Because I've checked our DB and here's what I got:
Red General 4231 v1.0
424 CF Cabuloso
433 Tsukyomi I
According to your table, "Red General 4231 v1.0" tactic is ranked "8th" and "424 CF Cabuloso" tactic is ranked "1st" when it come to 4-3-3 formation but according to our numbers the difference is about 1.5 points which could be just "statistical noise".
If you ask me then I think it makes sense to "rank" tactics only when there's a meaningful difference between them, I'd say it should be at least 10-15 points difference but when the difference is just 1-2 points then it can be easily "statistical noise".
Also, according to your table "433 Tsukyomi I" tactic is ranked "1st" when it comes to 5-3-2 formation but that's not true. Expand
Thanks for you feedback, as always I appreciate it. I accept some of your commentary. There was a minor error in my calculation Which I have now fixed. I have also moved the scale back to the 38 games that you guys use. I have checked the calculation and I still calculate Tsukyomi as ranked first. As I said, I only used the first 6 tests, so is your data showing the average from all tests where you have tested 4000 matches or only the first 2400 matches?
I accept the statistical noise for less than 2 points. I was just trying to look at how I can use a combination of tactics through the season to my advantage.
So here are my results from the first 3 games of the new season. We are expected to be mid table.
Genie Scout Team Ratings and Odds for the season.
Game 1 we are playing against a 433 side who are slight favourites. I think this formation is slightly weaker in defence so I went with Deformation as it is most attacking of the formations that work against 433. We led 2 Nil and hit the woodwork 3 times so we probably should have won by more.
Game 2 up against a 442, with us slight favourites. Tossed up between 424 and 4231. Ended up with the 424 CF Cabuloso because of its defensive advantage. We went down a goal but had a huge win.
Game 3 against 442 Diamond, with us again very slight favourites. I went with the 433 Tsukyomi I. We totally outplayed them but led 2 to 1 when they scored against the run of play. Scored another 2 quick goals to lead 4 to 1 and then swapped to 424 to shut the game down and scored twice late to smash them.
Does this approach work. Still too early to say but seems promising.
dzek said: I think this is a good test and so we will know which formation performs better against the 3 formations used by the FM-Arena tactic testing league. This would be even better if we tested it on the other formations that the game has too.
For the AI formations that haven't been tested by FM Arena I am using the reverse approach eg if I play against a 442 diamond I look at the worst result for the only HOF 442 diamond and go for the worst performance against it by the AI formation. In this case either 523 or 433. So I will play with the 433 Tsukyomi I as my squad is better suited to this formation than a 523.
If I was playing against a 4222 or 424 I would use a 4231.
I hope this makes sense.
What would be more interesting would be a home and away split.
I have had a quick look at the top FM Arena HOF tactics against the 3 formations FM Arena are testing against, 433, 4231 and 523.
I have only looked at standard type tactics with only one asymmetric tactic being the most successful, Deformation II. Because the testing uses 2 of each of 433 and 4231, and only one 523, I have averaged the scores for all tactics to make the 523 comparable. I also only looked at the first 6 tests (ie 2400 matches) for each HOF tactic to ensure they were also all comparable.
The selected tactics were:
4231 RedGeneral v1.0 424 CF Cabuloso 424 BBW v4 424 Deformation II 4222 Jigen Box I 41212 Ol Reliable 343 Italian KO 433 Tsukyomi I 433 Inverted Samba Shield X3 442 BOBW x Scunny v1
I have listed the average points per season and then the ranking against each AI formation.
I have no idea if this will help anyone but I have just started using it.
Please let me know your thoughts and any results if you use it
This is very much based on @A Smile Icebreaker CF II tactic. I have dropped the wing line back to a 442 instead of 424 and used inverted wingers.
Tested with two teams in 3 leagues. All did well, but as I play lower leagues I am just showing VNL. Bromley and Kidderminister both performed well above expectations with some great wins through the season. Being knocked out by Man Utd in the FA Cup was also a massive financial boost for the club.
