Problem is, that angle after the brass hits the defector, has nothing to do with the angle it coming out the ejection port to begin with.

So the basics (before the spent case strikes the upper receiver deflector),
Extractor goes to pull the spent case out of the chamber on action stroke, holds onto the brass rim to keep the spent case on the bolt face for the pull to the rear, bolt stops on the rear extraction stroke stall in front of the back edge of the ejection port window to allow a clean pivot of the spent case off the bolt face, and then depending on the amount of dead blow stall effect timing of the B/C at back stall, the tension of the ejector pushing the rim off the bolt face at back of stroke, and how cleanly the spent case rim can release from the extractor on the pivot, will dictate the ejection angle out the port.

So if you keep these basics in mind, then it easy to solve when a rifle is not initially ejecting the spent case out of the ejection port at the 4:30 position, before the spent case hits the upper receiver deflector.  

The glitch, its after the case hits that upper receiver defector to send the case more forward (where the spent case pile ends up on the ground), that they look at, which has little bearing to the true ejection angle before the spent case hits the defector instead.

To which I will add, in a tech pub by Armalite, while they recommended against tuning, if a gunsmith/armorer felt the need to do so, it can be done with ejector and extractor springs:  weaker spring rates shift ejection aft/toward 6.

So if one really wants to tinker...