Announcement

Collapse
No announcement yet.

Home Advantage

Collapse
X
  • Filter
  • Time
  • Show
Clear All
new posts

  • #46
    Originally posted by Gonira View Post
    talking about side effects, be cautious with the acceleration slider, as it is directly linked to fatigue. if too low, players never get tired. if too high, players get tired too fast.
    That's a great tip, never heard that one, good to know.

    Comment


    • #47
      wow, i'm just thinking i might get very realistic accuracy on this without even needing to go too drastic.

      after some preliminary testing, +/- 7 pass error seemed to make the advantage too exaggerated, so i dialed it down while checking some stats until i settled in +/- 3.

      i also kept the line height offset, but i changed it to +/- 1 instead of +/- 2. personal preference tbh. in my set, center backs of the home team sometimes got a little bit crazy going to attack when this was two points up.

      then i played a 2015/16 champions league campaign with psg.

      results:

      home win 2-1 vs shakhtar
      away loss 1-2 vs malmo
      home loss 1-2 vs real madrid
      away win 3-1 vs shakhtar
      away draw 1-1 vs real madrid
      home win 2-0 vs malmo
      home draw 1-1 vs bayern
      away draw 0-0 vs bayern

      gg. god, i hate the away goal rule.

      home results: 2W-1D-1L, 6GF, 4GA
      away results: 1W-2D-1L, 5GF, 4GA

      this doesn't tell much. the results of 8 games aren't nearly close to enough to have big conclusions. happily tough, shot stats, because the numbers are bigger, converge much much faster than final results and can give me a more accurate idea without having to play a shit ton of games.

      so in these games, home teams, on aggregate, scored 10 times while away teams scored 9.

      the shot count of home vs away was:

      home: 78 shots, 42 on target
      away: 60 shots, 31 on target

      for true-to-life realism, home teams have to shoot 25 to 30% more than away teams. in the long run, this will also make them score something like this, 5 home goals for each 4 away goals, or 4 to 3, something in this ballpark.

      in this low sample i got, the home numbers were 30% more total shots, 35% more shots on target, 11% more goals.

      a very good start.

      i'm gonna repeat the experiment to have a bit of a bigger sample just to be sure. if it goes well, i might just settle, not in the mood to test this for several days and i'm already lightyears ahead of when i was wasting my time with those useless ini's. oh, and this time i'll go with united, i swear!

      what i like the most about the way i made this is that the changes are perfectly symmetric, so the same tweaks would probably also work super fine on every version of fifa that has sliders, regardless of your baseline slider set or if you play manual or assisted.
      Last edited by Gonira; 13-10-2016, 06:50:AM.

      Comment


      • #48
        i stopped using the line height offset, because the cpu conversion rate per shot started to get way out of control in my home matches. it seems the extra point in line height makes the holes in my defense give them many better oportunities for clear shots and they're too good in finding the back of the net in these cases (at least much better than i am).

        other than that, the results of the -3/+3 pass error experience is inconclusive, even after playing 26 games with it. it might be just right or it might be too much.

        results:
        11 home wins, 6 draws, 9 away wins

        shots:
        283 home, 208 away: 36% difference

        on target:
        164 home, 112 away: 46% difference

        goals:
        38 home, 34 away: 12% difference

        it's clear to me that:

        1. it works. yay!

        2. if i keep wanting to nitpick on finding the magic numbers, i'll need help. i can't just play this many games and i'm so obsessed with this right now that my productivity will go down the drain.

        i started a thread at the operation sports forums for people to try the pass error tweak and post their stats so we try together to get there. it's a nice way to see if this tweak also goes well in different versions of fifa and different slider sets. i decided to suggest -2home/+2away there to see the effects, as the shot count i got gives me an impression that -3/+3 is probably a bit too much.

        the thread is this one. if someone wants to test the settings for at least a couple of games and share their results there, it will be of much help:

        http://www.operationsports.com/forum...ting-fifa.html

        Comment


        • #49
          I love the innovation and effort, you're a dying breed. 15 years ago, sports game forums were full of guys sharing their individual tweaking discoveries. Now, people just sit around and complain about the game while waiting for Fidel, Paul or regularcat to fix it for them lol.

          That said, I really can't see how a couple of slider ticks are going to help with this. Because the moment the CPU AI boosts kick in, all those slider settings go out the window. You haven't noticed how the CPU always attacks like crazy in the first few minutes and last couple minutes of the first half, or at the end of the game, etc.? That stuff goes on throughout the game, with the specific degree of crazy depending on which teams and competitions you're playing. There are no sliders that can offset all that pre-programmed nonsense, other than pausing the game when one sees the boosts kick in, and lowering CPU speed and accel sliders by 10 or 15 points ... which I have honestly sometimes done in the past when playing on Legendary.
          Last edited by bangus; 15-10-2016, 09:55:AM.

