If he is basing his thinking off of improving the likelihood that a different brand bullet is more likely to work if another brand fails, that is just statistically incorrect. Each round is an independent sample. Assuming each are of equivalent quality and reliability, then they all have equivalent chances of firing. And if you know ahead of time that one is less likely to fire, then it shouldn't be in the gun.
If he is basing his thinking on one round in the magazine showing improved expansion characteristics than the others, then he must be planning to shoot the entire magazine whether he needs to or not. That will impress people reviewing the shoot. And if he doesn't shoot it all, then his theory, which is incorrect anyway is flawed. Again, research the reliability and expansion info and pick one.
Pick one and train with it because, as noted, each of these will have slightly different POI versus POA. Does it matter much? Probably not. Because he is likely less accurate than the variance in the rounds he chooses (assumption here, but if he loads this way it's a good bet).
Finally, that no other professional shooter does this is a big clue.
To the extent that it's his life, his gun, and his attorney, he has the right to do this or anything like this. But is it wise? Clearly not.
B