You are viewing a single comment's thread from:
RE: Potential Bandwidth Issue Fix is in the Works - Softfork 19.1
@smooth, can you explain the bots you use in the SBD/STEEM market? I can't figure out the purpose of the bot always trying to buy 125.000 STEEM, but adjusts the buy price every minute or two. I saw that @teamsmooth-mm likes this method. There's no rewards for providing liquidity, is there?
In any case, I am interested in the bot and how it runs... and why... and whether it is profitable.
The purpose is to provide liquidity to users who want to exchange small to moderate amounts of STEEM or SBD at a small spread. You are correct there is no longer any system reward for liquidity, but market making can still be naturally profitable. For example, the current market as I write this has the bot buying 102.067 STEEM for 125 SBD and selling 101.615 STEEM for 125 SBD. If one user were to buy the maximum and another user were to sell the maximum, the bot would exactly break even on the SBD side but earn a profit of 0.452 STEEM.
In practice the bot is not very profitable, and requires a fairly large stake of SP to keep up with bandwidth demands (also enough trading capital to insure that it never runs out of STEEM or SBD). I mostly run it as a service to the community. Perhaps in the future if volume were to increase a lot it could become profitable.
Man, @smooth, that helps a lot. I had to track down what was happening and scratch my head at it. If you wouldn't mind, I'd like to someday make a post about it and maybe interview you about it. I have a couple comments for later. But for now, this one comes to mind: have you considered that it may be counter-productive at times to manual traders? I'd invite you to go to the market, come up with a plan, and try to set up some manual trades... there are times in which the bots just dominate as they swap/delete/hedge/buy/sell while the humans are sort of like... Huh? There is an aggravating factor to them. I'd even like to game your bots, but it's good to know about the bandwidth criterion.
Could you (at minimum) try a few different tweaks here and there and see if they can work and play well with humans? You may not think about the bots much, but if I get going with my idea, I think I could out-bot them when you least expect it. I'd like to discuss the particulars elsewhere, if you don't mind.
Also, what would happen if you shut down the bots? Have you studied that? Does the market all of a sudden go into meltdown or something?
Thanks again for replying, and thanks @timcliff for this post.
The bots are down occasionally, usually due to technical problems (server runs out of memory, etc.) or the account running out of bandwidth. What generally happens isn't any sort of meltdown but spreads definitely widen a lot and for casual users the cost of exchange goes way up. It might be good for other active traders though, I can't really say.
As for out-botting, go right ahead. It's a free market, and I have little doubt it is possible. If my bot loses at too high a rate I'll improve it or shut it down.
For discussion elsewhere, I'm occasionally on steemit.chat but the best thing to do there is PM me and check a day or two later for a reply. I try to not fall too far behind on PMs (not always successful).
@smooth
i vote you :)
Thank you for the vote!
you are welcome,,, @smooth
please vote my post
come to my blog :D
nice to meet you