I believe the fuel system stays pressurized. That does not really matter either though because hydraulics change very, very little when pressurized so again, there just isn't much fuel that would be released if the system de-pressurized (as if through a fuel injector or anywhere else).
What are the overnight temperatures like when the bike starts badly? How about when it starts and runs normally? We are having extremely variable temperatures here in the northeast so if that is the case where you are, it could be what is starting the whole 'poor running' situation.
Does the high idle ever work on your bike?
I still believe there is some simple, direct cause of your problem rather than a complicated, involved cause. Further, things like faulty injectors or coils <usually> do not come and go- they just fail (although it is certainly possible to have a weak coil). The other thing that makes me thing that this is something simple and direct is that the bike runs correctly in all circumstances other than the very first few moments of a cold start, and not always even then. Weak spark components and erratic fuel delivery usually show up under high engine loads or high engine loads at high RPM.
Again, I would find out why the high idle mechanism is not working. Be the cause a direct mechanical fault or the ECU being misled by a faulty temperature sensor would handily explain your current problem.
Brian
If the system stays pressurized, and there is a injector that is dribbling out fuel, it would leak more fuel if the system was under pressure then if there were no pressure, that's just basic physics. Without pressure, the only force would be gravity, and I doubt it would leak much in that case. Still, given the hazardous nature of working on pressurized EFI systems, I would expect an abundance of warnings in the FSM if it were pressurized (to avoid lawsuits). If there are any, I didn't see them, and I was specifically looking for them.
The bike is parked in a insulated garage, so the overnight temps seldom get below 50F, and haven't for several months.
When I started it last night, the high idle worked fine (see my post above, 4th paragraph, for details).
I totally agree with looking for the simplest solution, but I also look for the solution that fits within a given theory supported by the observed facts or data points. So, do you really believe that if the high idle was not working, the amount of fuel from the ECU would be so much that the bike would flood to the point that opening the throttle slightly would kill the engine? I have trouble buying into that premise, since I've been around engines all my life, and usually when you have a mildly rich condition, adding extra air to the mix has an immediate positive affect. Not saying it can't happen, just that it seems unlikely in this case. I would think the amount of extra fuel would have to be significant.
I also agree with your comment about the coils and injectors typically failing under load. That is almost always the case with anything electrical or mechanical, and how I usually find out a component is failing.
There's also another remote possibility that the little rodent that moved in there 2 years ago, munched on another wire that I missed (although I was very thorough in my inspection of all the wires and hoses after that). Perhaps there's a sensor wire that has damage to the insulation, and with a little corrosion, it's increasing the resistance that's reported to the ECU? Not sure why that would come and go though, or not affect the bike under all running conditions.
To be fair, the last two times did not reproduced the exact conditions where I observed the problem (by just starting the bike, running it for a few minutes, then shutting it off). The real test is to ride it for at least an hour. park it overnight, then restart it. I'm riding it today, so starting it tomorrow morning will be a good test case.