fixed, teams have better fg%'s but points are a little low now(obviously). rebound %'s are off, steals and blocks too high, turnovers too low, free throw attempts are too few, and assists are also low. looking to see what i can do to fix these as i think as these problems are fixed small tempo adjustments would be all that's left to have accurate sim stats(with the exception of some teams shooting less threes).

the only thing i worry about with gameplay though is that most every team will have a tempo less than 20 to do this. player tendencies should control the pace of the game though, so hopefully the difference in team play styles is still apparent.

[wrestlinggod1]

Re: Neon1 2k10 (360)

so what slider set is best?

[Neon1]

Re: Neon1 2k10 (360)

hey neon, been working on the sim fixes for your roster, but it doesn't look like it can be done without lowering tempo all the way to 0 for most teams. and on top of that, the teams that shoot alot of 3's need their tendencies toned down to prevent them from having too many fga's.

how it works, tempo seems to control the amount of shots per game that are not 3's. so if you reduce every player on a team's 3 point tendency to 0 and the team's tempo to 0 they will shoot around 5500 total shots(team total). If you max out the players tendency for 3's, they will end up shooting a rediculous amount of 3's but still shoot around 5500 non-3 point attempts. in real life, teams averaged around 5200 total non-3 pointers for the season. so teams that had less than 5200 inside, close, and mid range shots for the season would need their individual player 3pt tendencies scaled back to make up for the way too many shots taken.

with the tempo low and tendencies fixed, teams have better fg%'s but points are a little low now(obviously). rebound %'s are off, steals and blocks too high, turnovers too low, free throw attempts are too few, and assists are also low. looking to see what i can do to fix these as i think as these problems are fixed small tempo adjustments would be all that's left to have accurate sim stats(with the exception of some teams shooting less threes).

the only thing i worry about with gameplay though is that most every team will have a tempo less than 20 to do this. player tendencies should control the pace of the game though, so hopefully the difference in team play styles is still apparent.

Are you actually testing the IN-GAME effects of stuff like this? If so, whats going on with it?


Maybe you should just completely edit 4-6 teams and test it out. thats how i do it. Get a variety like fast paced GSW,PHO..the super slow, POR,CHA,BOS,CLE....and the chuckers ORL,NYK etc.

I guess im just trying to get a grasp on what you guys are actually trying to accomplish...

Is this an effort to 100% replicate team stats for simulation purposes only or is the main focus on actually improving the IN-GAME (cpu vs cpu or cpu vs hum) simulation aspect?

[bussiejr]

Re: Neon1 2k10 (360)

well i could care less about cpu vs cpu really(although it is cool sometimes), i just figured watching cpu games was a fair way to test rosters. what i'd really like is to be able to play associaton with my in-game season stats comparable to league sim season stats. and of course, for those stats as close to realistic as this game will allow.

and no, i haven't really tested anything with gameplay lately. i've focused solely on trying to figure out what attributes/tendencies/coach profile changes all do to affect simulation stat output. as far as dropping tempo, for some teams this may need to be rectified with the pace slider, or even run plays amount. but then again maybe not, tendencies may control pace. i think i'll try what you said, do up a few complete teams and see how it plays.