In the liveletter, Kaiten's removal cited it being an being an "Action Bloat" problem.
The JP word used was "操作量", the "quantity of operations", I interpret this as not relating to the number of buttons mapped on the hotbar.
The objective of the post is not to engage in subjective arguments on the aesthetics, depths of use, or thematic importance of the skill. There is a quantifiable metric that can be used to measure the "quantity of operations", that being CPM.
With that in mind, let's examine if Kaiten really is an action bloat problem.
In 4.x, Kaiten was not frequently used due to using Hagakure to convert Sen into more Shinten was the preferred rotation.
In 5.x, Tsubame was introduced and Hagakure was nerfed to give rise to the strict 1 minute rotation, you would do 4 Kaitens per minute in the standard rotation. There were no complaints regarding Kaiten being too busy.
In 6.0, Meikyo received a 2nd charge, and Ogi Namikiri was added. This resulted in 1 extra Midare cast on the opener, and 1 extra Ogi cast every 2 minutes. If we take a look at the logs of the best performing Samurais in the savage encounters, the average cast per min for Kaiten is just shy of 5.
So, 4 Kaitens per minute was not an issue, but once you cross the line at 5, it needs to be removed? This makes no sense.
Further more, the best performing Samurai in P4S P2 casted 33 Kaitens over 6:43, that is 660 Kenki spent, which is equivalent to 26.4 Shinten casts, thats 3.93 CPM.
So, if all else in kenki remains equal (which it might not, which is another can of worms), removing Kaiten would effectively lower Samurai's overall CPM by roughly 1. This solved nothing.
(Funnily enough on a side tangent, the GNB in the same log had even higher CPM than SAM)
Adding to that, the 3 highest performing SAM in Eden's Promise P1 had CPMs of 47.6, 47.8, and 47.1
Whereas in P4S P2 the top 3 SAM had CPMs of 46.1, 45.3, and 46.1
Action Bloat wasn't a problem for SAM in 5.x where the CPM was even higher than now, I don't think one can, in good faith, make the argument that currently SAM has an action bloat problem.
I will give that, sometimes during 2-minute burst windows, I run into the situation where I'm setting up for
Midare -> Kaeshi -> Sen GCD -> Higanbana -> Ogi
Where I already have 1 Meditation stack, meaning between Higanbana and Ogi I need to fit in a Shoha and Kaiten, but I only have 1 oGCD due to Higanbana being a cast. I deal with this simply by adding another GCD between Higanbana and Ogi, or shift the sequence so that Ogi comes earlier.
There are many ways to alleviate this particular scenario that does not invlove removing Kaiten.
All in all, I am strictly examining whether or not Kaiten contributes to action bloat and thus needs to be removed, and my conclusion is that it does not, nor will its removal reasonably alleviate the non-existant notion of action bloat on Samurai.