"...DeeAnna, the disappointment I feel in you for your mistake is acute, just crushing. You are not allowed. Such a fall from grace!!..."
<sob!>
"...But seriously, looking back at that chart now, maybe the 3-4% difference in the stearic is significant, although now Robert just SB'ed the OO recipe a fair amount and had no separation. I thought perhaps the SB'ing was important in breaking the first emulsion and getting it to reverse. Damnation!!!..."
I think the issue of separation probably has several triggers. Is one sufficient to cause an emulsion switch or an emulsion failure? I truly don't know. And the uncertainty of the unknown is why there are so many superstitions in soaping -- "stir in only one direction" is an old one (but I can see that advice might keep a newbie soaper from splashing lye all around, hadya thotaboutdat?), but we have our own modern day versions.
I do believe temperature and mixing intensity (or lack thereof) are definitely factors. It seems possible that the fatty acid makeup comes into play too. We may never exactly know.
Phase of the moon? Ley lines? A stick blender sacrifice to the soaping goddess? ;-)
PS: My notes show I poured the first batch (100% OO) at 94 deg F (34 C). I did a LOT of SB'ing, and it did go through a rubber-ball stage. This is the one I CPOP'ed and it separated in the mold. Would it have separated if I hadn't CPOP'ed? Who knows!
The second batch (HO safflower & lard) was poured at 90 deg F (32 C). I SB'ed and hand stirred rather like Robert did. The batter looked "normal" throughout -- no rubber-ball stuff. I left the mold out in the open -- no covering, no insulation. It did okay and looks a lot like Robert's.