Wow this is cool, these are obviously factors found through sieve as well not only P-1 and ECM. But it's important.

I think there are two things to think about when it comes to B1:B2.

First is the B1:B2 ratio, when considering using prime95 to determine these factors not sieve... Prime95 spends equal time in stage1 and stage2 when you chose a 1:14 ratio.

Second is how many factors fall below a particular B1 value. Double that value how many etc.

It might be useful to only consider factors 9M<n<20M and P>2^47 or 48, see if p makes a difference.