Dynasty focused [Simulation Engine] patch

[Chris_23]

[Simulation Engine] Patch

Spent another day studying simeng.txt and ended up updating the game and finalizing something to release. This is a bit technical, but an average computer user should not have a problem installing the patch.

This is Dynasty focused and uses default rosters to get more-or-less accurate results around the league. This does not mean that it only works for rosters, but it does mean that iy works best with rosters that have at default the stats distribution similar to that of default rosters (aka if you have already reduced fatigue in rosters, this will decrease it a bit more)

A note of warning though. NBA Live 07 uses -overall- stats to calculate game results and when results are known, the points are shared between players and other statistics generated and then shared based on player ratings. This means that there is no way to just simply make a player score 30 points. This means that the rest of the players in his team need to hold the ball less. Using default rosters, LeBron James does not score many points for a single simple reason - he has two other strong offensive players that get a share of the score due to NBA Live 07 algorithm. Since you cannot edit Primacy in game, you can tweak player ratings to your liking before applying the patch, the patch uses Offensive Awareness to calculate how much the player gets the ball (why? Because player gets the ball -only- in the offense, Primacy does not affect rebounds, steals, blocks etc)

Do NOT expect super realism here. It's strong improvement on original simulation engine and stats in dynasty, but I was merely focusing on dealing with assists and PF/C dominance. To get realism, you need to tweak players offensive awareness in the team in the roster/dynasty before applying the patch. The better offensive players CPU gets into their team, the less one player dominates their offense. It is just the way NBA Live 07 deals with it, there is no -team leader- or -leader duo- in NBA Live 07, and statistics are purely recalculated when a team gets a new star into the team, making them split their scores.

Anyways, what this patch does and how to install it, is follows:

1. SIMENG.TXT
Mainly tweaks assists. This ends up league leaders being around ~10APG.
How to install:
make a backup of your original simeng.txt in case you wish not to use the patch in the future. Then copy the simeng.txt from simeng folder in *.rar file into NBA Live 07's simeng folder

2. FATIGUE FIX
The patch redistributes Fatigue ratings in the league, making 0-99 ratings being changed to 0-84. This fix is necessary so that every player could get fatigued when playing the game.
How to install:
make a backup of your saved dynasty or roster file. Extract all files from ratingsfix folder into your roster/dynasty folder and execute FATIGUE FIX.BAT. You should execute this patch only once per roster/dynasty.

3. ROOKIE FIX
This patch is meant for rookies. You should -not- run it in the first year of your dynasty because the first year fatigue fix is already handled by FATIGUE FIX. However since FATIGUE FIX is only meant to be run once, ROOKIE FIX is used to make sure that NBA Live 07 new generated rookies would also have their Fatigue affected like the rest of the league has. This patch should only be used in dynasty (at the beginning of season every year after the first).
How to install:
make a backup of your saved dynasty or roster file. Extract all files from ratingsfix folder into your roster/dynasty folder and execute ROOKIE FIX.BAT.

4. MAIN FIX
This patch makes sure no player develops above 84 fatigue, and also redistributes Primacy around the league. It uses Offensive Awareness to calculate players Primacy (since in defence, rebounds etc the game simulation does not seem to take into account Primacy anyway) and improves guards and SF primacy while holding PF/C primacy down more. This results in guards being able to be among MVP contenders.
How to use:
Extract all files from ratingsfix folder to your dynasty saves folderThis patch should be run always when a team in your dynasty makes a trade, has a player injury and so on. Because of this, it is only really good for player-played dynasties and not only-simulated seasons (where constant updating would be a problem). This is so because NBA Live 07 automatically recalculates Primacy based on teams lineup changes. This patch should only be used in dynasty.

