Hi @paulag, I've noticed there might be a couple of problems with your analysis.

You seem to have used totals over 15 days rather than 14, which means you didn't use the same number of each weekday. I think if you correct this, you'll find it makes quite a difference to your weekday graphs.

It probably affects two days, Saturday and Sunday because you started at 21:00 rather than on a day boundary.

Our provisional graph for votes over days (last 4 weeks) looks like this, which is surprisingly different from yours:
http://www.steemreports.com/vote-days/?mode=stacked

Our times distribution matches pretty well though (again 4 weeks):
http://www.steemreports.com/vote-times/