Tendency edits that are game changers for realism?

[King Gro23]

Anybody having trouble connecting to 2k servers?

Sent from my VS987 using Operation Sports mobile app

[King Gro23]

Sent from my VS987 using Operation Sports mobile app

[vannwolfhawk]

Update on a few tendency findings…

1. POST UP MOVES - I tried everything with post up tendencies to get players to use player specific moves. I zeroed out everything, went 100 everything, went 0 in moves they didn’t use and 100 they did. I went high in moves they did and 0 in ones they didn’t. I messed with shot IQ. Nothing was working and they were doing moves I had them zeroed in. Then I tried averaging everything at 100 total for all the moves they would use and BINGO I got it. So, for example I’d put R hook at 25, L Hook at 15, Drop Step 10, Up & Under 5, Drive 5, Spin, 5, Right Fade 25, Left Fade 5, Post stepback hop shot 5 (these numbers are examples) and they all equal to 100. I then saw only those moves with hooks being the priority. The only bad thing here is it will be a time consuming Edit. I tested with the Greek Freak & M. Gasol and finally got great results after a few days. I have a few teams done but still need to do this league wide.

2. DRIBBLE EDITS - I have mentioned these before as I said I eliminated all the DRIVING half spins, spins, double crosses, etc. Well I now eliminated everything including the basic crossover. The only thing I left was behind the back at 5 for most everyone but bigs or bigs that don’t use them. I did keep setbacks for those who use them and high with J. Harden, C. Paul, I. Thomas, etc. The reason for this is 3 fold. Number 1 it slowed guys down when making a move on the dribble leading to slow layup animations. Number 2 the moves looked bad and no one uses all these moves when attacking the hoop on a drive. In a setup yes but rarely on the drive while attacking. 3rd is the behind the back dribble is used perfectly and looks great as they use it when cutoff by a defender. The crossover even though at 0 is still used as if they change directions they use it to switch hands anyways. They are both used appropriately and it doesn’t kill the players momentum. Lastly is the setbacks create space but then can regain momentum for a faster layup animation. It plays 1000x better with the league globally edited like this. I have already done all this in my roster.

3. PASSING TO OPEN MAN - This is a huge one and tough one to balance, but I have no players below 40 in this area. To create the drive and kick’s for wings depending on the player should be between 55-65 and better passers higher than that. Bigs should be between 40-60 unless guys like Jokic, etc… It really is a huge game changer as now you will see guys like Tony Allen receive a pass on a cut, drive and kicks to the corner for 3. Most guys in the NBA can penetrate and kick. The balance is making sure some guys aren’t always threading the needle, throwing full court passes, or standing at top of key and rifling passes cross court to the open man because the help defender is sagging too much. But the great part along with the drive and kicks is guys will feed the post better, make those flare passes, have less idle situations, or drive and put nose down and force way to the hoop for 5 seconds (ties into drive strong tendency as well). I’ve don’t and scaled the whole league like this as well….

I did try for testing purposes the ball control at 25 (lowest it could go) as that was mentioned earlier in the thread. I wasn’t a fan for 2 reasons. Number 1 against a user it doesn’t translate and too many steals and turnovers although you could adjust sliders to compensate but more importantly it severely slows their ball handling and moves down.

[King Gro23]

Ps4 test file is up
Tendencies and rating adjustments for gameplay, minor minor change mainly post shots on stars

Psn : TopshotGwaup15

Available teams: Cavs , warriors , thunder , t wolves , pelicans
Wip

I use Armor swords purple set w tweaks.

Roster base was the OS community edit based of Oct 28th with - Vanns playbooks. Advanced rotations,

All credit goes to community and their efforts.



Sent from my VS987 using Operation Sports mobile app

[SonicMage]

Quote: Originally Posted by vannwolfhawk

Update on a few tendency findings… 1. POST UP MOVES - I tried everything with post up tendencies to get players to use player specific moves. I zeroed out everything, went 100 everything, went 0 in moves they didn’t use and 100 they did. I went high in moves they did and 0 in ones they didn’t. I messed with shot IQ. Nothing was working and they were doing moves I had them zeroed in. Then I tried averaging everything at 100 total for all the moves they would use and BINGO I got it. So, for example I’d put R hook at 25, L Hook at 15, Drop Step 10, Up & Under 5, Drive 5, Spin, 5, Right Fade 25, Left Fade 5, Post stepback hop shot 5 (these numbers are examples) and they all equal to 100. I then saw only those moves with hooks being the priority. The only bad thing here is it will be a time consuming Edit. I tested with the Greek Freak & M. Gasol and finally got great results after a few days. I have a few teams done but still need to do this league wide.