ORDER OF INSTALL:
IN FIRST DYNASTY SEASON
0. make backups of your roster and simeng.txt
1. place simeng.txt to simeng folder in installation directory
2. extract ratingsfix folder into your roster/dynasty save folder
3. run FATIGUE FIX.bat
4. run MAIN FIX.bat (and run it again after every ingame week or so)
5. enjoy the game

IN 2+ DYNASTY SEASON
0. make backups of your roster and simeng.txt
1. place simeng.txt to simeng folder in installation directory
2. extract ratingsfix folder into your roster/dynasty save folder
3. run FATIGUE FIX.bat
4. run ROOKIE FIX.bat
5. run MAIN FIX.bat (and run it again after every ingame week or so)
6. enjoy the game

UPDATING ROSTER ONLY
0. make backups of your roster and simeng.txt
1. place simeng.txt to simeng folder in installation directory
2. extract ratingsfix folder into your roster/dynasty save folder
3. run FATIGUE FIX.bat
(this only updates the Fatigue ratings so that every player has at least some fatigue decrease in game)

Example pictures:

Shows the improved assists per game (full season). Yes it doesn't show anyone in 10+, if you want more assists per game, increase float values in simeng.txt Assists section a little bit. I prefer current range since ingame the computer calculates assists for players differently from how they are calculated in real NBA, thus it's very hard to get 10APG even with Nash. Reason why Nash does not lead the assists even if he has 99 Offensive Awareness and 99 Passing is purely because Phoenix has alot of offensive players and NBA Live 07 calculates assists in share-basis, depending on every players ratings. And reason why Arenas is leading this is because Washington is vice versa, Arenas is one of the few there. This cannot be changed with any simengine update, but you can change it by changing players ratings in the team.


The only way to really get NBA Live 07 have players score in 30's is to have a team that has a lone offensive superstar in the team, like Kobe. Sadly the default rosters, as I explained before, have LeBron in a team that is way overrated by EA Sports, with Hughes and Ilgauskas rated as amazingly strong offensively, thus LeBron scores merely 22. If you want to improve this, decrease Hughes and Ilgauskas Offensive Awareness before using the patch.


The sweet MVP list. Arenas features here again since he posts great individual stats being a player for his team that he is. Sadly NBA Live 07 does not take into account team success for MVP Race (last 15 years the MVP has been won by a player of top-3 teams 14 times), but you see good guards and forwards here without forward/center dominence anymore.

It probably still requires updates and fixes, but it's ready to be released to people for testing. Just remember to back up your roster/dynasty before testing it out.

If you do not like the changes and the patch, just delete the patched simeng.txt and roster/dynasty folder and restore the ones you backed up. It works for me and I'm happy with the results, hopefully it'll increase your enjoyment of the dynasty mode as well. Sure, having to run MAIN FIX.bat after every trade/injury is troublesome, but I personally just do it always before running NBA Live 07 so it's never much of a problem.

I also need feedback, let me know how it works for you.

DOWNLOAD:
http://kristovaher.pri.ee/nbalive/simeng_patch/chris_simulation_engine.rar

[hallzofillusion]

wow looks amazing Chris keep up the good work. i admire your intelligence.

[Chris_23]

Oh, darn, there seems to be a problem with Fatigue patch, for some reason it updates all fatigues to 12. But it's not important for stats ingame and is only used in game, so refrain from using it right now and sorry for the inconveniance, I'll try to find a solution asap.

EDIT: I found the solution. It was a result of loop that downgraded all fatigue ratings because of previous changes and the listing needed to be exactly the opposite. Will update the fix soon.

EDIT2: Updated. Sorry for that little thing I missed about the order of fatigue ratings update. It works now the way it was intended

[Patr1ck]

Looks very good, Chris. You took into account that the rosters had a big impact on the outcome. I could have never made those batch files. I'm not too surprised to see Arenas at the top of the MVP voting, because when I got the sim engine to where I wanted, he was the only guard showing up consistently as an MVP candidate.

[Ryan]

This works like a charm. The only problem for me so far is Lebron not scoring enough, but I think it is because of the playing time % in the game. The cpu made his playing time like 59 because Marshall (77 ovr) is his backup.

EDIT: Also PF's and Centers like Camby, Kenyon Martin, and Drew Gooden are only averaging between 6-9.5 ppg in my dynasty so far.

[Chris_23]

I suggest you increase players offensive awareness with dbf editing if you wish to make them more of an offensive force and vice versa if you want the opposite. Camby was considered not much of an offensive force by EA thus his awareness was low and the primacy in my patch depends on awareness.

LeBron situation is due to EA seriously making Cleveland better than it really is. Cleveland has three great players with amazing offensive awareness and they score alot thus taking away points from LeBron (but he still does score around 22 instead of 17 before), and sadly it seems that minutes can -not- be changed at all unless anyone knows where are the files that affect rotation? Gooden has made a consistent improvement in this season and EA's default rosters have him based on his weaker previous years which is why he's underrated.

