Zolov said: are we able to get the league file to trial and error individually? Expand At the moment we're not planning to share it from the start but we might change our mind on that in the future.
Here are some highlights of FM-Arena tactic testing league:
- All players in the FM-Arena testing league have their morale and conditions set and "frozen" at the highest possible level, it ensures that the morale and physical conditions don't change and stay the same throughout a testing process, also, injuries are being removed if they occur. These adjustments greatly reducing the random elements of the game.
- Tactics regardless their formations/roles/duties are being tested with equally good players which ensures that the results aren't affected by the difference in the players' quality.
- All teams in the testing league have the "Team Cohesion" and "Tactic Familiarity" maximized.
- All players have their "Player Traits" removed, which ensures that they play according with the tactical instructions and not according with their "Player Traits".
- The variety of the AI tactics in the league ensures that we test tactics against the most optimal set of AI tactics.
- In our testing league the AI managers can't get sacked during testing due to poor results so they stay in charge of their teams until the end of testing. Keeping the testing environment unchanged till the end greatly reduces the game randomness.
The above are just some features of FM-Arena tactic testing league and there are much more.
If you want to check the league settings you can do that by downloading a game save:
I want to let you know that we are going to start a new tactic testing for the 21.4 patch.
As you might notice there are just few insignificant ME changes in the 21.4 patch and one of them changes the way long throw-ins work, but at the moment almost all tactics use long throw-ins, so we've retested few tactics under the 21.4 patch and found 21.3 and 21.4 results were slight different, even the difference was a quite small we still decided to start a new tactic testing for the 21.4 patch.
Please note, that we'll be testing a tactic only if we can be sure that this tactic is approved for the 21.4 patch but the author.
You might noticed there's a slight difference in the testing results between the 21.2.2 and 21.3.0 patches.
I want to inform you that the difference is not due to the ME changes in the patch 21.3.0 because there weren't any ME changes in the 21.3.0 patch. The difference is due to small changes(improvements) that we made to our testing database.
To be sure we retested some tactics using the old database and there wasn't any difference in the results between the 21.2.2 and 21.3.0 patches.
We've done some interesting internal testing and it looks like Acceleration and Pace attributes are among the most important player attributes in FM21:
Of course, other player attributes are important too but changing Acceleration and Pace attributes makes the most difference in our testing.
For example, let's take Phoenix v3.0 tactic if you look at the table then its points per match is "2.114" but if we decrease the Acceleration and Pace attributes of the Shadow Strikers and Wingers by "3" points from "16" to "13" and re-test the tactic then we'll get the following results:
As you can see the points per match have dropped from "2.114" to "1.657".
If we also decrease the Acceleration and Pace attributes of the Inverted Wing-Backs by "3" points from "16" to "13" and re-test the tactic once more then we'll get the following results:
As you can see the points per match have dropped further from "1.657" to "1.467".
So decreasing the Acceleration and Pace attributes across the positions just only by "3" points has turned a winning tactic into a loser, the points per match have dropped from "2.114" to "1.467" (about 31% decrease).
No doubt, such information is very important and we're debating how to better represent something like that in the future.
After each patch that includes ME changes we'll be internally re-testing few tactics for 912 matches and if we see there's a significant difference in the results then we'll make a new table for the new patch and if the difference is small then we'll keep the current table.
We've decided to get rid of "Standings" and use "Ratings" instead.
We found that "Standings" concept can be quite confusing, for example, there could be a big difference in standings between tactics but a very small difference in the results.
We find that the "Ratings" concept gives a better view on the difference between tactics.
Yes, that's correct.
Hi,
You can check the league configuration yourself, here's a game save ( the end of a test )
Download - FM-Arena Tactic Testing League Game Save
Zolov said: are we able to get the league file to trial and error individually?
At the moment we're not planning to share it from the start but we might change our mind on that in the future.
Here are some highlights of FM-Arena tactic testing league:
- All players in the FM-Arena testing league have their morale and conditions set and "frozen" at the highest possible level, it ensures that the morale and physical conditions don't change and stay the same throughout a testing process, also, injuries are being removed if they occur. These adjustments greatly reducing the random elements of the game.
- Tactics regardless their formations/roles/duties are being tested with equally good players which ensures that the results aren't affected by the difference in the players' quality.
- All teams in the testing league have the "Team Cohesion" and "Tactic Familiarity" maximized.
- All players have their "Player Traits" removed, which ensures that they play according with the tactical instructions and not according with their "Player Traits".
- The variety of the AI tactics in the league ensures that we test tactics against the most optimal set of AI tactics.
