Haven't read the whole thread so if this has been mentioned, just consider it another vote if not, well then, I get to be unique
What ever you grand wizards of theory and game design come up with, I'd like to see training level (from laws), unit experience and the skill level of the commanding officer play a much much larger role in both org regain/loss and delay times.
Experienced units rally faster, have experienced NCOs and officers that are experienced at such things as impact both org and delay. They've practiced a lot under real conditions. That's what experience is.
Units with more invested into their initial training and officer/nco selection processes will be better geared toward adjusting quicker to the unusual events that might crop up from time to time on a battlefield. Events such as having to get their crap back into one bag after a battle, or not getting lost on the march.
Higher skilled commanders would tend to be less tolerant of incompetents and/or slackers in the chain beneath themselves. Soooo...
PS. I'd also like to see experience removed from the training laws. I know it's too late to do that but for next game...
Experience is only gained by doing the thing in the real. Morale, elan, esprit and all that sort of thing can be gained by training the hell out of troops and stricter selection protocols for officers and nco's is what should be mimicked. No matter how tough the training, the first go in the real is still a shocker for all involved. Experience fixes that over time. Hard and long training makes that first go in the real a bit more survivable though.
So, I'd prefer it if training laws provided a modifier on shatter tolerance and possibly casualties/str loss rates and some impact on org regain/loss and delay issues. Experience should only be gained through doing. The impact of experience would be heavier toward org loss/regain and less toward str loss mitigation.