whereas in fact it was the other way around

Yes, that was my novelist alter ego going overboard with me ;-)

But I stick to the theory that it were the Romans and their high priests who made leap years popular (see this other Wikipedia article). And i stick to the leap day being February 24.

How could I forget the year started in March! Yes, I had read that, too, and it coincides with the fact that my favourite algorithm (from 1963) (partially explained here) (in ALGOL) starts out with

if m > 2 then
   m := m - 3
else begin
   m := m + 9
   y := y - 1
end
[download]

thus re-scaling the year from 1=January .. 12=December to 0=March .. 11=February (only for easement of the calculation, not because of history, but that makes it even nicer)