Generating randomly distributed column values

By 'averages to 80%' do you mean 'is equal to 80%'? I ask, since your random number formula should also average out to 80% over many iterations, so it seems like that should work, but you say it does not. To guarantee 80% of the 60 lines (which is 48) are a 1, you could create an array of 48 1s and 12 0s. Put a random number next to each one and then sort by the random number and then copy those newly ordered 1s and 0s over to your work list. Of course, this is assuming you want the 1s and 0s randomized, otherwise, you could just put a 1 in the first 48 work items.

I've been thinking about this, and no simple solution with a formula comes to mind. I thought of something like this:
=rand()<=(.8-(number of 1s above this cell/total number of cells)
to reduce the odds of getting a 1 as you approach 80%, and that would kind of work to limit the number of 1s but does not prevent the number of 0's being > 20%. As I tried to counter that issue, it just messed up the other part. Seems like there should be something along those lines, but it never came to me.

Of course, there are always macro solutions for something like this, and that may be your best one-stop solution. However, I would also argue that even at 1000s of lines, using the fill command to fill the top 80% of cells to a 1, then the bottom 20% to a 0, fill the next column with rand(), and then a sort. Maybe a minute? Actually, I just tried it for 10000 rows and it was 80 seconds. So, if this is something you do every day, then go to the macro forum and make a request. Who knows, someone may come up with a formulaic solution.