There is no rational reason to have the ASD set to anything other than its max value (10 minutes) for a compressor based system. If you are triggering the ASD during normal operations, there are mitigation options available other than killing your compressor.
There are all manner of variables that influence what the average, max, and min temp of the liquid will be. The most important thing is to have the probe mounted appropriately to/in an appropriate vessel, and the diff set to your tolerance. Proper probe placement outweighs any of the other uncontrollable variables affecting liquid temps. To get to the level of variance you seem to be shooting for, you will need some extra probes and DAQ gear to see what is really going on, not just what the controller probe is showing. This will all change based on ambient conditions, so you need to add that to the analysis, and control protocols.
I just tape the probe to the side of a vessel, insulate, set the temp and diff to what I want, and call it good. There is nothing wrong with a little variance when fermenting, as long as it doesn't cause the yeast to freak out. Tight temp control (<1F) may not even be what is best for ferming beer.