- In our testing league the AI managers can't get sacked during testing due to poor results so they stay in charge of their teams until the end of testing. Keeping the testing environment unchanged till the end greatly reduces the game randomness.
The above are just some features of FM-Arena tactic testing league and there are much more.
If you want to check the league settings you can do that by downloading a game save:
We've changed the way we represent the tests results.
We get rid of the ratings and translated all the results into "a typical 38 matches season".
We believe that way it gives a better gauge of the difference between the tactics.
I hope you like it.
Cheers.
https://fm-arena.com/table/15-patch-22-2-0/
It looks like that tactic was tested a moment ago - https://fm-arena.com/tactic/1639-fanecas/
Later today or tomorrow we'll start testing FM22 tactics.
Thanks for your support!
Cheers.
I want to let you know that we are going to start a new tactic testing for the 21.4 patch.
As you might notice there are just few insignificant ME changes in the 21.4 patch and one of them changes the way long throw-ins work, but at the moment almost all tactics use long throw-ins, so we've retested few tactics under the 21.4 patch and found 21.3 and 21.4 results were slight different, even the difference was a quite small we still decided to start a new tactic testing for the 21.4 patch.
Please note, that we'll be testing a tactic only if we can be sure that this tactic is approved for the 21.4 patch but the author.
Also, all tactics will be tested for 912 matches!
Thank you for your support.
You might noticed there's a slight difference in the testing results between the 21.2.2 and 21.3.0 patches.
I want to inform you that the difference is not due to the ME changes in the patch 21.3.0 because there weren't any ME changes in the 21.3.0 patch. The difference is due to small changes(improvements) that we made to our testing database.
To be sure we retested some tactics using the old database and there wasn't any difference in the results between the 21.2.2 and 21.3.0 patches.
I hope this helps.
Cheers.
We've done some interesting internal testing and it looks like Acceleration and Pace attributes are among the most important player attributes in FM21:
https://fm-arena.com/table/9-player-attributes-importance/
Of course, other player attributes are important too but changing Acceleration and Pace attributes makes the most difference in our testing.
For example, let's take Phoenix v3.0 tactic if you look at the table then its points per match is "2.114" but if we decrease the Acceleration and Pace attributes of the Shadow Strikers and Wingers by "3" points from "16" to "13" and re-test the tactic then we'll get the following results:
As you can see the points per match have dropped from "2.114" to "1.657".
If we also decrease the Acceleration and Pace attributes of the Inverted Wing-Backs by "3" points from "16" to "13" and re-test the tactic once more then we'll get the following results:
As you can see the points per match have dropped further from "1.657" to "1.467".
So decreasing the Acceleration and Pace attributes across the positions just only by "3" points has turned a winning tactic into a loser, the points per match have dropped from "2.114" to "1.467" (about 31% decrease).
No doubt, such information is very important and we're debating how to better represent something like that in the future.
thanks for offering your help but at the moment something like that isn't technically feasible
Hi,
We're going to re-test only some tactics from the top and haven't decided yet how many tactics it'll be
After each patch that includes ME changes we'll be internally re-testing few tactics for 912 matches and if we see there's a significant difference in the results then we'll make a new table for the new patch and if the difference is small then we'll keep the current table.
I just want to let you know that we've started tactic testing for FM21.
Our tactic testing system is even more accurate than before.
Stay tuned!
I've added an additional option to the advanced tactics search. Now you can filter tactics by name.
Enjoy.
Enjoy!
https://fm-arena.com/table/6-Patch-20-4-1/
I want to inform you that a new cool feature "Search tactics by Formation / Author" is coming soon!
Let's say you want to search for 4-4-2 tactics then just set it this way and here you go
or if you want to search for strikerless tactics then just exclude the STCL/STC/STCR positions and you'll get all strikerless tactics
or if you want to search for any tactic with 5 defenders then you can do that this way
so stay tuned... I hope you like it!
I don't think you are interested in knowing the actual formula because it looks geeky
The ratings just make it much easier to gauge the difference between tactics and it's as simple as that
We've decided to get rid of "Standings" and use "Ratings" instead.
We found that "Standings" concept can be quite confusing, for example, there could be a big difference in standings between tactics but a very small difference in the results.
We find that the "Ratings" concept gives a better view on the difference between tactics.
We hope you like it.
Can you test this one, please?
Thank you.
Tactic thread: https://community.sigames.com/topic/523253-chrissy-rosss-442-prem-win-with-bournemouth-1st-season-no-transfers/
Download: https://www.dropbox.com/s/k5ds4dimz5bnn8x/Chrissy%20Ross%27s%20442.fmf?dl=0
Hi,
it's been tested already https://fm-arena.com/tactic/276-ChrissyRosss442/