there's not really much interaction there. if your game plan is good and the opponent should be shooting 40% (to simplify things), and they shoot 40%, then feedback is neutral... if they shoot poorly say 30%, then they are getting a boost on each fga. however, this boost is still just based on what they are supposed to be shooting. maybe instead of 40% shooting, the second half is tending towards 43% to make up for that 30%. but its not like the feedback mechanism has ANY consideration about whether a game plan was good or bad, its simply (using fg% as an example), how well should the players have shot relative to how well they shot - with some adjustment later. if your 2nd half game plan was somehow terrible (not sure how this exactly would happen but still), and now the oponent should shoot 60%, that 60% is still going to be upped or downed based on how well the guys did in the 1st half, relative to expectations.
so, fundamentally - feedback itself, at a mechanism level - has no conception of or relation to the game plan.
but, when you zoom out a bit, does knowing feedback exists impact game planning settings? it definitely can - not super important, but i would say yes. if you are really likely to win a game, then if you are losing with 6m left, feedback is probably on your side - this encourages you to run uptempo or at least normal, if winning with 6m left. if you are really liked to lose, and you are winning with 6m left, even by 1 - you almost definitely want slowdown. i mean, you probably were running slowdown anyway - and running slowdown if winning late against a great opponent, it really makes sense regardless of feedback existing or not, right?
feedback does more directly impact my view on +/- adjustments. i've always been on the edge of the community in the sense that i basically refuse to let the sim make +/- adjustments in the post season. i am basically either losing by 10, or never, for my +/- in the NT. i tend to allow it in the regular season because i miss so many game plans, but essentially, if i am game planning, i don't want the sim to adjust. this has always been the case, but feedback makes it more the case. if my opponent is taking a lot of 3s, and making a lot of them, relative to 'normal expectations' - then probably, feedback is on my side. jacking up the + setting isn't really that productive in this case. this is really assuming i am either on par or better than the other team, where i can interpret losing by 5 or 10 as a thing that shouldn't be the norm.
you could apply a similar line of thinking to the tempo auto adjustments, its really the two auto adjusts and the end of game settings, where this makes a lot sense. you could also argue that there's some relevance to overall settings like +/-, where perhaps a - opens the door to the opponent getting lucky on 3s - taking and making a lot - and that feedback curbs the extreme side some, making it an easier risk to take (if the opponent is worse - or a harder risk to take, if the opponent is better).
overall, you can pretty safely pretend feedback does not exist and do everything you would do anyway. but if you are interested in exploring it further, the above should give you some ideas?