Thanks, I hadn't really worked out the maths. I would have thought though that the n^2 curve would somewhat counter the loss from votes getting in before you...
You are viewing a single comment's thread from:
Thanks, I hadn't really worked out the maths. I would have thought though that the n^2 curve would somewhat counter the loss from votes getting in before you...
I actually don't know the formula precisely, but curation rewards are not distriuted on the n^2 curve.
Theyre based on the voters rshares (not vshares), the rshares that got in before and the rshares that come in after
one of the reasons that @biophil was so successful with his bot is that he doesn't seem to vote on material with even moderate support at the 30 minute mark.
The whale penalty for bandwagoning is huge, as is the penalty (because of front running bots) for voting on the same author multiple times. But its only huge as a percentage of fairly small potential curation reward.
someone like dantheman or smooth isn't going to put a huge amount of effort into increasing their curation reward from 150 steem\week (which is about where they are) to say 450 because its cheeseburger money relative to their balance.
Incidentally, BP's method is somewhat resistant to frontrunning, inasmuch as if someone cracks his algo, it would be easy to anticipate his votes and come in ahead, but he would no longer vote for those posts (because the front runner has changed the equation)
But you could argue that an aggressive front runner with his algo would be able to force him into his second and third choices.