5G toolbox NR SSB Beam Sweeping example

In the example, the decided receive beam seems to always choose the beam which points to the exact opposite direction from the scatterer. Not sure if I am doing something wrong but cannot figure it out. Is this intentional behaviour?

I tested using both FR1 and FR2. When using elevation sweep the decided beam often just points 90 degrees up, even though the scatterer is at the same height.

Below is an example image with FR2, 8x8Tx, 4x4 Rx with no elevation sweep. The scatterer is at 45 degree angles for accurate SSB angle matching. Instead of choosing the angle towards the scatterer (135 degrees), the simulation chooses the opposite angle of -135 degrees.

EDIT: Fixed the angles of the example. I think the reason the example in question chose the angle -135 instead of the exact opposite angle of 45 is because of the mirrored pattern happening to have slighty better response. I also tested this with the NRRectangularPanelArrays using NRAntennaElements. With these the response of the receive array was not symmetrical, and the resulting receive beam was pointing at exactly 180 degrees away form the scatterer.