During this dip storm let us have a reasonable discussion about trading bots. Does any of you have some experience with programming a crypto trading bot? What is your technology stack? Are you hardcoding the usual trading strategies or rather use machine learning? If the latter, then which factors are meaningful besides the obvious ones like a volume and a price? Do you know any reliable sources where I can read that stuff up?
Fun fact: this bot twitter.com
>Fun fact: this bot twitter.com
But that bot's only been tweeting since March...
docs.google.com
Here you go, all transactions listed.
>programming a bot
from what I heard you are better off renting one. Some are like 0,5 BTC per 3 months. So you know, if you can afford 1 BTC (half for renting half for trading) you can get profit.
Do you mind posting some more info? What are reliable sites with such bots? Do they really make 100% profit in three months to make it profitable? Seems unlikely.
I don't have experience programming a trading bot, but I'm trying to do it. Currently only programmed a warner bot which tells me about sudden volume spikes, similar to CryptoPing, but much faster.Currently programming everything in Python, because.. I don't know, seemed easy for prototyping and for querying the APIs. Currently trying to find high volatility charts which a bot maybe could daytrade for ~4% per trade automatically. But don't really have time. Maybe some (~10?) abled anons form biz could open a private bitbucket project and try it together?
are there any you can just download and run?
I dont know anything else right now sadly. Entered the thread out of curiosity, but I could probably ask the guy that I heard talking about bots for more info. He claimed he was doing 500$ a day thanks to the bot (he apparently has a big portfolio it shouldn't be that hard to make those figures, still on a daily basis is pretty rad) and that he got ROI in a little time.
But again, I'm like you I want to know more. Also if everyone starts using bots it will be like in normal finance where everyone is using supercomputers already to get that 1 nanosecond advantage over the competence.
ppl should make an opensource bot :/ where can i find more about reliable bots ?
That's interesting. I also wanted to start with such a warner and then proceed with margin trading manually to ride the whale's waves. Honestly I don't believe in such a project.
where can i find more about succesful bots ?
Good question, I found this one on one of pro trading subreddits. But this is the only one. Actually I suspect that the google doc I linked with all the transactions listed together with past history of the market should be enough to come up with a utility function somehow reminding the Vicky's one. Didn't dive into data yet, tho. I guess I'll try to figure out its long/short position patterns in July, once I'll be done with my finals.
leonardo trading bot
>mfw i could probably program a bot but i don't know anything about finance
just have someone give you the important parameters and you're golden. make moonbot a reality.
Not that easy. The deal is to actually come up with a good utility function made of market's signals.
like a crystal ball
Have anyone tested this bot? It seems rather like a trading platform + some trading scripts. About 100 bucks for a lifetime license seems unlikely to provide you with a cutting edge margin trading bot.
just bought it, been using the demo for a week or so.
Just set my new bots up and going out for a ride. Hopefully there will be some gainz when i return.
Check out the ANN on bitcoin talk. The developers are one there constantly taking suggestions and implementing them quickly. Gets better every week.
I'd definitely be interested in seeing results. Feel free to make a new thread in a week or something and I'll join.
Atm I'm a little bit skeptical due to cheapness of this bot comparing to what said.
opensource bots are usually trash good luk finding one. i use gunbot it costs .1 btc and i use the bollinger band setting for buys and sells 27% and 25% high and low respectively
And what are your results using gunbot? Did you try to compare them with profit obtaining just by holding coins the same amount of time? While over 600x results of the bot I mentioned in the first post may seem gargantuan, it is "only" 5x more then one would obtain simply by holding ETH for that period of time. I'm curious about gunbot performance.
obtained* shit
>$1000 into $619000
Oh. My. God...
I need to figure this out
WHAT DID SHE MEAN BY THIS?
>1000 into 619000
You could have just bought ETH in February and held it.
Nope. On 8-11-15 it had buying power of 1477.5 ETH worth 1000$. Had they just held until 5-12-17 it would be worth $135,722. With trading, they gained 5.08x.
I've held ETH for a year now, just not enough of it. As user further up I know how to program something like this, just need to figure out the trading/ financial apsect.
Heck, if you can make just 1% profit a day, thats 37 times your original investment in a year with compound interest. Seems doable. But then again if it were easy everyone would do it.
zenbot. I couldn't get it to make any money paper trading though.
To follow up on this, try downloading zenbot and changing the config file around. I only messed with it for an hour or so, but I lost money every time with backfilled trades and using the default config file. If someone can figure out how to make it work, I will send them .02 btc
Good thread user
I can't program for shit (I wish I could) but I'm getting into finance and I think a well programmed one can obviously make bank. The future is a bot which not only look at financial cues but also to social ones, like google trends, Wikipedia pageviews, forum keywords and the like.
>But then again if it were easy everyone would do it.
I'm noooot thaaaat sure about this. It seems that crypto market is either too fresh or too different from regular markets, where companies like JaneStreet are spending gazzilions of bucks yearly just in order to optimize trading speed by another nanosecond. I mean, I see two options (especially after some of replies here) that we're either lucky enough to be early to the game or it or bot trading is not doing well in cryptos for some reason (I'd bet their volatility). But I think that Vicky actually contradicts the latter, and it seems to be a project done by a bunch of geeks in their basement, not billion dollar worth companies.
preddy gud
Try out this bot, set it to 1min or less periods and get 10000x profits in 3 months with the right settings, backtested.
And its open source.
github.com
Yeah I guess I don't really believe it myself seing as how I'm going to try to make my own. Will probably post here on any results I get through the summer.
Really curious as to what it would trade on. Looking at charts it seems like you could get pretty good results going on nothing but MACD and volume really.
>with the right settings
Yeah, that's what the game is about. Any hints?
I'll install it next week and try to play a lil.
post settings
Bump, since I'm going to bed.
GIB SETTINGS
he's meming. that bot can't make shit no matter the parameters.