Do you know specifically which post tendencies need to add to 100? For instance, is aggressive backdown included? Shimmy shot? How about face up?

[vannwolfhawk]

Quote: Originally Posted by SonicMage

Do you know specifically which post tendencies need to add to 100? For instance, is aggressive backdown included? Shimmy shot? How about face up?

Only the actual post moves. You can still put 100 on backdown, aggressive backdown, face up (although I use 100 split between backdown and faceup). Shimmy I use the same value as the highest fadeaway so that also does not factor into the 100 split between the post moves.

It is really odd though. It was bugging me that I wasn't seeing any hook shots or specific moves which is what made me go to town on this. I'd like others to test and see for themselves but I was seeing more drives out of the post with it set to 0 and hooks at 100 or 99. Same for drop step's, step back's, fades, etc. It's like it got zeroed out and you would never see it. With 100 averaged out I got the top moves more with lower ones mixed in and great variety...

[SonicMage]

Quote: Originally Posted by vannwolfhawk

Only the actual post moves. You can still put 100 on backdown, aggressive backdown, face up (although I use 100 split between backdown and faceup). Shimmy I use the same value as the highest fadeaway so that also does not factor into the 100 split between the post moves. It is really odd though. It was bugging me that I wasn't seeing any hook shots or specific moves which is what made me go to town on this. I'd like others to test and see for themselves but I was seeing more drives out of the post with it set to 0 and hooks at 100 or 99. Same for drop step's, step back's, fades, etc. It's like it got zeroed out and you would never see it. With 100 averaged out I got the top moves more with lower ones mixed in and great variety...

I just did a little testing with this, and I don't think the count up to 100 method is correct, though I can certainly see why it would look like it is.

It has to do with the way the AI is written. By doing a count up to 100%, you're kind of assuming that each post move type has a chance relative to each other. So 25% of the time they would hook left, 15% they would hook right, etc. However, this would actually end up with a very awkward looking AI, in my opinion. The player would be rolling the weighted dice behind the scenes, and then forcing themselves into position to perform whatever move was the winner. But that's not really how the mind of a real life player in the post works, and I'm sure you know that from coaching. A player doesn't predetermine what move they're going to make ahead of time, and then "stick to the plan", so to speak.

No, what I think the tendency values represent are likelihoods. So they have nothing to do with how often they perform relative to each other, but rather how likely a player is to perform a certain move *when they are in position to do so*. And this makes more sense when thinking about post play in real life terms. In a split second a player will instinctively think to themselves "I'm in a good position to make this move, so should I do it?" Feel free to correct me if I'm wrong, but that's how my mind works when playing basketball.

I'll tell you one test I performed that confirms my hypothesis. I set 100 to hook left, and 0 to all the other post shot types. I had 0 for all the post position types of faceup, back down, and aggressive backdown. So what happened? My post player would catch the ball in the post and spend several seconds rapidly switching back and forth between facing up and backing down (which confirms my "likelihood" hypothesis for these positioning tendencies as well) before finally either forcing a shot or passing back out, and about every tenth possession or so I'd see a hook over the left shoulder because they just so happened to get into position for one by good fortune. Then I performed another test. All I changed was raising the backdown tendency up to 100. Now I see hooks over the left shoulder much more often, because the post player would often be backing down and in position for one.

So there's something else that can be done to test this theory. Try taking those tendencies that you had in your example, and doubling all of them. If you see those post moves being performed more often, then it's more likely that my theory is correct. If it's about the same frequency as it was, then your theory of "add up to 100%" is more likely. I'm curious to hear the results.

[J9mils]

alright first off ill start with something easy. POST GAME. First off you wanna have strength set to 0 in global sliders. since ive done this the post game is a lot better. really that's all I did. I did set all of durants moves to equal value, but never touched Horfords. But both make great decisions and show a variety whem posting. actually thought I had more to show, but sometimes I get side tracked and don't record when I thought I did.(sorry youll hear my kids in the background lol) but this is just some examples ill defintly get more. and again I know my graphics suck lol.

https://youtu.be/ppZIenW-508
https://youtu.be/3kIs02ffSzk
https://youtu.be/22FgxmbsV8o
https://youtu.be/91Wx836agTU
https://youtu.be/4mqJf21V_qM
https://youtu.be/jInA-Zfewgg
https://youtu.be/LGQWbkxdzws

https://youtu.be/hijKRiA1Glc