Tactical experiment: morphing a 3-4-3 defensive formation into a 2-2-4-2 in attack

[WhyMe]

Like a lot of people I've been enjoying playing around with the new positional play aspects of the FM24 match engine and since full release I've been playing around in a test save trying to get a libero based 3-4-3 to work effectively in the way I want it to.  This thread is about what I have learnt so far with my experiment and the tactic I have landed on.

First I started with a few principles of what I was trying to achieve.

1. 3-4-3 formation as the base
2. High press
3. Possession football
4. Double pivot
5. 2-2-4-2 on attack

So the first part of the attack system was the box defence on the transition and in possession.  The furthest forward of this box provides the double pivot.  The  main idea here was to create passing options in the centre for whomever had the ball at the back, but also options to progress the ball forward through the middle.  I experimented with various roles for the back 3 and wing backs and settled on this for my back 5 + GK:

 

The SK is on attack so he will join in with the 3 centre backs a little higher up the pitch to form a diamond in early transition to give the defenders an out ball if needed if they get pressed.

The wide centre backs on defend could probably be any role really given they come narrower in possession because of the libero but the WCB role does progress the ball well without taking too many risks (such as a BPD).

The Libero is the left hand side of the double pivot, I did experiment with him in the left CB position, but I found that he moved too central in the early transition rather than sticking to the left hand channel when moving into the defensive midfield strata which made for messy transitions and often led to giving the ball away in dangerous positions, particularly as the central CB would move to the top of the D when the GK had the ball instead of left to cover the advancing libero.

The IWB on defend is the other half of the double pivot, he will move into the right hand defensive central midfield position, pushing the libero a bit further to left of centre (though not quite as far as I would like).

4 defend roles isn't normally something I go for but in this case I find that if I put either the Lib or IWB on support they sometimes get too far up the pitch to form the double pivot effectively and both on support is defensive suicide on the counter attack.  I do sometimes put one on support against specific opposition formations, but never both.  Libero on support against a 4-5-1 can be very effective for example.

The left wing back is there to provide the width on the left both in the transition and in the final third.  He has PIs to stay wider and run wide with the ball to further emphasise that width

Next we have the midfield 2

The central midfield 2 were probably where I have experimented the most, I knew I wanted 1 creator and 1 runner and that I didn't need to worry too much about defensive cover as I had 4 defend roles in the back 5.  The creator needed to be close enough to the attack that they would be able to thread passes through to the wide forward or striker and the runner needed to be a goal threat.  Initially I thought I would have the striker drop deeper for an attacking runner and experimented with the Mez-a and the CM-a but this just didn't work consistently enough, but after re-reading @Rashidi's excellent thread on the positional rotations for the different roles I settled on the box to box midfielder with get further forward and move into channels PIs.  This proved to be a good compromise, providing a goal threat arriving late in the box, but not going mad about it too early in transition.  For the creative role, I tried a variety of roles ranging from a PI'd CM-s, a DLP and the RPM, but again after another look at Rashidi's thread I settled on the AP-s to get him close to the attack in the final third (with the Roaming PI)

Last but by no means least, the front 3.  I knew that the right hand wide man had to provide the primary width for that flank so his role was a no brainer, winger on support all day long.  However, the other 2 were trickier to decide on.  As I mentioned earlier I initially experimented with a deeper striker to support both the wide forward and the attacking runner from midfield, but this didn't work consistently and didn't really provide the 2 man spearhead that I wanted consistently (tested DLF-a and s and a CF-s here).  So I eventually swapped to a spearhead striker, the Pressing Forward on attack.  The final part of the system was the left hand wide man.  I wanted this player to get into the opposition box and move towards goal without the ball attacking the left hand channel, essentially forming an offset 2 man strike force with the central striker.  Obviously this meant the winger and inverted winger roles were out and I didn't have the player to try a wide target man on attack (though this could be interesting) so it was either an IF-a with sit narrower or Raumdeuter,  I love the concept of the Raumdeuter and have a go at trying to get one to work in every FM (usually unsuccessfully) and so I thought why not give it a whirl here (this time inspired by @04texag's positional play thread).  Much to my great surprise it worked quite well, though I think some player traits would help it work even better (namely gets further forward, gets into opposition area, moves into channels and probably likes to beat offside trap).