CPU takes Player Rotation into account when determing minutes in the game, and every position has rotation based on Overall rating and player position. Unless it's an option to tweak how computer calculates player rotation, LeBrons game time gets cut as a result. 77 Overall is absolutely great for a rotational player, so that's what happens.

If theres anyone who knows whether Player Rotation CPU is hardcoded or not and if it can be tweaked or not should let me know since currently the CPU does not take into account in simulation any fatigue rating (fatigue is seemingly used for ingame fatigue when playing the game, but not simulated game minutes, and it's used for overall tiredness rating in dynasty).


Anyways, as for offensive awareness being the one that distributes primacy (but not 1 to 1). The reason why I made it so is because simulation engine does not take offensive awareness into account much. It's only used in offense and might, together with FG ratings and Passing ratings, determine how many points or assists player scores.

But points were a problem already, thus I changed primacy so that forwards with high offensive awareness would also have high primacy, but that's only the top of the iceberg forwards (like Garnett). The rest of the forwards and centers have lower primacy if their awareness is low as well.

And assists were not dependant on primacy at all and merely used the already existing offensive awareness and passing to calculate the assists, which is why offensive awareness was the best way to go about it to calculate simulation engines primacies in roster. I only wish EA had another way to deal with how much a player is used in a team instead of recalculating primacy. Hells, I'd be happy if I could tweak primacy calculation formula directly, but it seems to be hardcoded.

[e-mat]

Sadly, Donyell is not the problem. LeBron plays about 65% of the game even if his backup is a 60something Ira Newble...

I got a question Chris - can I run the rookiefix.bat in the first year and not run the fatigue fix at all?

btw, judging by this season, Yao is the main option in Houston. Not Tracy.

[Chris_23]

You don't have to run Fatigue Fix, but the thing is that fatigue fix fixes the whole roster so that 0-99 fatigue is in 0-84 range (not capped, just re-rated into the other range).

Rookie Fix does the same thing but only to players with no years of experience (if a player does not play a single game, he doesn't have his years exp increased, this means that the rookie fix also updates -again- fatigue of players who did not play the previous season. But that's not of an issue since those players are usually around 50 overall and don't get game time anyway).

You don't have to run fatigue and rookie fix at all. MAIN FIX also caps fatigue to 84 maximum without rebalancing anyone who's below that. It's just that FATIGUE fix and ROOKIE fix both redistribute fatigue ratings in a new range, while MAIN fix merely caps it and holds it on 84 at best.



About T-Mac and Yao. Yes, you are correct, this season Yao is the main man and T-Mac is not. But default rosters have overrated T-Mac due to cover-guy aspect, and Yao is making his breakthrough season right now in offense.

Yao actually has a decent offensive awareness rating already, but to get more offense from Yao you should increase his offensive awareness to 90. The primacy rebalance makes it so that every center and forward with 90+ offensive awareness get primacies between 77-99 and make them the offensive power in points and MVP race.

[chaodck]

What I usually do to make LeBron play a whole lot more is to eliminate all SF's from his team. For instance, make Marshall PF/C, and make Ira Newble SG/SG, and if the game has got no SF's to backup James, it'll make Hughes (SG/SF) his replacement, but as he is a starter he'll get only like 5-10% playing time, which means James is playing almost 90% minutes (which is kind of accurate).

You can do this with any team, with a star player playing low minutes.

Hope it works!!

APB.

[Walleris]

Dude, sorry, but I don't understand what this patch do??

[Chris_23]

This patch redistributes fatigue ratings (so that every player in game could get fatigued, currently every player with 84+ fatigue can play the whole game without any fatigue loss, this just redistributes everything to 0-84).

This patch also updates simeng.txt so that assists are simulated properly in CPU vs CPU games.

And this patch updates new rookies fatigue ratings as well.

And this patch rebalances primacy calculation around the league so that guards and small forwards can score in simulated games as well.

(was it -that- hard to read through the original post in this thread Walleris?)

What I usually do to make LeBron play a whole lot more is to eliminate all SF's from his team. For instance, make Marshall PF/C, and make Ira Newble SG/SG, and if the game has got no SF's to backup James, it'll make Hughes (SG/SF) his replacement, but as he is a starter he'll get only like 5-10% playing time, which means James is playing almost 90% minutes (which is kind of accurate).