Given the worst tactic I could come up with had all the players on the outside, I thought I would try all players on the inside, as much as I could. After several attempts I got to one that seems to work. I would probably drop the mentality back to Balanced for away games though or change the Vol Su to DM De as it was a bit weaker away from home.
Something a bit different to all the 4231 tactics around. I wanted to limit Defending players to 3 and have 4 on Attack to try and outscore opponents.
I really don't think I can help. Make sure you have the latest versions. Zealand and Priisek both have videos on it and there is a support section on the FM Base site.
fikaaydin said: no i am doing true but always got same result chinese regen and some regens dont show. when i select all newgans it's okay but when i opened the config file some id dont show because there is a problem for mapping some nations and their ids. i cannot see their ids in config file. when i do manuelly i can see them but i have to adress their id's to african not asian or chineese. that's ridicoulus. Expand
The only other thing I can think of is that all of the numbers in your screenshots above are only 8 digits. Most if not all regens should start with "200" and be 10 digits. Are you sure you used the "is newgen search filter" when you were finding newgens for your text file.
Also the mapping message says it has no faces left for that ethnicity. This seems to indicate that there are not enough faces in your face pack which might happen if you are selecting too many players in your search ie not limiting the selection to only newgens.
If you cant figure it out you should follow the process on the Newgan Support page, copied in the post linked.
fikaaydin said: i tried -UltraRealistic NewGan Face CutOuts by RnR & FM.Z- and also tried FMNEWGANv2 + Update 1 but i always got same result. for 2 days i got crazy. I coulnt solve this problem. Expand
A few questions:
Do you still have both these face packs in the graphics directory?
Which pack are you currently using?
Can you upload the screenshot of the sub directories of the face pack you are currently using?
It does look like whatever face pack file you are using isn't compatible with Newgan, might be incorrect nomenclature.
fikaaydin said: some nations dont change like china but japan change i tried ten times may be. all the same chinese player didnt change. but japanese alwayws changed. i didnt understand whats wrong. Expand
I don't know if I can help you but will try. If you can upload two screenshots as follows:
Go to C:\Users\"Your Windows user name"\AppData\Local\Programs\NewGAN-Manager\app\newganmanager
and open the newgan.log file. Scroll to the very end of the file and take a screenshot.
Secondly go to the config.xml file in your newgen facepack directory. Open the file and again scroll to the very bottom and take a screenshot.
fikaaydin said: hello guys i did what you say on the comments already. But some newgan faces change but some doesnt change for example asian ones changes chinese dont change. how i can solve this problems Expand
It could be filters in your search when you ran the text file. Hard to tell from the info provided.
You can try the following:
Make sure you only have one face pack in the graphics directory because if there are conflicts it will revert back to the original FM face.
Turn on player IDs. Check the ID in the config file to see if it is there is a line with the ID and if so note the picture reference. Look the picture reference up in the face pack. If either is missing this will be your problem
crazyfmguy93 said: thanks for answering. you mean if Raphinha is 18/20 for AMR positon and Damsgaard is 20/20 for AMR positon, I should choose Damsgaard? Expand
Yes, actual current positional ability Damsgaard stays at 83.50% and Raphinha drops to 80.35%.
@Yarema is right. If you play and train Raphinha at AMR he may end up slightly better in the position. This doesnt always happen though and it may depend on age and Potential Ability
crazyfmguy93 said: Hello, MDW22 is fantastic for me in FM22. are you using MDW22 for only positional rating? what do you think about role ratings?
Im using ZaZ - Blue 4.0. For AMR positon I have Damsgaard and Raphinha.
Positional Rating Raphinha 84 Damsgaard 83,5 Role Rating For Winger Support Role Raphinha 88 Damsgaard 91
Which player do you choose? Expand
Role rating doesnt really matter. Positional rating is important but it also matters if the are Natural for the position or 20/20 for the position in Genie Scout rating.
crazyfmguy93 said: One more there are 3 different rating file YKYKYKY MDW22 Zaz
Which one is better for Zaz Blue? Expand
ykykyky was the one developed using machine learning for ZaZ Blue tactic. I still used MDW22 and still do.