The final tactic (including instructions is:

The only PI I haven't mentioned is take more risks on the winger (simply because I had the player to take advantage of it).

Team instructions wise its relatively self explanatory.  The out of possession instructions are pretty default for a high press without going full gegenpress, transition wise I am not using play out of defence and I want my GK to play to centre backs and I was testing this with a top team which would have 6 players forward so counter press made sense.  In possession does vary a bit depending on the opposition formation but the ones shown are my default.  Fairly narrow is to get the Raum a bit closer to goal by default and the rest as relatively standard progressive possession style instructions.

The with ball position map usually looks something like this:

As you can see the the wing back support gets high and forms the left hand side of the 4 in attacking midfield and the striker forms the spearhead.  The Raumdueter isn't quite as narrow or as high as I would like him to be ideally, but it plays out pretty well in general and since he starts the transition out wide it does make sense.

So what are the strengths and weaknesses of the system and where does it fall down?

I'm not going to share the league table or specific results as this was an experimental save to try and get a specific style of play working so I selected a top team (Bayern) which had most of the players I needed to fit the system so winning was always going to be expected, suffice to say I did well in all competitions.

The positives were in offense I scored 89 goals in the league (16 higher than the next best), which is good but not ridiculous given Bayern's relative stature, but only conceded 15 goals, less than half the next best of 34.  Goal difference wise that is an impressive 74, almost 3 times the next best of 26.  Now like I said this was an experimental save so when problems were identified with the system there were a small number of games (3 or 4) I would play several times to test changes to the base tactic as solutions, so take this with a pinch of salt, but still pretty happy.

Goals wise while the central striker did score twice as many goals (33 from 40 starts and 8 sub appearances in all comps) as any other position that did include 10 penalties.  So the goals were shared across all of the front 5 with 4 of the starting 11 hitting double figures and the starting AP just behind on 9.  The 2nd highest was (slightly disappointingly) not the Raum as the first choice BBM chipped in with 16 goals from 39 starts (7 sub).  However, the Raum was the most rotated position and the players that started there also started games as the right winger so this may be why.

Assists were also evenly spread again with 4 of the front 5 getting double figures for assists (this time with the BBM just short on 9) with the first choice left wingback also chipping in with 14 and a scattering from the libero and IWB (half a dozen each).

 

All of which sounds lovely, but what about weaknesses?

The system is heavily dependant on having players capable of playing the key specialist roles (Libero, Raum, IWB) and getting a top quality Raumdueter in particular is hard, even for a team like Bayern.  You need players capable of playing multiple roles and multiple duties for when things aren't quite working in the base system.  i.e. libero or IWB on support as well as defend, CMs who can also play AM or DM if you need a more radical change if things really aren't working (I also had a version with the CMs as DMS and a 4-2-3-1 for when nothing was going according to plan).

The base system in particular struggles to create much against very narrow formations with 2 DMs (e.g. the 4-2-2-2 Brazilian box or the 3-4-2-1 DM - AM narrow) and you usually need to play around with some roles, duties or team instructions to get things to tick against managers who use these systems.  Finding solutions to this was where most of the replayed test games occurred and even then I am not convinced I have solved it entirely.

I am still not really happy with the striker role, but this may be because Harry Kane is well suited to the original plan of a deep dropping striker role, but doesn't really suit a spearhead role (lack of pace and jumping reach), I have considered a poacher for the spearhead role, but I am concerned a loan poacher won't work well.

Last but not least, in an ideal world I would like the Raum to be getting more goals and I do wonder if a left footer on the left would work better here (i.e. gets through in the channel and then smashes it across the keeper for a goal or squares it for the spearhead), but I don't have the player to test this (had to sell Sane).