RE: This has been a demonstration of optimal posting strategy under Hardfork 20.
I don't know that I can answer all the specifics, though this might be close enough. I don't have time to read through the code today and figure out the methodology; I might do that next week.
Right now what happens is that the early vote bonus comes in the form of assigning a higher percentage of the post rewards to the author. My guess is that they're just going to cap that at 75%, so any value beyond that goes to the pool. But I'm not sure about that.
As far as I can understand it the main goal is to stop people from doing what I'm doing: self-vote and/or buy a chunky bot vote in the early minutes of a post to increase the percentage that goes to the author, and then buy a larger, adjusted vote later to take advantage of the middleman's curation adjustment being set to 25%.
It's probably not a bad idea to stop that exploit, but this is a terrible way to go about it.