letsgo9 said: They released a small update, 200mb one, anyone knows the content? Expand Minor Update 24.2.1 Changelist (Steam, Epic & Microsoft)
Licensing / Database * - Removed some duplicated player photos in Japan. - Changes to the database in Germany.
Match Engine - Fixed a problem where all players appeared to look the same in game.
Bug Fixing - Various stability fixes. - Korean currency now shown in correct short/long format. - Restored the option to use custom manager profile picture. - Updated NewGen UID’s in backend. - Post match analysis report only showing one key finding. - User remains in charge of B teams after resigning.
Editor - Fixed issue with custom scores not appearing in game.
*Licensing and Database changes require starting a new save game for them to be applied.
Chriswin4 said: That is a very helpful suggestion! I’m looking for a way to see how attributes influence match engine performances. Like looking at the chalkboard stats in game, the best performing players rarely have the attributes I believe to be necessary according to FM Arena testing. Expand To be able to say for each attribute which is better etc you need to use machine learning with AI that tries all possible ways and all combinations of attributes. It might take you as long as it takes to get the new FM out and more.
Chriswin4 said: Having seen the explosion of Python brilliance into the game this year, coupled with the excellent efforts made by FM Arena in finding what attributes most impact team performance, I'm wondering if as a group we can find a standard for player recruitment.
I know we've had the YKYKYK Balanced filters for previous Genie Scouts, but I'm wondering if we can find a different way. Obviously the most important two attributes are pace and acceleration, irrespective of possession.
However, based on the python model of attributes weighted 5, 3 and 1, I'm wondering if there are filters we can come up with, because even with this system, a lot of the players thrown up are still unattainables like Haaland.
Yet, if you mess with the filters too much, i.e. lower things, you get too few players who don't look great.
I wonder if there is a way we could determine and identify what things equate to match engine performance in individual positions.
I know this could vary from person to person, like I could value passing in my DM, whereas someone else could just want a Lee Cattermole style destroyer with no real ability.
A lot of this is a bit of a ramble, but I'm looking mostly for a discussion, see other people's opinions and see if we can come up with something. Expand Another way to decide on the selection filters for the players you buy is to look at the average attributes for the league you play in and create your own filters based on them. Keep in mind though that these will change from time to time so you will need to adjust them accordingly every 6 months I believe.
CBP87 said: Aye I get that, I wanted to use deformation as my example but I couldn't remember what it scored after 1200 matches, think it 65 but couldn't remember. But my point is that tactic although it has been tested for 4000 matches, its potentially dropped 3 points and now probably doesn't get the attention it should due to other tactics around it scoring similar but over less matches.
If we put the 1200 matches into perspective then we are potentially looking at 17 seasons (if we were looking at a 70 game season) and by the point people have hopefully built an OP team so the swing in RNG should be reduced but I just feel the more and more matches tactics are tested for then the OP tactics are sort of lost within the tactics that haven't been tested as much. Expand I understand what you're saying and I agree. My honest opinion is:
1. Reduce the tactics in the Hall of Fame - increase the range of scores that get in there. 2. All tactics in the Hall of Fame should be tested in 4000 games and if a tactic falls below the second best in the same formation, the second best tactic should be tested in 4000 games, etc. until a "winner" is found. 3. 2400 matches for a tactic to be done in some cases - when the creator wants it (and the admins agree of course) to see the difference in some changes he made to his tactics.
letsgo9 said: The revolution is alrdy made, now is comming a new test wave with thousands of underlaps, haha Expand After so many of my own tests and others I think that the Overlaps/Underlaps and Focus Play instructions have more to do with what formation you are using so I don't recommend that people try it on all formations unless they know exactly what the strength of the formation they are using is.
That’s my opinion.
CBP87 said: @Zippo@Droid would it possible to be add a filter to the tables for 1200 game, 2400 games and 4000 games please? I'm only asking because I feel that tactics which have been tested for more games are somewhat penalised in a way.
I know it might sound petty but I think it will give a better idea of which tactic is better over 1200 matches, I mean even this tactic which is a belter dropped 1 point when tested for 2400 matches and is now showing as being on par with others when in fact over 1200 it scored higher.
Sorry if I haven't explained it properly but in summary would it possibly to see results in the table for 1200, 2400 and 4000 matches please Expand I'll try to answer briefly but surely the admins can answer it better.
Tactics from a certain point onwards, for example say score 57, are all good in general.
Now regarding tactics that score 62 in 2400 games and others that score 62 again in 1200 games I think a decision should be made on how many games all tactics will be tested. But actually that is not the real problem. The problem is that most authors make zero to minimal changes to every tactic they bring up and that causes this mess on the table.
Peter78 said: Enjoying this so far, 2nd season I am moving from a 4-3-3 and this 4-2-3-1 plays nice football, is there any room for experimenting with the CDM roles? Expand I'm glad to hear that
You could do the following: Aggressive approach: Change one of the two to VOL(Su).
Defensive approach: Change one of the two to DLP(Su). He'll help more with possession of the ball with more passes but may make a few turnovers. Also he holds his position better than the DM role defensively.
Filer974 said: Thank you for your help. You're probably right, I'll check it out Expand This is just an example. You have to build it to work as it should.
nice to find a tactic with less intensity, do you tried with balanced mentality ? Expand Hi,
No I didn't and I don't know if there will be a change in terms of the score here at FM-Arena. You can always change it to Balanced when you want to maintain a lead or relax your players
Filer974 said: 52 great. It's the best 3241 so far. It just goes to show that you have to persevere. I'm going to keep thinking about how to keep improving this tactic... If you have any ideas? Expand To be honest if you want to achieve this shape when your team has the ball you can play like this but be aware that what you see there, it's the out of possession shape and its very leaky in defence from wider areas of the pitch because you don't use normal FBs.
