Jump to content
Sports Interactive Community
FM Base

Let's talk about tactic testing formula

Recommended Posts

Hey everyone, as you know - FM Base has entered the tactic testing scene this year (long time coming ey). 

I'm looking to gather expert opinions on what makes the most accurate testing environment. 

Should we test with one team or five teams, should we test over 90 games or 180 games? How does reputation fit in to everything?

We're currently using a combination of the Heffem and Mr L formula but we have an expert database editor and we're willing to make adjustments. 

I look forward to hearing the excellent points I'm sure this community will present. 😊👍

Edited by FM Base
Grammar

Share this post


Link to post
Share on other sites

Not exactly an expert opinion but I do have a question about your testing. How do you play out the games? Do you actually play all the games, do you use 'Instant Result' or do go on holliday?

Share this post


Link to post
Share on other sites
17 minutes ago, burnum said:

Not exactly an expert opinion but I do have a question about your testing. How do you play out the games? Do you actually play all the games, do you use 'Instant Result' or do go on holliday?

Holiday for 90 games

Share this post


Link to post
Share on other sites

Ok, so results basically show how effective a tactic is when simulating ('holidaying') games. How much does that correlate with actually playing the games?

Share this post


Link to post
Share on other sites
5 minutes ago, burnum said:

Ok, so results basically show how effective a tactic is when simulating ('holidaying') games. How much does that correlate with actually playing the games?

Quite accurately, but of course it doesn't account for team talks, making subtle tweaks to counter weaknesses during a game and using shouts. As a way to get a baseline result though, it's effective. 

Important to remember a lot of people plug and play tactics 

Share this post


Link to post
Share on other sites

That sounds like a brilliant idea. My suggestion would be a testing environment where all the teams have a full balanced squad of identical players. A global transfer ban for all clubs until 2024, instead of a 90 game holiday, I would propose a 5 season deep tactical analysis in a completely level playing field so that would be in the prem. 38 games x 5 which is much more realistic and comprehensive. Over 150 games in total. You would sim half a season at a time coming back to ensure the squad is happy and the assistant is given a new contract. Record comprehensive data at the end of each season and deliver for each tactic a full 5 year deep analysis. Of course it wont be the same as a human playing all the games so e.g you might find if a team wins the champs league and fa cup, the domestic season may dip where a human manager would have used rotation to better effect but it will still give you a concrete idea as to which tactic is best. Its not brain surgery, indeed common sense says that when the complete field is level then it can only become down to tactics. When Arsenal plays Arsenal. Who wins? The one with the best tactic of course! 🤣

Edited by Fenech

Share this post


Link to post
Share on other sites

I'd be very curious to see how the default styles with basic roles stack up against the created tactics, bit of a pest to test them all though!

Share this post


Link to post
Share on other sites
42 minutes ago, Fenech said:

That sounds like a brilliant idea. My suggestion would be a testing environment where all the teams have a full balanced squad of identical players. A global transfer ban for all clubs until 2024, instead of a 90 game holiday, I would propose a 5 season deep tactical analysis in a completely level playing field so that would be in the prem. 38 games x 5 which is much more realistic and comprehensive. Over 150 games in total. You would sim half a season at a time coming back to ensure the squad is happy and the assistant is given a new contract. Record comprehensive data at the end of each season and deliver for each tactic a full 5 year deep analysis. Of course it wont be the same as a human playing all the games so e.g you might find if a team wins the champs league and fa cup, the domestic season may dip where a human manager would have used rotation to better effect but it will still give you a concrete idea as to which tactic is best. Its not brain surgery, indeed common sense says that when the complete field is level then it can only become down to tactics. When Arsenal plays Arsenal. Who wins? The one with the best tactic of course! 🤣

Quite like the sound of this.

With the test above I'd also like it to be forced so different teams are forced to use different formations and styles against you. I think by changing the manager's details for xxx team that should cause them to play in that way.

Though I'd also like to see this flipped to in a more practical way to compliment since in real FM no two players have the same exact attributes, nor do teams go full seasons without suspensions and injuries (okay they can but it is extremely rare). So using the same team that compliments varied tactics/creating a team that has a lot of variety that can be used, however this will be harder as you're not in control of the variables. To further compliment matters, you'd have to go with a poor team / mid team / top team to for further statistics to how the tactic performs on an overall basis. Especially over a 5 year test.

To add another equation in would also be the stance on set-peices. Would be interested to see what happens to a tactic when the set-pieces are defaulted in a seperate test along side and if the results vary, what happens if you take the routines from the best performing set pieces and give them to the best performing tactic with defaulted set pieces? (bearing in mind player positions/roles).

Inpractice you'd use a theory test of say where a team is 120CA / 145 CA / 170 CA and for the practical teams that roughly equal that, like Norwich / Everton / Liverpool (I'm just guessing not actually looking at their CA but I'm sure their are closer canidates).

Essentially statistics from a "theory" test and a "practical" test along side an 2 enviroments of default / custom set pieces, you should cover a lot of angles and get very decent quite conclusive results. Getting a few statistics from there go a long way to, such as "goals from corners / fks / throw-ins, possession, cards, main scorer, highest average rating etc". The theory would also tell you roughly that this tactic/shape worked well against 442/451 but suffered against 5-4-1 (just examples, would probably be worth if possible having the mentalties of the tactics it did well/suffered against).

It would also somewhat be good to have a blurb on the testing table stating that this is a test, and the games were holidayed. To get the best performance, help the players build links (regular game time together/mentoring/units), and keep morale up.
 

Spoiler

And still worth noting, this still won't prevent people from having an issue when they're using the best tactic on there either.

