I think that this perforated part of the structure may interfere stress contours. That's why you should try to adjust the scale.
Good that you've mentioned reaction forces. They should be checked in such cases. The easiest way (assuming the simulation doesn't take a lot of time to compute) is to replace this encastre BC applied on face with coupling constraint on this face connected with reference point in the middle. Just constrain this reference point to easily measure reaction forces (no need to sum nodes in this case, just one point to evaluate for RF).