The polarity shift helps adjacent transponders not interfere with each other. It also helps with adjacent satellite slots. Same frequency on the same satellite on opposite polarity? That isn't going to work well.
It sounds like you're describing how how Ku transponders are laid out, where you have transponder 1 at 12.224 GHz LHCP, transponder 2 at 12.239 GHz RHCP, transponder 3 at 12.253 GHz LHCP and so on. Directv's Ka and reverse band use identical frequencies for LHCP and RHCP, so for reverse band you have transponder 1 at 17.33 GHz LHCP, transponder 2 ALSO at 17.33 GHz RHCP, transponder 3 at 17.73 GHz LHCP, and so on.
They're broadcasting the "same frequency on the same satellite at opposite polarity" already, and have been doing so for over 10 years now, so I think it is proven to work well Perhaps in the past what you're suggesting was true, and was the original reason behind Ku's stair-step frequency/polarity plan.
Since Directv is going to be broadcasting on identical frequencies with opposite polarities anyway, the question that drives the choice of bonding pairs (from an interference perspective) is: in the event of something that interferes enough to cause signal loss, can only specific frequencies or only specific polarities be affected? Given that the only thing that causes signal loss Directv really has to worry about is rain fade, which will affect all reverse band frequencies and polarities if it affects any, there doesn't seem to be any reason to prefer one method of pairing bonded transponders over another from an interference standpoint. Due to SWM, there isn't any reason to prefer one method over another from the customer end.
So it seems to me the choice will come down to whatever makes it easier for Directv on the delivery end. Whether that's identical frequencies / opposite polarities, some other fixed assignment strategy, or bonding pairs that are flexible and adjusted from time to time for reasons only they know, I have no idea.