My mistake I put after 3, but I meant on the 3rd. As Rotation 1 you'd Twin, then 2nd you'd True and 3rd you'd True again, only seeing the benefit then.
Your main point was this rotation was intended (to which you've gone back on that slightly now for 5.0). My point is it isn't with how tight it is with the room for error. Also, my original reply was to someone mentioning the 5.x rotation, you replied to that involving yourself in the 5.x double true discussion. Anything that was feasible in the past was irrelevant to what was being discussed. I still highly doubt double true was the intended rotation devs had planned as they seem incapable of developing anything with such foresight into how the jobs will play at the end of an expansion because they don't test those GCD tiers. And because there are those few players that do break the GCD tiers and find a different, alternative combo to be taken advantage of doesn't mean it was intended. Such as the Double Full Thrust rotation in HW for DRG. No way in hell it was intended and required insane amounts of SkS to pull off but again, that required end of expansion (EoE) levels of substats to pull off. And I'm guessing prior to 5.x MNK would also have needed EoE substats to pull off a double true rotation, not to mention it would have been completely cucked by RoF slow back then forcing the rotation to revert back.
We can probably argue this from both sides until we're blue in the face. If you think it was intended then all the more power to you, but from my observations across the job design, it wasn't intended. Especially so with the tight margin for error. Sure, you can break the intended rotation with huge amounts of SkS. That doesn't mean the devs intended it. I can get stacks of GL from an Anatman opener; doesn't mean it's intended. I would love an extra 1-2s on Twin. I'm sure on a reddit post I mentioned they should embrace the double true rotation and extend Twin slightly but they won't because this is MNK we're talking about.
I never said Twin would fall off unless you factor in the margin for error such as clipping or disengages.