So i think it's better to use a 433 out of possession(what you on tactics panel) and transform it into 3241 in possession(with specific roles).
Minor Update 24.2.1 Changelist (Steam, Epic & Microsoft)
Licensing / Database *
- Removed some duplicated player photos in Japan.
- Changes to the database in Germany.
Match Engine
- Fixed a problem where all players appeared to look the same in game.
Bug Fixing
- Various stability fixes.
- Korean currency now shown in correct short/long format.
- Restored the option to use custom manager profile picture.
- Updated NewGen UID’s in backend.
- Post match analysis report only showing one key finding.
- User remains in charge of B teams after resigning.
Editor
- Fixed issue with custom scores not appearing in game.
*Licensing and Database changes require starting a new save game for them to be applied.
To be able to say for each attribute which is better etc you need to use machine learning with AI that tries all possible ways and all combinations of attributes. It might take you as long as it takes to get the new FM out and more.
So good luck
Added: Shorter Passing, Focus Play Through The Middle
I know we've had the YKYKYK Balanced filters for previous Genie Scouts, but I'm wondering if we can find a different way. Obviously the most important two attributes are pace and acceleration, irrespective of possession.
However, based on the python model of attributes weighted 5, 3 and 1, I'm wondering if there are filters we can come up with, because even with this system, a lot of the players thrown up are still unattainables like Haaland.
Yet, if you mess with the filters too much, i.e. lower things, you get too few players who don't look great.
I wonder if there is a way we could determine and identify what things equate to match engine performance in individual positions.
I know this could vary from person to person, like I could value passing in my DM, whereas someone else could just want a Lee Cattermole style destroyer with no real ability.
A lot of this is a bit of a ramble, but I'm looking mostly for a discussion, see other people's opinions and see if we can come up with something.
Another way to decide on the selection filters for the players you buy is to look at the average attributes for the league you play in and create your own filters based on them. Keep in mind though that these will change from time to time so you will need to adjust them accordingly every 6 months I believe.
Go to Squad Planner > Report > Comparison
Hope this helped
TIs
Removed: Shorter Passing, Pass Into Space
If we put the 1200 matches into perspective then we are potentially looking at 17 seasons (if we were looking at a 70 game season) and by the point people have hopefully built an OP team so the swing in RNG should be reduced but I just feel the more and more matches tactics are tested for then the OP tactics are sort of lost within the tactics that haven't been tested as much.
I understand what you're saying and I agree. My honest opinion is:
1. Reduce the tactics in the Hall of Fame - increase the range of scores that get in there.
2. All tactics in the Hall of Fame should be tested in 4000 games and if a tactic falls below the second best in the same formation, the second best tactic should be tested in 4000 games, etc. until a "winner" is found.
3. 2400 matches for a tactic to be done in some cases - when the creator wants it (and the admins agree of course) to see the difference in some changes he made to his tactics.
After so many of my own tests and others I think that the Overlaps/Underlaps and Focus Play instructions have more to do with what formation you are using so I don't recommend that people try it on all formations unless they know exactly what the strength of the formation they are using is.
That’s my opinion.
CBP87 said: @Zippo @Droid would it possible to be add a filter to the tables for 1200 game, 2400 games and 4000 games please? I'm only asking because I feel that tactics which have been tested for more games are somewhat penalised in a way.
e.g This tactic https://fm-arena.com/thread/7622-katana-4231-106pts/ achieved 63 points over 1200 matches, but dropped 2 points after 2400 matches, however this tactic https://fm-arena.com/thread/7723-4231-hegemony-through-dzek-tweak-v2/ has scored 62 over 1200 and is placed higher in the table.
I know it might sound petty but I think it will give a better idea of which tactic is better over 1200 matches, I mean even this tactic which is a belter dropped 1 point when tested for 2400 matches and is now showing as being on par with others when in fact over 1200 it scored higher.
Sorry if I haven't explained it properly but in summary would it possibly to see results in the table for 1200, 2400 and 4000 matches please
I'll try to answer briefly but surely the admins can answer it better.
Tactics from a certain point onwards, for example say score 57, are all good in general.
Now regarding tactics that score 62 in 2400 games and others that score 62 again in 1200 games I think a decision should be made on how many games all tactics will be tested. But actually that is not the real problem. The problem is that most authors make zero to minimal changes to every tactic they bring up and that causes this mess on the table.
An updated version of Scalifornia V3
FB --> WB
IW --> IF
Tweaks on some PIs.
An updated version of Scalifornia V2
DLP --> RGA
WB --> FB
Tweaks on some PIs.
I'm glad to hear that
You could do the following:
Aggressive approach:
Change one of the two to VOL(Su).
Defensive approach:
Change one of the two to DLP(Su). He'll help more with possession of the ball with more passes but may make a few turnovers. Also he holds his position better than the DM role defensively.
So far that's all.
4-2-3-1 (Out of Possession)
3-2-4-1 (In Possession)
No instructions
4-3-3 (Out of Possession)
3-2-4-1 (In Possession)
This is just an example. You have to build it to work as it should.
nice to find a tactic with less intensity, do you tried with balanced mentality ?
Hi,
No I didn't and I don't know if there will be a change in terms of the score here at FM-Arena. You can always change it to Balanced when you want to maintain a lead or relax your players
To be honest if you want to achieve this shape when your team has the ball you can play like this but be aware that what you see there, it's the out of possession shape and its very leaky in defence from wider areas of the pitch because you don't use normal FBs.
So i think it's better to use a 433 out of possession(what you on tactics panel) and transform it into 3241 in possession(with specific roles).
gohan said: boss pls give me ur PI
Check the OP.