Tactic Testing System by TFF

Thanks TFF for all your contributions to the tactics part of FM! Things like these sure make me enjoy the game more.
 
I still don't think we have a good idea of the standard deviation. There is a 25 point difference between Laldy's tactic! Was the 90 point season the outlier? Was the 105 point season an aberration? We really have no way of knowing. 96 games is nowhere near enough matches and until someone wants to put in the time to measure a tactic over 1000's of matches, I don't think we will ever be able to have a definitive test.
 
Some quick meta-analysis (points per match).

[TABLE="class: grid, width: 282"]
[TR]
[TD]Tactic[/TD]
[TD]TTF Test[/TD]
[TD]Lang/Fran/Luc Test[/TD]
[/TR]
[TR]
[TD]TFF Dev V6[/TD]
[TD] 2.22[/TD]
[TD] 1.85[/TD]
[/TR]
[TR]
[TD]Laldy DMs high[/TD]
[TD] 2.03[/TD]
[TD] 2.10[/TD]
[/TR]
[TR]
[TD]MrL 4-4-2 HM2[/TD]
[TD] 1.94[/TD]
[TD] 2.03[/TD]
[/TR]
[TR]
[TD]Mahluk 1[/TD]
[TD] 1.66[/TD]
[TD] N/A [/TD]
[/TR]
[TR]
[TD]Mahluk 2[/TD]
[TD] N/A [/TD]
[TD] 2.00[/TD]
[/TR]
[/TABLE]
 
These tests are getting a bit silly now. They all magically show their creator's own tactics at the top. Hard to trust any of them. Just use the normal database, take the team expected to finish 1st, mid table and bottom and run the simulation 3 times per team.

Do it properly.
 
I think that these tests are good to have basic information about a tactic's competitiveness. But until these are done with instant result, we won't have the full knowledge. I know that even using only commentary mode would take take ages to test let's say 50-60 matches.
 
ralphs said:
Some quick meta-analysis (points per match).

[TABLE="class: grid, width: 282"]
[TR]
[TD]Tactic[/TD]
[TD]TTF Test[/TD]
[TD]Lang/Fran/Luc Test[/TD]
[/TR]
[TR]
[TD]TFF Dev V6[/TD]
[TD] 2.22[/TD]
[TD] 1.85[/TD]
[/TR]
[TR]
[TD]Laldy DMs high[/TD]
[TD] 2.03[/TD]
[TD] 2.10[/TD]
[/TR]
[TR]
[TD]MrL 4-4-2 HM2[/TD]
[TD] 1.94[/TD]
[TD] 2.03[/TD]
[/TR]
[TR]
[TD]Mahluk 1[/TD]
[TD] 1.66[/TD]
[TD] N/A[/TD]
[/TR]
[TR]
[TD]Mahluk 2[/TD]
[TD] N/A[/TD]
[TD] 2.00[/TD]
[/TR]
[/TABLE]
Click to expand...

Never tested TFF's V6..
 
Wakers said:
These tests are getting a bit silly now. They all magically show their creator's own tactics at the top. Hard to trust any of them. Just use the normal database, take the team expected to finish 1st, mid table and bottom and run the simulation 3 times per team.

Do it properly.
Click to expand...

This. The whole custom database thing just moves the whole test further away from what a user can achieve, and therefore actually doesn't tell anyone what the best tactic is.
 
votupehovi said:
I'd say that there's a key difference between TFF's test and "other" creators tests. TFF is the only who shared his DB to other people to try it and confirm that his testing is true.

Anyone can post "fancy" tables that show how cool his tactic and it doesn't take much time to make such tables but without sharing your DB to other people to confirm your test all that is total bullshit that is not worth any attention.
Click to expand...
I always share my database's but I dont make tactics so I will always be impartial.
 
FULL SEASON

TACTIC | TFF DEVASTATOR V6




Highlights

Biggest Win - 5-0 (vs Real Madrid)
Biggest Defeat - 1-4 (vs Barcelona/Juventus)
Highest scoring game - 6-3 (vs Barcelona)

Most games won in row - 4
Most games lost in row - 1
Most games without losing - 11
Most games without winning - 3
Most games without conceding - 2
Total games without conceding - 13
Most games without scoring - 2

Top goalscorer - 23 goals

View attachment 199462

View attachment 199461

View attachment 199460




TACTIC | TFF DEVASTATOR V6 (defense tweak)




Highlights

Biggest Win - 6-0 (vs Man Utd)
Biggest Defeat - 0-5 (vs Barcelona)
Highest scoring game - 6-1 (vs Chelsea)

Most games won in row - 7
Most games lost in row - 2
Most games without losing - 8
Most games without winning - 2
Most games without conceding - 1
Total games without conceding - 14
Most games without scoring - 2

Top goalscorer - 20 goals

View attachment 199451

View attachment 199449

View attachment 199448





CONCLUSIONS


It's not easy to take serious conclusions with such a small sample size. Only one season (even few seasons) would not be enough in my opinion. Also, there are many random external variables that might contaminate these results. This is in fact a really useful tool provided to devote some time trying to make your tactics better but everyone should ask themselves if this is the one that fits better what you're trying to test. From my experience, I honestely prefer test by playing and analysing the game in real time and moving on accordingly to my personal belifs, needs and philosophy. That being said, based on few different results people released so far seems to not exist any statistical evidence between these two versions. In my perspective, original version is extremely efficient the same way tweaked one is pretty solid (which I embrace more based on my personal playstyle).





​
 
TFF, if I'm correct your test would say the Devastator tactic is the best with top-tier teams (looking at the player description in the topic description) and other tactics like Mahluk and the 4-2-2-2 Gatling Project for more mid/low-tier clubs? Or am I wrong?
 
Replayed one season, not quite so good. Overall though P96 W59 D13 L24 FOR221 AG124 GD 97 PTS190 which puts it in a fairly good place compared to some others. For a 'overload' tactic pretty good!


View attachment 199210
 
Run my tactic on your test season just using instant result.

2iu2s5f.jpg
 
Knap said:
Run my tactic on your test season just using instant result.
Click to expand...

Hey Knap, good seeing you around here. I hope you release it, I remember your tactics for FM 11/12, they were very good.
 
Instant result is new to me, so I have no idea how that compares to a proper season. If real results mirror instant then I may release tactics but as a ME update has arrived, no doubt changes will be required.
 
You must log in or register to reply here.

FMBase

Beta

Resources

Back
Top