Could you by any chance test my setpieces? I already posted them somewhere in the forum, but I didn't upload them to the set-pieces section. I'm interested to see how it will compete. Expand
If you upload it to the Set Piece Sharing Section I will test it in the next few days.
RNG would still be at play here. This routine led until the last test and there was a bizarre low score from Team E. The worst of all testing. The only score outside the top 6 for this routine. When I looked into it they lost 14 players to injury for more than a month through the season. In all the other measures this routine performed well. If I take out the lowest and highest single performance I think it wins.
You are welcome to use whichever you choose from the results table. There are many ways to interpret these results.
Han106 said: Took the best set pieces from @Mark's list and I Frankenstein all of the best performers combined into one with credit labeled onto each set piece.
@Delicious's Attacking Corners Zimsmula's Defensive Third Defending Throw-ins and Attacking DFK @AideNGomiinaM's Attacking IFK @CSTG KANE's Defending Corners and Defending DFK Chou's Defending IFK
This would have to be tested to see if it has the best set of set pieces, but it's my best interpretation and attempt at it. Expand
I definitely recommend you to test this. In a season in which I scored 191 goals, I scored 58 corner goals with this routine. Kusi-Asare, with 20 jumping, scored 80 goals in the league that season (: Expand
Han106 said: I combined everything from Zim's that is not Attacking Corners and Chou's Attacking Corners for now. Expand
Start a new post on the Set Piece Sharing forum and I will add it to the schedule
Sorry for the delays. I decided I wanted to extend it to 4000 tests. I have now done 3 set piece routines and updated the original post. I removed standard deviation as I didn't think it offered much . I added a few other columns to try and make sense of the testing and updated the Legend at the bottom of the OP. I have also given my thoughts and the schedule for testing. I hope this helps someone. At the end of the process I might look to take the best from each of the SPs and build something for testing.
dzek said: Just to clarify something here. My test league uses the same player attributes as it is in FM-Arena tactic test league V1.0. I don't know if there are any changes in V2.0.
Other than that congratulations Mark for the initiative and I imagine that would help a lot of people.
Thank you! Expand
Only the small changes I mentioned on the test league forum
I am going to use @dzek testing database to look at a few of the recommended set pieces. I will be doing 4000 test matches with each set piece routine. The base tactic will be Katana 4231 v 1.3 which I think is close enough to the best tactic this year.
I will test all the set piece tactics in this forum if I am am able and update this OP with the results. Wish me luck.
So here goes.
The baseline: Katana v1.3 for 4000 tests.
Testing table Update:
Thoughts:
I have now come up with a rating and ranking system. I think that anything that is 10% above the baseline should be excellent and worth using, so there are a lot of excellent candidates for you to use. The stand out top 6 are Blue, Han106 1-24, Azure V3, Chicken Wing v1.8, Han 1-21 and Delicious 4 inside 34. I will be using the top ranked Blue Routine.
Previously I was approaching the results from 2 different ways.
The first was overall impact on results, as you cant tell how many goals have come as a result of the play after the set piece has occurred eg counter attacks, and also the throw ins. I am unsure how to find results from throw ins. So if you prefer this one then you should look at the points difference from the baseline tactic. Blue just leads Zim and Chciken Wing v1.8 from Han106 1-24 on this measure.
The second way to approach it was the overall set piece impact (SP +/-). There are 5 that standout on this measure. Han106 1-24 SP Best Hits seems to be the overall winner here from Han106 1-21 SP Best Hits, Delicious 4 inside 34 who is slightly in front of Chicken Wing v1.8 and CSTG KANE 4231VIFS. I think you could use any of these 5 or Blue, Azure v3 or Zim with confidence.
I would like to thank all the routine developers and sharers that gave me something to test.
Update:
I have added an overall score and rank.
Schedule:
I uploaded and tested Knap MC40, Chicken Wing v1.8 and Azure V3. There are now no more to test, Back to my Scunthorpe save.
Legend:
PPM Points per match GFPM Goals for per match GAPM Goals against per match GDPM Goal Difference per match Corners Goals from corners DFK Direct free kick goals IFK Indirect free kick goals CornersA Goals against from corners DFKA Direct free kick goals against IFK Indirect free kick goals against Pts Dif The PPM difference from the baseline tactic Corner Dif Difference between goals from corners and goals against from corners per match DFK Dif Difference between goals from DFK and goals against from DFK per match IFK Dif Difference between goals from IFK and goals against from IFK per match SP +/- total of the 3 Dif columns, so overall impact of attack and defend set pieces % SP For The percentage of goals scored from set pieces % SP Against The percentage of goals scored against from set pieces
Updated data getting more than 6000 matches for the AI 433 and 4231 tactics and more than 3000 matches for the 523.
I used 8 base formations. The data comes from FM Arena Patch 24.2 V1.0.
There are 2 or 3 formations that stand out against each AI formation
For the AI 433 I couldn't split 4231 and the 4222 def (deformation/levante) formation, just in front of 4222.
For AI 4231 the 4222 def was just in front of 4231 and 424.
For AI 523 the clear winner was 424 from a grouping of 3, 4231, 343 and 4222 def.
I know with the amount of matches and using data across tactics with different TIs and PIs and tactical styles that this is not perfect. However I do think it is usable if you wanted to try such an approach.
