Like many of you, I love FM and i didn't like fm26 tbh so i returned to fm24, but I’ve been getting really frustrated with the arcadey feel of the default Match Engine. I got tired of seeing laser-guided ping-pong passes and shots, defenders doing random suicidal slide tackles, and goalkeepers acting either like Prime Buffon or completely forgetting how to dive.
So, I decided to take matters into my own hands. I've spent the last few weeks digging deep into the ME files, running endless tests (simulating full seasons in Full Detail across different leagues like the Portuguese Primeira Liga and German 3. Liga), and comparing the in-game stats directly with real-life Opta data.
The goal wasn't just to make the game "harder"; I wanted to make it play like actual, real-life football.
What’s changed:
Ball Physics & Passing: The ball actually has "weight" to it now. Short passes aren't hit like powerful shots anymore, and long cross-field balls have a realistic, natural trajectory. Build-up play finally looks like real football.Defending Behavior: Defenders are much smarter. They stay on their feet, jockey properly, and only commit to tackles when it makes sense. Say goodbye to those ridiculous, game-breaking red cards from mindless sliding tackles.
Goalkeepers: I've tweaked their reaction times and agility so they behave normally. They make the saves they should, but they won't pull off superhuman, teleporting dives every single time.
Direct Free Kicks: I found the sweet spot for this. Set-piece specialists can actually score now, but at a lower realistic rate. Top teams and specialists will get around 0-5 free-kick goals a season (exactly like real life), instead of 15.
The Results: I didn't just eyeball this. After running multiple full-season simulations, the match stats are virtually a 1:1 match with real-life league averages. Pass completion (around 82-84%), shots on target, foul counts, and set-piece goals are exactly where they should be for modern football.
If you enjoy winning 6-5 every weekend with 40 shots a game, this ME is definitely not for you. But if you want to feel like a real manager where your tactical tweaks matter, player attributes actually show on the pitch, and the game flows realistically, give this a try.
I want to give a massive shoutout to the guys in the modding community who have been digging into the ME files and sharing their findings and raw data freely. I didn't discover all these variables myself. I basically took their incredible groundwork and decided to go one step further using Gemini 3.1 Pro & ChatGPT 5.4.
CI hope you enjoy and i would love some feedback so we can improve it together. Btw I use it alongside with Daveincid realism pack + my injury pack.
Installation instructions: make a backup and then delete the simatch.fmf file in the installation data directory. Then put the simatch folder as it is without chaning anything in the name in data installation folder. Enjoy. New save is not required.
MacOS file also included.
I runned my final Simulation test and my new results were stunning. i am gonna give them to you so you can check what TrueSim has achieved.
Where i have put the R7, it means the reale top 7 europe leagues average number.
**** Direct freekick goals R7: 0,75 goals/team/season VS TrueSim_1.1: 0,77goals/team/season
**** Cross Success Rate R7: 16%-23% with an average 19% VS TrueSim_1.1: 18,8%
**** Goal ratio from Headers Ration R7: 16% VS TrueSim_1.1: 15,9%
**** Successfull Pass Percentage R7: 78%-90% with average 84,5% VS TrueSim_1.1: 85%
**** Shots per game R7 11-16,5 with an 13,5 average VS TrueSim_1.1: 10,5-17,2 with an 13,8 average
**** Shot conversion rate R7: 9%-15,6% VS TrueSim_1.1: 8,8%-15,4%
**** Fouls per game per team 11-13 R7 Vs TrueSim_1.1: 12,3 average
**** Yellow Cards per team per game average R7: 2,1-2,4 VS TrueSim_1.1: 1,8-2,6
**** Goals from corners R7: 5,5-7,5: VS TrueSim_1.1: 5,9 average (I could boost this a bit but because i know how corners could be exploited, i left it there)
**** Indirect Freek kicks goals R7: 3,5-4,5 per team per season VS TrueSim_1.1: 4,2 average.
Update 1.4:
Sacrificed a bit of delivered stats in order to Improve the match flow and motion feeling.
Now the game motion feels more natural and smooth, more attacking & beautifull football is being played.
To keep scoring realism up as much as i could, because FM24 overperforms in scoring, i had to pump the GKs just a bit.
Update 1.5
Fixed some visual bug headings occured after a cross has happened.
Fixed some really powerfull driven crosses that were impossible to defend.
Lower the effectiveness of gegenpress a bit, now being punished more if used when you shouldn't.
A txt with installation instructions is also placed inside the zip file.
For the optimal visual experience, it is recommended to play at +1 speed.
The version 1.5 will be the last till late May since i think only a lot of feedback can make me change things.
![FM26 2025-26 Real Fixture & Results [23-4-2026]](datas/users/realresult_thumb_25_26_fm26_257759.png)
Discussion: True Sim 1.5 (Final) for FM24
45 comments have been posted so far.
Thank you so much for your hard work and dedication in trying to create the most realistic match engine.
I have played these games for 40 years now. I find your mod to be so good on the eye in terms of modelling and simulating real life football.
I really look forward to trying version 1.3 if you have any details on when it will be released?
I appreciate your input regarding my opinion about the simatch. I assume you might be one of the people who work on it. However, let me clarify what I meant.
I have played five seasons in the Premier League with Man Utd using that mod, and I watched all the matches in 3D. I have not simulated any matches or let the assistant handle them. In those five seasons, I have not seen a single lob (as in vanilla), and I have not seen a nicely placed shot. No matter the PPM of the player, they always smash the ball into the back of the net. Also, the team tends to play the same type of football regardless of the tactic.
Overall, the match experience feels boring and not visually different from vanilla. For example, in the Korean mod, players move and turn more gracefully, especially the more gifted players.