Chriswin4 said: That's a very good point. However, I do wonder if there is a baseline we can reach where we know a player is going to succeed to a certain degree in the match engine. Expand
I play lower leagues. At the start of each season and at the mid season mid Point i look at clubs ratings on genie scout (i either use MDW or ykykyky). I record best, worst and the mid point. I do this for league up to and including EPL.
I then rate my players against the mid point of each league and get rid of any that won't make the mid point of the League we expect to be in either this season, at the start, or league above at mid season if we look like getting promoted. My spreadsheet then rates the players current and future ratings as a league value.
I have a calculation for potential rating that applies a likely improvement rate based on age and potential positional improvement.
It works for me. No idea how you would code that with python.
Are you sure your numbers are correct?
Because I've checked our DB and here's what I got:
Red General 4231 v1.0
424 CF Cabuloso
433 Tsukyomi I
According to your table, "Red General 4231 v1.0" tactic is ranked "8th" and "424 CF Cabuloso" tactic is ranked "1st" when it come to 4-3-3 formation but according to our numbers the difference is about 1.5 points which could be just "statistical noise".
If you ask me then I think it makes sense to "rank" tactics only when there's a meaningful difference between them, I'd say it should be at least 10-15 points difference but when the difference is just 1-2 points then it can be easily "statistical noise".
Also, according to your table "433 Tsukyomi I" tactic is ranked "1st" when it comes to 5-3-2 formation but that's not true.
Thanks for you feedback, as always I appreciate it. I accept some of your commentary. There was a minor error in my calculation Which I have now fixed. I have also moved the scale back to the 38 games that you guys use. I have checked the calculation and I still calculate Tsukyomi as ranked first. As I said, I only used the first 6 tests, so is your data showing the average from all tests where you have tested 4000 matches or only the first 2400 matches?
I accept the statistical noise for less than 2 points. I was just trying to look at how I can use a combination of tactics through the season to my advantage.
do you use MDW22 or ykykyk files in genie scout 23? or are there any rating file fit better for 23?
Both are good. I have been using my MDW22 and still am
Genie Scout Team Ratings and Odds for the season.
Game 1 we are playing against a 433 side who are slight favourites. I think this formation is slightly weaker in defence so I went with Deformation as it is most attacking of the formations that work against 433. We led 2 Nil and hit the woodwork 3 times so we probably should have won by more.
Game 2 up against a 442, with us slight favourites. Tossed up between 424 and 4231. Ended up with the 424 CF Cabuloso because of its defensive advantage. We went down a goal but had a huge win.
Game 3 against 442 Diamond, with us again very slight favourites. I went with the 433 Tsukyomi I. We totally outplayed them but led 2 to 1 when they scored against the run of play. Scored another 2 quick goals to lead 4 to 1 and then swapped to 424 to shut the game down and scored twice late to smash them.
Does this approach work. Still too early to say but seems promising.
Good idea @Mark!
For the AI formations that haven't been tested by FM Arena I am using the reverse approach eg if I play against a 442 diamond I look at the worst result for the only HOF 442 diamond and go for the worst performance against it by the AI formation. In this case either 523 or 433. So I will play with the 433 Tsukyomi I as my squad is better suited to this formation than a 523.
If I was playing against a 4222 or 424 I would use a 4231.
I hope this makes sense.
What would be more interesting would be a home and away split.
That is correct
I have only looked at standard type tactics with only one asymmetric tactic being the most successful, Deformation II. Because the testing uses 2 of each of 433 and 4231, and only one 523, I have averaged the scores for all tactics to make the 523 comparable. I also only looked at the first 6 tests (ie 2400 matches) for each HOF tactic to ensure they were also all comparable.
The selected tactics were:
4231 RedGeneral v1.0
424 CF Cabuloso
424 BBW v4
424 Deformation II
4222 Jigen Box I
41212 Ol Reliable
343 Italian KO
433 Tsukyomi I
433 Inverted Samba Shield X3
442 BOBW x Scunny v1
I have listed the average points per season and then the ranking against each AI formation.
I have no idea if this will help anyone but I have just started using it.