If you come up against any other formations, using the reverse approach ie which AI did better could possibly work, and in all cases it appears the AI 4231 outscores the other AI tactics. So for say 442 you should probably go with 4231.
It is working for me but it could be just the team I have put together and using the top tactics from the formations that is working. I still enjoy using this approach as it uses the different formations rather than set and forget.
Enjoy, and let me know your thoughts and if it works or doesn't for you.
Calculation
Could you by any chance test my setpieces? I already posted them somewhere in the forum, but I didn't upload them to the set-pieces section. I'm interested to see how it will compete.
If you upload it to the Set Piece Sharing Section I will test it in the next few days.
You are welcome to use whichever you choose from the results table. There are many ways to interpret these results.
I have now tested this routine and highly recommend it. The OP has been updated with the results from this.
I do
@Delicious's Attacking Corners
Zimsmula's Defensive Third Defending Throw-ins and Attacking DFK
@AideNGomiinaM's Attacking IFK
@CSTG KANE's Defending Corners and Defending DFK
Chou's Defending IFK
This would have to be tested to see if it has the best set of set pieces, but it's my best interpretation and attempt at it.
Well done
The OP has been updated with the completed results.
The updated @Delicious routine looks very good. Only just behind in overall points but leading or close to it with all other stats.
Zim has 2 so @Han106 might not have removed the second one.
I definitely recommend you to test this. In a season in which I scored 191 goals, I scored 58 corner goals with this routine. Kusi-Asare, with 20 jumping, scored 80 goals in the league that season (:
Han106 said: I combined everything from Zim's that is not Attacking Corners and Chou's Attacking Corners for now.
Start a new post on the Set Piece Sharing forum and I will add it to the schedule
As always please let me know your thoughts
Original tables
Updated tables
yes
Described here
A copy of the league here
Other than that congratulations Mark for the initiative and I imagine that would help a lot of people.
Thank you!
Only the small changes I mentioned on the test league forum
Good work as always
I left it with the staff
Can you tell me about the player attributes in the said database? What's the jumping reach of the players?
@dzek tells me that they are the same as the fm arena database
I will test all the set piece tactics in this forum if I am am able and update this OP with the results. Wish me luck.
So here goes.
The baseline: Katana v1.3 for 4000 tests.
Testing table Update:
Thoughts:
I have now come up with a rating and ranking system. I think that anything that is 10% above the baseline should be excellent and worth using, so there are a lot of excellent candidates for you to use. The stand out top 6 are Blue, Han106 1-24, Azure V3, Chicken Wing v1.8, Han 1-21 and Delicious 4 inside 34. I will be using the top ranked Blue Routine.
Previously I was approaching the results from 2 different ways.
The first was overall impact on results, as you cant tell how many goals have come as a result of the play after the set piece has occurred eg counter attacks, and also the throw ins. I am unsure how to find results from throw ins. So if you prefer this one then you should look at the points difference from the baseline tactic. Blue just leads Zim and Chciken Wing v1.8 from Han106 1-24 on this measure.
The second way to approach it was the overall set piece impact (SP +/-). There are 5 that standout on this measure. Han106 1-24 SP Best Hits seems to be the overall winner here from Han106 1-21 SP Best Hits, Delicious 4 inside 34 who is slightly in front of Chicken Wing v1.8 and CSTG KANE 4231VIFS. I think you could use any of these 5 or Blue, Azure v3 or Zim with confidence.
I would like to thank all the routine developers and sharers that gave me something to test.
Update:
I have added an overall score and rank.
Schedule:
I uploaded and tested Knap MC40, Chicken Wing v1.8 and Azure V3. There are now no more to test, Back to my Scunthorpe save.
Legend:
PPM Points per match
GFPM Goals for per match
GAPM Goals against per match
GDPM Goal Difference per match
Corners Goals from corners
DFK Direct free kick goals
IFK Indirect free kick goals
CornersA Goals against from corners
DFKA Direct free kick goals against
IFK Indirect free kick goals against
Pts Dif The PPM difference from the baseline tactic
Corner Dif Difference between goals from corners and goals against from corners per match
DFK Dif Difference between goals from DFK and goals against from DFK per match
IFK Dif Difference between goals from IFK and goals against from IFK per match
SP +/- total of the 3 Dif columns, so overall impact of attack and defend set pieces
% SP For The percentage of goals scored from set pieces
% SP Against The percentage of goals scored against from set pieces
Zims Best FM24 Set Piece Routine
I used 8 base formations. The data comes from FM Arena Patch 24.2 V1.0.
There are 2 or 3 formations that stand out against each AI formation
For the AI 433 I couldn't split 4231 and the 4222 def (deformation/levante) formation, just in front of 4222.
For AI 4231 the 4222 def was just in front of 4231 and 424.
For AI 523 the clear winner was 424 from a grouping of 3, 4231, 343 and 4222 def.
I know with the amount of matches and using data across tactics with different TIs and PIs and tactical styles that this is not perfect. However I do think it is usable if you wanted to try such an approach.
If you come up against any other formations, using the reverse approach ie which AI did better could possibly work, and in all cases it appears the AI 4231 outscores the other AI tactics. So for say 442 you should probably go with 4231.
It is working for me but it could be just the team I have put together and using the top tactics from the formations that is working. I still enjoy using this approach as it uses the different formations rather than set and forget.
Enjoy, and let me know your thoughts and if it works or doesn't for you.