          Comment


          • #50
            I watched this animated short years ago called "Home On The Rails." The scene is a senior lady's home. She's knitting, and drinking tea, and a fly buzzes about the room. All very ordered and mundane. Suddenly the doors of her house fly open and a freight-train thunders through her living room, in one door and out the other, 30 seconds of chaos and ear-splitting noise. And when the train's gone, she goes back to her knitting and tea while a fly buzzes about the room. We're the senior lady trying to come up with the perfect slider set. And EA's CPU AI boost programming is the freight-train.

            Comment


            • #51
              Originally posted by bangus View Post
              I love the innovation and effort, you're a dying breed. 15 years ago, sports game forums were full of guys sharing their individual tweaking discoveries. Now, people just sit around and complain about the game while waiting for Fidel, Paul or regularcat to fix it for them lol.

              That said, I really can't see how a couple of slider ticks are going to help with this. Because the moment the CPU AI boosts kick in, all those slider settings go out the window. You haven't noticed how the CPU always attacks like crazy in the first few minutes and last couple minutes of the first half, or at the end of the game, etc.? That stuff goes on throughout the game, with the specific degree of crazy depending on which teams and competitions you're playing. There are no sliders that can offset all that pre-programmed nonsense, other than pausing the game when one sees the boosts kick in, and lowering CPU speed and accel sliders by 10 or 15 points ... which I have honestly sometimes done in the past when playing on Legendary.
              this things in fifa 17 is less marked than in 16, but the center of gravity of team is lowest too low than 16.
              change tactic but nothing change.
              Legend level is less competitive than fifa16(modified not clean)


              Gonira good test!bravo

              Comment


              • #52
                Originally posted by bangus View Post
                That said, I really can't see how a couple of slider ticks are going to help with this. Because the moment the CPU AI boosts kick in, all those slider settings go out the window. You haven't noticed how the CPU always attacks like crazy in the first few minutes and last couple minutes of the first half, or at the end of the game, etc.?
                when you think about it, it's not just a couple ticks, because in fact the difference going from home to away is you changing 4 ticks to one team and 4 ticks to the other in the opposite direction.

                for example, i'm going from 51/65 neutral to 49/67 home and 53/63 away. so in my set, pass error has an 18-point gap home and a 10-point gap away. that's 8 points of difference, in what in my experience is one of the sliders that are most influential in difficulty. perhaps it's even mild in fact, at this point everything is kinda close, but for me it's already proven by my results that -3/+3 definitely does the job, at least in fifa 14, and perhaps even "too well".

                in '14 i only notice the cpu attacking like you're saying at the last 10 minutes of the game, if they're losing. there's definitely some stats boost going in this situation, tactics alone wouldn't make them so furious. but i live with this, especially because the same thing was 100 times more intense in '13, i remember. almost a game breaker.

                anyway, i think it's wrong to assume these scripts would override sliders to the point of making futile any effort to manipulate the game. it's all about what happens on average in the full 90 minutes.
                Last edited by Gonira; 16-10-2016, 12:21:AM.

                Comment


                • #53
                  i had success with -2user/+2cpu pass error home and +2user/-2cpu pass error away. it seems to be about right to replicate home advantage realistically.

                  i'm quoting the post i made about my conclusions at the os thread, because i think i can't put it better than i did there.


                  20 games in.



                  8 home wins, 5 draws, 7 away wins

                  30 home goals, 23 away goals (home +30%)



                  i'm right at my target. the win/loss distribution is almost even (definitely not what we want), but this is where 20 games is not even near to remotely close of being a sample.

                  out of curiosity, i checked the last 20 games of the biggest leagues just to show how this can vary wildly in such a sample:


                  premier league

                  7 home wins, 8 draw, 5 away wins
                  30 home goals, 19 away goals (+58%)


                  la liga

                  10 home wins, 5 draws, 5 away wins
                  41 home goals, 28 away goals (+46%)


                  bundesliga

                  10 home wins, 8 draws, 2 away wins
                  30 home goals, 17 away goals (+76%)


                  serie a

                  8 home wins, 5 draws, 7 away wins (same as me)
                  24 home goals, 29 away goals (-21%)


                  ligue 1

                  12 home wins, 4 draws, 4 away wins
                  29 home goals, 23 away goals (+26%)


                  so this is why i decided to primarily look at shooting instead of wins or goals. bigger frequency = less variance = bigger confidence.


                  and this is the breakdown of these 20 testing games:

                  as i said,

                  8 home wins, 5 draws, 7 away wins

                  30 home goals, 23 away goals (+30%)

                  shots:
                  172 home, 142 away (+21%)

                  on target:
                  109 home, 78 away (+37%)

                  possession:
                  49.3% home, 50.7% away

                  shot accuracy:
                  63.4% home, 54.9% away

                  pass accuracy:
                  80.7% home, 81.9% away


                  the tweak has proven to work in providing home advantage in my games. i came as close as i could come to real life shooting and scoring stats, under the limitations of this sample size. anyway, the worst case scenario here is "not quite there but close enough". i can live with that, quite happily. no one has ever done something so analytical in this subject, so i'm certain this is the closest solution we have for fifa at the moment.