Back to the MatchLab mod: in one season I played three finals (UCL, Carabao Cup, and FA Cup) and lost all of them 1–0, even though I was the dominating team with high xG. That is not RNG that feels broken, especially since I obliterated everyone in the league and only had one loss in the whole season. As I said, I played and watched all the matches, changing tactics on the go. Also I have never seen turnarounds something like 2-0 half time 2-3 at the end no matter the attacking power
Since I do not save scum, I continued into the next season, where I lost two more finals 1–0 (the Carabao Cup and the UCL again).
The MatchLab mod definitely feels broken to me, because this cannot just be RNG the pattern is too obvious. I feel that version 1.4 does not bring anything visually new to the match experience, aside from some strange ball speed movement. In the most important matches, the repetitive script kicks in where everything supposed to be decided by slim margins, there is no football spectacles something that happens in vanilla version
That being said, I am still grateful for the work the team put into the mod and the effort to improve the match experience though my opinion is that they might have improved something but ruined something else as the mod makes the game predictable especially in important matches like the finals where is super scripted.
Unfortunately using AI for this process can only lead to major mistakes and lack of understanding. Reading your post is clear that the AI misled you and that's just the natural consequence of using AI because AI is not trained to understand the inner workings of making a match engine mod. The descriptions to not match what each parameters does and the way you changed the parameters, unfortunately, are more copy pasting previous work than anything truly original.
Besides the fact that when downloading the zip file, what it brings is a folder which is not what the game reads and the game cannot read a folder.
Another point is that your testing is deficient. You only use the top 7 leagues; the FM Match Lab team used all the leagues in the game, all the divisions of each league, and all the players in the database to avoid the results being only focused on the better teams. Lower teams and lower divisions also matter. The way your testing is done is only focused on the top teams and top leagues which means any change you make will mean, ergo, the match engine is focused on the top leagues and top teams, and it is not an experience that it is replicated across the whole world in FM.
You also mentioned that you did multiple seasons. For context, the FM Match Lab run tests between 50 and 100 seasons which means smoothing out the variables. If not, the RNG takes hold and it messes up with the testing. It seems you also do not have controlled groups which, in itself, doesn't bode well of your testing.
A match engine mod that only is focused on stats is bound to fail and that's the biggest lesson here. Adding to that, the AI will lead you the wrong path because the AI is not properly trained. The FM Match Lab team did the mods without any AI help because, honestly, is horrendous. It can help you to gather data, but it doesn't help you to interpret the data and that shows in your post.
I commend your effort but unless you change the methodology approach then you cannot say this is a True Sim. Finally. you are missing the key ingredient here. Changing one file is not enough for a "true sim", for that you need to change much more. There's a reason that FM Match Lab had a team of almost 25 people working on the mod, and that's because it needs manpower, and much more to make something worthy of being considered a "true sim".
Regarding winning 7-0 to Juve using a Cautious mentality can be considered a good thing since winning with a Cautious mentality is not easy. The FM Match Lab mod opens up the tactical possibilities of the game. I think you are seeing this in the wrong way. Plus not seeing long range strikes or players lobbing the keeper, that's just bad luck. I can send several goals where screamers are scored and players lob the keeper.
The Korean mod is a nice mod for those who prefer a defensive match engine. At its core, the Korean mod doesn't change one single offensive parameter while changing defensive parameters in a very humble way. Compared with the default match engine it is a good mod, but it lacks the knowledge accumulated in the meantime.
At the end of the day, any player will choose a match engine that suits his needs. And that's fine, but the analysis should be done properly and based on huge amounts of data and not single games.
Firstly, I want to say thank you for your work. I would like to share my findings so far.
In the past I tried the latest Matchlab Simatch and I found it quite poor. Around 90% of the finals ended 1–0 in my long save. This happened multiple times, I played three finals in a single season and lost all of them 1–0. At first I thought it was just a coincidence, but it clearly wasn’t as it repeated in the following seasons
It also feels very unbalanced. For example, winning 7–0 to Juventus in the Champions League while using a Cautious mentality is far from what I would consider a good simulator. Additionally, players seem unable to perform certain actions regardless of their attributes they never lob the goalkeeper or score long-range screamers is like attributes don't matter
Because of that, I moved to the Korean Simatch, which is very good and might actually be the best one overall, although you will occasionally see unusual results like 10–0.
Regarding your Simatch, in my experience it currently feels unbalanced. In my very first match a defender scored a hat-trick, which never happened when I used the Korean Simatch (that version seemed to improve defensive behavior). I also had a match where I was leading 7–0 at half-time against Tottenham and finished the game with over 6 xG, 34 shots, 15 on target, and 6 clear-cut chances. That seems quite broken, especially with a simple 4-2-3-1 setup. And also lost the Carabao Cup final with 1-0
I also noticed that the statistics like XG, clearcut chances etc appear more broken than in the Korean version.
In my opinion, the Korean version might be the best base to work from and improve further. It had stronger defensive behavior, far fewer corner goals, and produced some very beautiful goals. However, sometimes the ball speed felt a bit strange. On the positive side, player movement and turning animations looked very good and that would make gifted players look apart in the match engine. I guess the engine has its limitations as I would like to see more killer balls especially when a counter attacking tactic is used.
@Leeds-1919 exactly.
As i wrote, you need to backup and delete simatch.fmf file in installation directory and just replace as it is with my simatch FOLDER as it is without renaming it. I wish you all test it and give some feedback in what you didn't like so i can improve it. Soon version 1.1 will be up with minor cross + head improvements towards realism.
remember to back up the original
Can you provide instructions? Which files go to which folders?