Yes, i think this particular moment is not representative of the way its supposed to work. A brutal change of direction here due to the high speed and very closed turn exit, plus hitting the curb, what we see is the reception of a jump basically together with big load transfer.
Considering the benefit of being closer to the ground at middle to high speed for the floor as suggested by comments of jenson button. What he said was that higher rake is more stable hence giving a big confidence boost for the drivers as we see also with the redbull drivers. He was talking about the braking phase which is a highly transient phase.
But still lower rake at high/mediumhigh does not as i thought reduce downforce as it is supposed to reduce drag and maybe the answer to this is this picture from the pirelli test of 2017 tires.
The way they increase downforce to match more or less what should be 2017 downforce level, as far as i can see the main modifications are those two little hanging pieces* that reduce side inlet and isolate a little more positive pressure potential from negative pressure potential hence accelerating the speed of vortices increasing induced downforce. Well its my interpretation it maybe wrong. So i suppose its possible to have this reduced inlet effect and as the rotation is in roll it doesn't affect as much the rake as the car rotate over a longitudinal cone.
So more peak downforce with lower rake at high speed which redbull achieve with non linear softer spring at the rear and more stability under braking hence confidence in the rear end staying in the back position which is usefull.
It seems that high rake at low speed benefit essentially the front wing as there problem is understeer. Softer spring at the back allow more load tranfer on the outside front wheel hence less understeer.
* a configuration i tested on a transat chair in ground effect