Please let me know your thoughts and any results if you use it
Tested with two teams in 3 leagues. All did well, but as I play lower leagues I am just showing VNL. Bromley and Kidderminister both performed well above expectations with some great wins through the season. Being knocked out by Man Utd in the FA Cup was also a massive financial boost for the club.
What does good look like in FM 24?
FM Stag Stats are now in a few Skins including SAS24.
I hope this helps someone. I find it quite interesting
Something a bit different to all the 4231 tactics around. I wanted to limit Defending players to 3 and have 4 on Attack to try and outscore opponents.
Results are below.
Try this
Newgan Priisek Tutorial
The only other thing I can think of is that all of the numbers in your screenshots above are only 8 digits. Most if not all regens should start with "200" and be 10 digits. Are you sure you used the "is newgen search filter" when you were finding newgens for your text file.
Also the mapping message says it has no faces left for that ethnicity. This seems to indicate that there are not enough faces in your face pack which might happen if you are selecting too many players in your search ie not limiting the selection to only newgens.
If you cant figure it out you should follow the process on the Newgan Support page, copied in the post linked.
but i always got same result. for 2 days i got crazy. I coulnt solve this problem.
A few questions:
Do you still have both these face packs in the graphics directory?
Which pack are you currently using?
Can you upload the screenshot of the sub directories of the face pack you are currently using?
It does look like whatever face pack file you are using isn't compatible with Newgan, might be incorrect nomenclature.
What newgen face pack are you using?
I don't know if I can help you but will try. If you can upload two screenshots as follows:
Go to C:\Users\"Your Windows user name"\AppData\Local\Programs\NewGAN-Manager\app\newganmanager
and open the newgan.log file. Scroll to the very end of the file and take a screenshot.
Secondly go to the config.xml file in your newgen facepack directory. Open the file and again scroll to the very bottom and take a screenshot.
Now upload these two screenshots.
It could be filters in your search when you ran the text file. Hard to tell from the info provided.
You can try the following:
Make sure you only have one face pack in the graphics directory because if there are conflicts it will revert back to the original FM face.
Turn on player IDs. Check the ID in the config file to see if it is there is a line with the ID and if so note the picture reference. Look the picture reference up in the face pack. If either is missing this will be your problem
Hope this helps
Yes, actual current positional ability Damsgaard stays at 83.50% and Raphinha drops to 80.35%.
@Yarema is right. If you play and train Raphinha at AMR he may end up slightly better in the position. This doesnt always happen though and it may depend on age and Potential Ability
Im using ZaZ - Blue 4.0. For AMR positon I have Damsgaard and Raphinha.
Positional Rating
Raphinha 84
Damsgaard 83,5
Role Rating For Winger Support Role
Raphinha 88
Damsgaard 91
Which player do you choose?
Role rating doesnt really matter. Positional rating is important but it also matters if the are Natural for the position or 20/20 for the position in Genie Scout rating.
crazyfmguy93 said: One more there are 3 different rating file
YKYKYKY
MDW22
Zaz
Which one is better for Zaz Blue?
ykykyky was the one developed using machine learning for ZaZ Blue tactic. I still used MDW22 and still do.
All the below should be in your original post so you will need to edit it:
You must provide a screenshot of the tactic/formation,
You must provide a screenshot of the league table. At least one full season must be played ( FOR/AG/G.D. stats must be visible ),
You must provide a screenshot of the team selection ( Goals/Assists/Ratings stats must be visible ),
You must provide a screenshot of the fixtures ( from the start to the end ).
I play lower leagues. At the start of each season and at the mid season mid Point i look at clubs ratings on genie scout (i either use MDW or ykykyky). I record best, worst and the mid point. I do this for league up to and including EPL.
I then rate my players against the mid point of each league and get rid of any that won't make the mid point of the League we expect to be in either this season, at the start, or league above at mid season if we look like getting promoted. My spreadsheet then rates the players current and future ratings as a league value.
I have a calculation for potential rating that applies a likely improvement rate based on age and potential positional improvement.
It works for me. No idea how you would code that with python.