You can do this with any team, with a star player playing low minutes.

Hope it works!!

Yes, this is a way to make it work. But I can't imagine myself writing a batch that does it automatically. It might mess up teams that do rely on bench more and it's a mess otherwise.

I've just accepted that James does not get so much time in such a team. Compared to Kobe for example.

[Noah]

man... nice work!!...id be pleased about a screenshot showing us how your patch affects rebounding! there are often 15 players or more who average more than 10 rpg...maybe this could be fixed ?

[Chris_23]

Decreasing rebounding you mean? I'll look into it, but the comparisons I did did not seem to differ too much from how it is in real NBA (with 13 players currently at 10+). I might look into it at some point.

I'm currently trying to see whether it's possible to write a patch that dynamically affects teams playbook depending on which players they have in the team.

[JJones]

This patch looks simply amazing I will make sure to check it all out ASAP and respond with my findings. Thanks Chris.

EDIT (didnt want to double post)

Decreasing rebounding you mean? I'll look into it, but the comparisons I did did not seem to differ too much from how it is in real NBA (with 13 players currently at 10+). I might look into it at some point.

I'm currently trying to see whether it's possible to write a patch that dynamically affects teams playbook depending on which players they have in the team.

oh i like that second part, it would be a godly idea. good luck i like what this has done to my dynasty so far. 2 Guards and 3 big man for MVP race. Alot better then before.

[Chris_23]

That is great to hear

The patch does have tendecy to favor in points players with offensive awareness now, so you may have players who usually don't score that much yet are rated good in offense scoring more points. I could have of course also checked for FG's in patching, but that would affect again other things etc etc. So basically offensive awareness was the best and closest thing to go with for basic guideline for primacy distribution and if people see some players scoring more for AI than they should, tweaking that down helps the issue before patching.

It is so akward that NBA Live 07 uses merely few ratings in simulation, not all ratings ingame (primacy has no effect ingame) and such, but we got to live with all that.

Also, you can see higher point records in box scores as well now. I've seen Kobe above 40 a few times by now which is certainly nice, and alot of houston games have both T-Mac and Yao 20+.

[---]

Chirs_23, I have a quick question for you. If I was to tweak my simeng.txt in NBA LIVE 2005, would my dyansty that is already 11 or so games feel the effect of it? I'm guessing yes but I would just like to make sure.

[Chris_23]

Yes it would, simeng is not roster/dynasty dependant.

[ericisgod]

thanks

[Chris_23]

I am working on v2 of this patch.

What you will hopefully see in the next version:

Better primacy distribution based on how good a shooter is. This will make it so that even if a player may be amazing in offensive awareness but cannot really shoot, he would not try that much in simulated games.
The same with forwards. Forwards with better inside scoring will get a bit better primacy, power forwards as well, while lowering those who really are not meant for offense in game.
Better points distribution across the team, which will hopefully fix the issue of certain teams being too balanced (like Cavaliers having everybody rack up good amount of points).

I will also look into how much do certain questionable ratings affect how a player performs in played games. This means that I'll test whether offensive awareness and primacy actually affect ingame perfomance (other than passing and fast breaks) in significant ways.

And the best thing about this patch would be that it's part of MAIN FIX, which means it can be applied whenever, to any dynasty in progress without any problems.

(gosh I wish NBA Live 07 would be open source.. I really would love to dig deep into it's code as a programmer)

[Noah]

chris23...what do u think about the idea that the amount of blocks in a simulated game is affected by their rating in defensive awareness ?

[Chris_23]

I don't see much wrong with current blocks. Good blockers seem to lead the league in blocks so I see no problems having to change that.

I will not make a patch that will increase players blocks based on defensive awareness because it would be pointless if the simulated engine does not care about that. I only used offensive awareness link to calculate primacy curve only because it was the most obvious thing to tie it with since Live 07 recalculates it based on hardcoded value every time.

Unless it becomes necessary to increase simulated blocks, then the way to go is simeng.txt, but other than that I don't see the point But thank you for the suggestion, I love suggestions and keep receiving many of them in PMs as well

[michaelc37]

I loved ur, primacy fix for sim stats. more specifically the Main.bat file.

