With your quoted example, it is difficult to deduce why that happens, but I can have a guess (because I nearly play a different formation every match for most 51 matches a season, probably 45 different pre-match ones at least, with lots of them tailor-made counters to attack opponent’s weakest player, flank, area and certain space, hence know a bit more about positioning).
In general, if we put our passing focus to 1-flank, that flank’s attacking arrow will be around 55-60%; both flanks then it would be 40% on each flank with the middle 20%; playing middle would usually have 60% middle, with 20% on each flank. This is general, assuming that we use symmetrical formations with most players at default positions.
However, when we put 4 or even 5 players onto 1 of our flanks and attack via that 1 flank only (eg. 3-1-4-1L/R-1, with DMC at DL region & ST at AML region, so effectively a 5-man-overlapping flank attack), match stats may show a 65 to even 70% attacking arrow on the left flank, and that is a direct reflection of average players positioning and their actual attacking threat. Same applies when we play a 3n-4-1-2 via middle with ML & MR at MC region, middle arrow will show 80%+ attacks there, simply because all players do move central with formation and order setting, and then player positioning dictates it further.
If the above is understood, I can confidently say that some players of yours, such as MC & ST are placed very central to have affected the % of play negatively onto the flanks. If all MCs are moved to ML & MR regions, all ST(s) move to AML & AMR regions, then the flank attacking % will be enhanced positively for a pre-set both flanks attacks. And my guess is you may have used mixed passing, instead of both flanks, so the players’ own form will further dictate where most play goes, instead of enforced order to flanks only.
And then comes to point of counter attacking. Other managers would call that risky, but I turn on CA for 80% of matches and so I may have slightly more experience with counter attacks than others. CA-on function simply means as it states: our team DMC or defenders will be willing to risk keeping hold of safe possession turnover upon successful defensive moves, such as tackling and pass interception, for an increased chance of launching counter attacks (eg. In animations, when defenders opt not to launch counter attacks, he will pass the possession gained to a teammate, typically sideway or back-pass, just to hold possession; when CA is on, there will be an increase of chances that an immediate long [cross-field] pass will be made upon gained possession, followed by team moves leading to a shot on goal, which could be a goal, save or miss).
And obviously it does not stop there. When our team has a significant quality advantage over opposition, our huge advantage in possession means opposition could hardly get hold of the ball, let alone deep into our half, hence using CA against very weak opposition is a waste of extra conditions no matter what mentality we choose; when our team faces a slightly weaker opponent (eg, opponent attacker whites of around 120%, and our defenders whites around 150%), then CA will be very useful when we turn our mentality to hard attacking plus long pass when it seems to be a silly move. Very hard attacking in such quality difference scenario would not only push our attackers to score more via riskier moves, but also purposefully losing long passes to be intercepted by our opponents who are just good enough to get the possession turnover and then do not possess the required attacking whites to threaten our defenders, and then our defenders expectedly get back the possession turnover to launch a counter attack for another goal. Needless to say, we can only play this kind of strategy when we already have fully analysed opposition threat (so it will be a counter goal to us, but not a possession turnover to opponent leading to goal against). This scenario will still happen with CA off, but just more frequently with CA on. And it goes without saying that this also is a very good strategy to get a 1-0 win when frontline and midfielders simply would not score via direct open play or set pieces.
All about of use with a combination of quality difference, mentality, passing type and if it is a suitable situation to apply it.
So, all in all, animations could both initiate our change of orders (eg. when 2 long passes are intercepted by opposition in first half, we should change to short), substitution (certain attacker missing all easy chances via animations), identification of opposition orders (press, pass & direction types) and also be a reflection of our orders (CA moves). Personally, I am more than genuinely happy if opposition treats animation as useless when they would have missed important signals and information which could have prompted in-match changes to benefit their teams.
More than welcome to play friendlies and test it out there