 

Edited by Danwolf

Share this post


Link to post
Share on other sites
18 minutes ago, Danwolf said:

Quite like the sound of this.

With the test above I'd also like it to be forced so different teams are forced to use different formations and styles against you. I think by changing the manager's details for xxx team that should cause them to play in that way.

Though I'd also like to see this flipped to in a more practical way to compliment since in real FM no two players have the same exact attributes, nor do teams go full seasons without suspensions and injuries (okay they can but it is extremely rare). So using the same team that compliments varied tactics/creating a team that has a lot of variety that can be used, however this will be harder as you're not in control of the variables. To further compliment matters, you'd have to go with a poor team / mid team / top team to for further statistics to how the tactic performs on an overall basis. Especially over a 5 year test.

To add another equation in would also be the stance on set-peices. Would be interested to see what happens to a tactic when the set-pieces are defaulted in a seperate test along side and if the results vary, what happens if you take the routines from the best performing set pieces and give them to the best performing tactic with defaulted set pieces? (bearing in mind player positions/roles).

Inpractice you'd use a theory test of say where a team is 120CA / 145 CA / 170 CA and for the practical teams that roughly equal that, like Norwich / Everton / Liverpool (I'm just guessing not actually looking at their CA but I'm sure their are closer canidates).

Essentially statistics from a "theory" test and a "practical" test along side an 2 enviroments of default / custom set pieces, you should cover a lot of angles and get very decent quite conclusive results. Getting a few statistics from there go a long way to, such as "goals from corners / fks / throw-ins, possession, cards, main scorer, highest average rating etc". The theory would also tell you roughly that this tactic/shape worked well against 442/451 but suffered against 5-4-1 (just examples, would probably be worth if possible having the mentalties of the tactics it did well/suffered against).

 

  Reveal hidden contents

And still worth noting, this still won't prevent people from having an issue when they're using the best tactic on there either.

 

That sounds great so essentially then you would have two 5 year deep tactical analysis one theory one practical covering all bases and angles with in the one test environment really sorting out the tactics from the craptics 👍😜

Share this post


Link to post
Share on other sites
3 hours ago, AcidBurn said:

I'd be very curious to see how the default styles with basic roles stack up against the created tactics, bit of a pest to test them all though!

We started testing default presets 

Share this post


Link to post
Share on other sites

If I was to put my tuppence in about this testing, I did have this issue last year with Knap in his environment about making it a very sterile environment.  i.e Injuries eliminated and fitness and match sharpness kept to maximums plus no transfers,  Is this your plan too?

I understand using a sterile environment will allow for a more level playing field, but it also takes away the human element and unpredictability.

But with that aside, in my opinion, the best way to get the best results from the testing would be with 5 teams and ideally the 5 weakest teams(as in the ones predicted to finish botoms) from the top 5 league in England over 100 games. 

Using this method, you can see how tactics work at different levels. Just cause a tactic makes Arsenal unstoppable, does not mean it will make Harrogate Town

Share this post


Link to post
Share on other sites

Would not be last year and even then any test league would be made by Zyndar. Also I don't really test tactics but set them up, and show the result over a season in a real save, apart from the year Zyndar made test leagues, which we both used.

In order to compare tactics it needs to be done in a "sterile" environment otherwise you make the test unpredictable. 

Pointless using weakest team because  most players either play with top teams or they strengthen squad, so the chances of anybody playing in a real save with weakest team is remote.

I have saves in lower leagues and never found a problem with tactics between leagues. Dafuge has probably played with same wide diamond tactic for many years in many leagues.

Share this post


Link to post
Share on other sites
6 hours ago, knap said:

Pointless using weakest team because  most players either play with top teams or they strengthen squad, so the chances of anybody playing in a real save with weakest team is remote

I'd counter that with the popular youth only saves that start in the lower leagues.  Initially the quality of the players coming through the club's academy will be worse than the existing players and the other teams in the league will be buying players to strengthen their squads.

Share this post


Link to post
Share on other sites

Usually you should be making your own tactics for these challenges. If not check out Mentality Matters that helps adapt to match situations.

Share this post


Link to post
Share on other sites

if i may add my mentality about tactic testing I usually move in this way... my focus on the game is building an universal tactic that could be useful for every team and for any league (like a 4141 of last FM)

 

when I build a tactic or testing other tactics i check it on 3 steps: (1) really bad bad team like Vanarama or Italian Serie D (of course i take last team according betting point); (2) middle team like Cagliari, Genoa, Bournemouth, Siviglia, Schalke04 and last (3) Top Team... of course i try to win everything in every step and all in the first year...

 

i know it is a really long test but the most (i think) efficient :) not like some people i see that only test tactics with top team and sometimes does not win nothing xD

Edited by Sirraf

Share this post


Link to post
Share on other sites
On 27/12/2019 at 10:30, Sirraf said:

if i may add my mentality about tactic testing I usually move in this way... my focus on the game is building an universal tactic that could be useful for every team and for any league (like a 4141 of last FM)

 

when I build a tactic or testing other tactics i check it on 3 steps: (1) really bad bad team like Vanarama or Italian Serie D (of course i take last team according betting point); (2) middle team like Cagliari, Genoa, Bournemouth, Siviglia, Schalke04 and last (3) Top Team... of course i try to win everything in every step and all in the first year...

 

i know it is a really long test but the most (i think) efficient :) not like some people i see that only test tactics with top team and sometimes does not win nothing xD

Kinda similiar with the principal of an underdog/subtop test, at the moment we just don't test the top team. Very similiar theory though!

Share this post


Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.


  • Recently Browsing   0 members

    No registered users viewing this page.

×
×
  • Create New...