( This is for my 2020 SRS bought new, with currently 45k miles and the original battery. )
My experience just for context is if I ride at average road speeds and stay off highways, I don't notice a rebound in SOC after riding.
But... if I get on the expressway for about 15 miles or so, speeds around 80 to 90 mph, it will jump up anywhere from 8-10%.
The degree of bounce back , magic charge, whatever you want to call it roughly correlates to how high the SOC at the start of the trip, and how fast I went. For example if I charged it to 100% until it stopped on its own, then rode 100+ mph for a good portion of the battery discharge, it might bounce back 20% or more after just a few minutes of sitting keyed off. But if it was at 60SOC and I rode slowly, no bounce at all.
I presume some cells are depleted more than others during high current draws, and some kind of balancing occurs between them afterward.
SOC confused me for a long time after buying this bike. I never could understand, how much it really represented.
For example... I set it to charge to 90%, say, come out, and it has reached 90%, but then it continues to charge at 1.3kW for the better part of an hour while staying at 90% before it stops. So I've learned there are two kinds of SOC... the "topped off" reading where it charged until it stopped, and the one where that's just where it is at the moment. You can see how weird this is by doing an experiment: Set your bike to charge to 80%. When it has reached 80% and continued charging for a while, key it on, override, then key off to let it continue charging. Suddenly the SOC will jump to a new level, 90SOC or whatever. You've gone back to the "what is is now" level and not the topped off target.
I got so annoyed by the whole thing I installed a kilowatt-hour counter on my charging box, and it shows me how much power the bike has drawn during charging. Its easier for me to understand. I wish their software offered such a thing as an option for one of the quadrants.