Ah, but you leave off the axis marks, so your graph is WAY zoomed in. Your spikes are in the 0.05-0.10 range on the doser, if even that high. So when you zoom out with 0 dKh at the bottom and say 12 dKh on the top, all your lines are relatively flat. Something like this. And I don't think corals are that sensitive to those level small fluctuations in alkalinity. I do NOT intend to invalidate your point, a CaRx does provide a more consistent value, I just don't think it matters.
EDIT: Please don't be intimidated by my free-hand graphing skills. I had to learn it for my engineering degree. 20 years later, and clearly it's paid off.