                  i'm glad to see pass accuracy itself wasn't affected by the tweak, since there isn't a big discrepancy in home vs away pass accuracy in real life too. it's more about the quality of the passes. i think this is the reason home teams had the edge in shot accuracy here. good passes just put them in better spots to shoot.


                  i'm done testing this for now. got where i wanted, no need to change the initial setting. i'll put fifa to rest until ariel releases an update for the moddingway patch with the new season rosters, then i'll restart my valencia career (i was sucking as much as they did in real life in the last season anyway). when i finish my first season, i'll post the home/away records here just to check what we got.


                  i know this was done solely in fifa 14 and there's a new engine, yada yada, but as i said, i believe i'm the first one to look seriously into home advantage in fifa, so if i wanted to have home advantage in fifa 17 or whatever, i'd certainly start with this same pass error tweak. the concept is as close to fail-proof as it can get.

                  in case someone does it, i would deeply appreciate some results being shared here. i'm very curious about how it would translate.


                  see ya'll soon!
                  Last edited by Gonira; 17-10-2016, 01:43:AM.

                  Comment


                  • #54
                    Originally posted by bangus View Post
                    That's a great tip, never heard that one, good to know.
                    bangus, i found a way to get over that problem of fatigue being linked to the acceleration slider, it might help you.

                    quoting kazemaky, from the os forums:

                    As i said in another thread, you can play with 45 acceleration if u want. Just put the half time you really want into the locale.ini like this:
                    "Head to CWherever you installed it\Origin Games\FIFA 17\Data\
                    Open the locale.ini file
                    Add the following in the end:

                    SKIP_BOOTFLOW=0

                    HALF_LENGTH=10

                    AI_SETTING_HALFLENGTH=10

                    //

                    Save the file and you are done."

                    Now Im not really sure what the bootflow does, but after editing this, you go to ingame options and use alower value halftime. This way game uses locale ini time, but ingame ht stamina and it balancea out with lower acceleration. Just have to find the corresponding values for most realistic stamina. Im using 8min ingame and 10 min real ht with 49 acc and its like 50 acc on 10min ht without file editing. Thus you shouldnt be overly concerned with acceleration being tied to stamina. It can be compensated.

                    i tried this in my cl.ini and it worked. the code overrides the half lenght chosen in the game, but the game is tricked into thinking you're playing with the lenght of the settings screen and applies fatigue accordingly.

                    Comment


                    • #55
                      hey, just want to update and put final numbers on my research.

                      i just finished my first full season using the pass error tweak.

                      -2home/+2away pass error is really where it's at for me, at least on fifa 14. very realistic home/away records.

                      these are the records i got on this season. playing as sampdoria, using the moddingway mod. finished 5th in serie a:

                      home: 10W-7D-2L, 28GF-16GA
                      away: 8W-4D-7L, 28GF-30GA


                      i had a previous sample of 30 spare games playing 15 correspondent home-away fixtures (mostry friendlies). the following numbers are a comparison of the overall home vs away performance of these 30 games plus the 38 league games of the season i've just finished:


                      68 games,

                      29 home wins, 20 draws, 19 away wins (+53% home wins)
                      goals: 99 home, 74 away (+34% home goals)
                      shots: 686 home, 598 away (+15% home shots)
                      on target: 387 home, 335 away (+16% home s.o.t.)


                      it's just about the bias i was looking for. i can't say how the intensity of the advantage would be translated to newer games (or even different slider sets), but the advantage would definitely be there, so this is the starting point i'd recommend to everyone.

                      if you prefer to go for a more home biased intensity, to emulate south american leagues or bring you closer to the results of fifa simulated games, -3/+3 pass error would be certainly better.


                      the testing was a pain, but the tweak is so simple and so effective. i can't understand why ea and konami don't put something similar in their games. just use a similar approach, make it an offline-exclusive "home advantage on/off" switch and everyone would be happy. the knowledge that your team is expected to overperform home and underperform away adds SO MUCH for immersion! i don't know how i could live without that for so many years.

                      Comment


                      • #56
                        one more season for the books, now i have a big sample on this.

                        following my good season with sampdoria, manchester city made me an offer to replace pepe, and i got it. fun season. won both national cups, 2nd place after united in the epl and eliminated by ajax in europa league r16.

                        shot difference between home and away stayed pretty even compared to the previous results, but there was no advantage in wins and goals.

                        this is the updated picture. single-legged cup fixtures are not counted.

                        116 games

                        46 home wins, 31 draws, 39 away wins (+18%)
                        160 home goals, 134 away goals (+19%)
                        shots: 1273 home, 1100 away (+16%)
                        on target: 710 home, 603 away (+18%)


                        definitive conclusion for me that -2/+2 pass error is still not enough for realism.

                        gonna try -3/+3 in the next season. had good results with this offset in the past, but over a very small sample. anyway, it will be better.
                        Last edited by Gonira; 23-02-2017, 08:13:PM.

                        Comment


                        • #57
                          Can't find cl.ini

                          Hi gonira, im an avid fifa 14 player like you. recently i saw this post of yours regarding the fatigue problem...but i cant seem to find the cl.ini file....please help

                          Comment

                          Working...
                          X