I TONED DOWN big men primacy by OFFABILITY, BUT added opportunity for BIG MEN(c,pf) to get more primacy based on the following: (i had GREAT RESULTS)

(1) DSHOTRANG->like kevin g, tim d, dirk n, lamar o.
(2) INSIDESC -> really good inside scorers deserve more primacy
(3) OREABILITY -> allows guys like ben wallace to get about 10PPG even thow he doesnt have alot of offability, his oreability allows him to get second chance points.
(4) DSTRENGTH-> guys like shaq has doesnt get extra primacy for dshotrange, but he is so strong. he gets extra primacy for his strength, as well as ben wallace.

I have had a range of different results, but for the most part they a realistic and there is alot of depth.

Next Update: Am currently workin on adjusting primacy for different types of point guards(based on superstar(not scorearea) and offability);e.g.:
EDIT: jason kidd, steve nash-> passer->less primacy than scorer PGs.
EDIT: gilbert arenas-> scorer-> more primacy than passing PGs.

Thanks to chris_23 for allowing me to modify and post his original file:

DOWNLOAD this fix here is the link.
http://michaelc.slylabs.com/mainfix.zip

Picture:

[Chris_23]

Nice to see you did not make any rookie-scripter mistakes in the patch, it looks good

I will personally not base any primacy on shooting ranges, because it just makes no sense. Offensive rebounds are a good improvement and I'll take that into account in my update as well, along with inside scoring.

How will you handle Arenas? In default rosters Arenas already scores alot in game so I don't see why he would need even more primacy? Arenas is not a player whom I'd see being such a high scorer in another team for example, in Wizards he is the prime shooter.

And how will you deal with those passers then? Just decrease primacy based on their passing rating? Nice idea actually, the more you pass the less options you have for shooting. You've given me some good ideas here, thanks, it's great that this patch has inspired people to tweak their own, since it only needs a little bit of scripting.

By the way what rosters are you using? I noticed LeBron scoring alot, which would be hard if the team has alot of good players.

[michaelc37]

You right arenas does score alot in the default rosters.
sorry i was not clear when i said talked about the primacy for point guards.
when i said:
jason kidd, steve nash-> passer->less primacy.
gilbert arenas-> scorer-> more primacy.

i really meant:

jason kidd, steve nash-> passer->less primacy than scoring PGs.
gilbert arenas-> scorer-> more primacy than passing PGs.

And the primacy isnt just based on shooting ranges. what am sayin is if the big man has has good shooting ranges. he gets a little extra amount of primacy(dirk,garnet,duncan,bosh,odom).
However, if he doesnt have good range,BUT has great strength strength(Shaq), he gets a little extra primacy.
Same with inside scoring...
In my version the primacy for big men is NOT just based on offability alone.

I can understand why one wouldnt see gilbert arenas as scoring on another team, but at the same time we have seen him play on another team, so its unfair for me to judge/predict him like that; my considerations on based on what he has done, and what he is currently doing in the league on his team.

in my next version. i am redoing the how primacy is distributed for PGs, based on a combination of default SUPERSTAR(realized i cant use scorearea) and OFFABILITY. As i now know SUPERSTAR can help determine if the point guard is a great scorer or a regular passer.

Notice how all scorer point guards superstar is 17 to 18:

[Chris_23]

Yeah I thought about superstar as well, but will probably stay away from it (since player can be many at once and I don't want to cut down on good scorer because he is also a playmaker. Yes, I can write conditions to check these things but it would make all this too long and not worth the effort). The passing idea is good though, I'll cut down on PG's points with it.


In [Simulation Engine] 2
Tweaked simeng.txt to adjust rebounds and assists
If player passes more, he scores less
Point Guard and Shooting Guard adjustments based on passing, dribbling, and shooting
Small Forward adjustments based on passing, dribbling, shooting strength and inside scoring
Power Forward adjustments based on rebounding, inside scoring and shooting, strength, dribbling and ballhandling.
Center adjustments based on rebounding, inside scoring and shooting, strength, and dribbling.
Adjustments based on years in the league (less hyped new players in the league tend to blossom few years later in the league even if their skills would be ready for more)
Bench Players tweaks, players in the roster who are on bench have less primacy in simulated games.
Other minor tweaks

Will be testing for a while, might release later today or early tomorrow.