Cold hardiness, this really is individual by individual named cultivar. The multicolor, multiple pattern variations trait has been bred into both relatively cold hardy azaleas, and into even more tender azaleas. You list yourself in zone 7a, which means most satsuki will be hardy when planted in the ground in your area. In a pot, you always loose a zone of hardiness, so in a pot, Satsuki that are hardy to zone 6a would be okay without heat being used in your 7a winter protection. Pots set on the ground out of the wind, should be winter hardy. "Should", myself, I am in zone 5b, I like using unusual pots for azalea, these pots are not reliably freeze-thaw proof. I protect all my azaleas, putting them in a well house that is dark, no light, but stays between 32 & 40 F, (0 to 4 C). Even though they are evergreen, if temps are below 40 F, (4 C), having no light does not cause any problems. I do have a fan running 24/7 in the well house to keep air moving and eliminate fungus problems. Still air is bad.
So summary - there is no blanket winter hardiness statement you can make about satsuki with the variable color patterns. Some are, some are not hardy.
For the next question, it is useful to find on line or books to give you the ''formal'' description of named varieties of Satsuki. This is if you want to be strictly correct. These descriptions, like "conformation to breed" scoring in dog show circles, will tell you what the "correct" percentage of dark selfs, and light colored selfs should be. A self is a flower of uniform color. Most of the multipattern Satsuki produce a certain percentage of single color light and single color dark flowers. Usually the dark is at a lower frequency. Callahan's book is out of print, but if enough requests come in Stone Lantern might get another printing authorized and available. At about $20 it was a great deal, current used prices are way out of line compared to its value. It won't have varieties developed after the first printing. So find as much info as you can about the "conformation to breed" for you named Satsuki, so you know what you should be aiming for. Note, just like the sectors of dark color on an otherwise light colored flower, the presence or absence of flecks of color is important in conformation to cultivar description. These flecks, sometimes call jewel marks also will vary in frequency. Keep track of whether they are supposed to be there or not.
Second, before pruning for conformation, remember young plants do not display the "mature" color pattern. It is very likely you won't see the dark colored selfs until the tree is over 5, sometimes 10 years old. So "conformation pruning" is not a big issue on young trees in training.
If you can't find official cultivar descriptions to give an idea of what good conformation looks like, remember, you want some of each color pattern.
I have 'Akemi no tsuki', as best as I can tell the ideal is about 75% of the flowers should be snow white, about 20% should be the various color patterns, and about 5% should be the dark, scarlet selfs. I will prune to try to maintain this. NOTE: usually, if at all possible, the apex of the tree should be predominantly the light colored self. Or a mix of the light with a few of the pattern flowers.
Branches that seem to produce all light colored flowers often will at some point start throwing the mixed patterns. Where sectors of the flower will be the dark color. This is especially true of young trees. You might identify a branch that never seems to throw the other colors, these should be marked, by a piece of colored wire, or yarn, of a bread twist tie, and then reduced or removed if you are trying to keep all the color patterns. For the light selfs, it takes more than two blooming cycles to be certain it is only throwing the light color. If you root a cutting from the branch that only throws the light form, you will have lost conformation to the cultivar name. Apparently in the nursery trade, Waka ebisu was propagated from the self, and most 'Waka ebisu' from landscape industry show only the solid dark form. But buy 'Waka ebisu' from a satsuki importer, where they were propagated for conformation in Japan, and 'Waka ebisu' will have flowers with dark sectors, dark selfs, and light coral pink selfs and occasional light coral pink sectors. So all colors from a dark salmon to pale coral pink can appear. In addition, 'Waka ebisu' is a hose in hose double. The trait is inconsistent on young plants, but should show on every flower in a mature plant. It is a 'flower in a flower' in terms of how it is double.
The dark selfs, like the full petal reversions of 'Kinsai' and 'Kegon' can take over. Once a branch begins producing only dark selfs, it will never (? I've been told never???) never show the light selfs or the multicolor patterns. Prune out branches that only throw the dark selfs, and if you root these cuttings they will never show the lighter colors, and should not be labelled with the cultivar name, because they don't conform.
In a Satuski, ideally the dark colors appear where one might want to have a shadow. Especially a branch dominated by dark colors, is okay if it is low in the tree where a shadow would be appropriate. Think a billowing thunderstorm cloud, cumulonimbus cloud. The dark should be on the underside. Or the dark should be scattered like stars or jewels. Yeah, poetry is not the same as pruning directions.
A branch that produces mixed marking flowers, colored sectors, is the most valuable to keep. Cuttings from this type of branch will be able to produce all the colors. The branches that only produce light colors, or only dark colors are less important, and as the design permits should be replaced with branches that maintain the whole range of flower forms.
That's all I can tell you. There is no reliable way to tell which branch will produce which color flower that holds true over multiple cultivars. Once you have had a cultivar for a number of years, you might notice a slight difference between branches that produce dark selfs, branches that produce multicolor flowers, and those that produce all light color selfs. But those subtle differences will be unique to that cultivar and could not carry over to other cultivars. You just have to get to know your own tree.
)
