You are viewing a single comment's thread from:
RE: Handful of Users Are Gaming the Liquidity Reward System And Earning 1200 STEEM per Hour
The reward is, in a sense, already proportional. Once an account wins the reward its score is reset and has to start from zero. Someone else who earns rewards at a slower rate will not win but their score is not reset. So they can continue to earn points to reach the top of the scoreboard after multiple hours. For example, in a contest between two accounts with one earning points at twice the rate of the other, the slower one will get one in three hourly rewards on average.
I made a test spread at 3.500005/3.500010, which is a ridiculously tight spread.
abit came in between this, self-traded and collected another $5000. basically $100,000 per day is being gifted to abit and his bot appears to swoop in to just get enough points to win that hour's reward.
since he has more capital than anybody else, he can make tighter spreads than anybody else and self-trade as much as needed to win in any given hour.
So what good is slowly gaining points when there is a self-trading bot that will always have more points in any given hour? Your assessment that an account getting double the points loses one third of the time seems to ignore the fact that there is an account that is getting 1000x the rate and just a few minutes is all that is needed to win, every hour.
This is basically a $100K/day hack and while I would be interested in a relatively fair liquidity providing rewards system, I cant even find the API docs to write a bot and there is no way to beat abit with the current rules.
I completely agree with you on that. My point was simply that it is somewhat proportional, not that you can compete with someone getting 1000x more points. You can't, but that is true under any proportional system. Perhaps under another system you have an easier time realizing your 1/1000 share, but it would still be negligible.
(replying to your reply below, which is at max comment depth)
if it is a balance based system instead of unspent outputs, it makes it a lot trickier...
what we need is some metric that consumes a finite resource and it doesnt have to be specifically coinage. Maybe something like (account balances / total activity) can be used. In the case of two accounts trading back and forth, the account balances stays constant while total activity goes way up, so the value of this shrinks to nothing at the abit level. The problem is that there also needs to be a way to pull in all the other accounts so an honest liquidity provider that is doing a lot of trading is able to get proper credits. If a snapshot of all the account balances involved in the market maker's trades can be made, then it could be used as a limiting factor.
anyway, there is a solution, just not sure the right details yet