Cryptocurrency Arbitrage Bot - Advice?

[mattzoing]

Hi people!

Long story short. I've developed a cryptocurrency arbitrage bot which analyzes over a dozen exchanges and look for arbitrage opportunities in real time for over a hundred pairs (both crypto/crypto and crypto/fiat). More details below, keep reading!

It takes into consideration all aspects in a successful and low-risk arbitrage strategy, here are some of the features I implemented.

- It calculates all fees involved to not fool itself and only take those opportunities that will be really profitable at the end (that is way more important and tricky than what you think it is)
- It considers the tradeable volume resting on the order books at different price levels (not only the prices or the top of the book)
- It is capable of analyzing possible trades on two or more exchanges exchanges (which maximize the return and also lowers the execution time).
- It consolidates trades and reports what kind of order, for what volume and at what limit price to place them (this saves execution time and lowers the risk involved)
- It filters the most profitable opportunities for each pair (you can select a minimum return of investment)
- Many more features

Here are some very interesting insights. First of all, most articles and youtube videos related to crypto arbitrage are just bullshit, most of them don't consider the fees which is plain wrong, some of them base their strategies on the price of the last trade not the order book, which is really stupid and completely wrong. I learnt there are arbitrage opportunities out there all the time, but it requires a lot of idle capital to get all of them.

All in all my bot works really well and now I'm considering what to do with it. Honestly, the problem is I don't have much capital to invest which is a must have if you want to do arbitrage, also I'm not holding many different currencies, only BTC and a few others so I just can't trade many pairs.

What other alternative ways of monetizing this project can you think of? I'm open to offers to sell this project or licences to use the bot, but I'm not sure where to ask for that, is there any appropriate forum to sell it?

If you're doing arbitrage yourself or you want to ask me something comment below or send me a PM.

Cheers,
Matt

 

[TheVigilante]

We are already doing same thing nd also filed patent for it

 

[Grimasaur]

Hello, i have some questions i have a project that its an arbitrage bot too
wich is currently on testing with an okay success
do you do arbitrage only between exchanges ? example , buy on exchange A and sell on exchange B so you get a profit there
do you have all your funds in one exchange after a succesfull trade? example you buy something on exchange A then sell it on exchange B and buy other thing and sell it on exchange C then move it all to exchange a again
do you have in consideration that on the time that you buy something someone may have took that buy order for you to sell it? what do you do then

We are already doing same thing nd also filed patent for it

how are you going to patent arbitrage? maybe i dont know a lot about patents but it seems like if you try to patent SEO or something like that

 

[TheVigilante]

It's totally AI
It's going to be automated you set what indicators you want and it does the rest making trades buy in and sales for you!!

 

[Grimasaur]

It's totally AI
It's going to be automated you set what indicators you want and it does the rest making trades buy in and sales for you!!

but thats not arbitrage , its automated trading right?

 

[MrAlex123]

I have read something like this in a few other forums as well, did someone actually succeed in coding something like this? and abhi007 I highly doubt that you can patent it

 

[TheVigilante]

but thats not arbitrage , its automated trading right?

Yup nothing like this we are the pioneers obviously.

 

[mattzoing]

Here are some examples of what I'm getting right now in real time.

OPPORTUNITY ON BTC/USD
> Volume 7.06663835 BTC
> Investment 63241.80674390178 USD
> Revenue 64234.9733743990 USD
> Commissions 199.58665418990259 USD
> Profit 793.57997630731741 USD
> Return 1.2508 %
> Orders
BUY on Bitbay 0.07270264 BTC at 8936.71 USD
BUY on Kraken 1.32 BTC at 8950.00000 USD
BUY on BTC-e 1.16587185 BTC at 8956 USD
BUY on Bitstamp 0.06486401 BTC at 8947.99 USD
BUY on BitFinex 4.44319985 BTC at 8956.3 USD
SELL on Hitbtc 7.06663835 BTC at 9062.19 USD

OPPORTUNITY ON XZC/BTC
> Volume 2.52316762 XZC
> Investment 0.0112117286319388 BTC
> Revenue 0.0113593910536800 BTC
> Commissions 0.0000337828483175576 BTC
> Profit 0.0001138795734236424 BTC
> Return 1.0127 %
> Orders
BUY on Cryptopia 2.52316762 XZC at 0.00445976 BTC
SELL on Binance 2.52316762 XZC at 0.00449000 BTC

OPPORTUNITY ON ADA/ETH
> Volume 2998.00000000 ADA
> Investment 1.0467251800000000 ETH
> Revenue 1.099188000000000 ETH
> Commissions 0.0021459131800000000 ETH
> Profit 0.0503169068200000000 ETH
> Return 4.7972 %
> Orders
BUY on Binance 2998.00000000 ADA at 0.00034941 ETH
SELL on Hitbtc 2998.00000000 ADA at 0.0003610 ETH

 

[mattzoing]

Wow, already some comments. Let me go one by one.

We are already doing same thing nd also filed patent for it

There's no way you can get a patent for an arbitrage bot, people have been doing this for ages in the forex market.

do you do arbitrage only between exchanges ? example , buy on exchange A and sell on exchange B so you get a profit there

Multiple exchanges at the same time, check the examples above.

do you have all your funds in one exchange after a succesfull trade? example you buy something on exchange A then sell it on exchange B and buy other thing and sell it on exchange C then move it all to exchange a again

This is another very important thing to consider, it's called rebalancing. This is done automatically if you wait long enough. At some point A will be cheaper than B, or either you can rebalance the accounts manually by transferring from B to A, there's a small FLAT fee associated with that.

do you have in consideration that on the time that you buy something someone may have took that buy order for you to sell it? what do you do then

I do, it's also a good question! The prices I reported above are limit prices, not the exact price for the orders I found on the order book. If you place orders using that price most of your order will be filled immediately, chances are you get 100% filled at that price. How I calculate that price is part of the secret sauce.

did someone actually succeed in coding something like this?

What do you mean? It's certainly working.

 

[greymoon]

It's totally AI
It's going to be automated you set what indicators you want and it does the rest making trades buy in and sales for you!!

Artificial Intelligence? That sounds interesting. Because if you use just simple indicators and bolinger bands so GUNBOT do this for years - but as I understand you are building something more interesting here. Am I right?

 

[TheVigilante]

Artificial Intelligence? That sounds interesting. Because if you use just simple indicators and bolinger bands so GUNBOT do this for years - but as I understand you are building something more interesting here. Am I right?

Exactly

 

[mattzoing]

Exactly

There are at least 5 other projects doing AI on crypto, there's no way you get a patent there either, in fact it makes no sense to request one. I don't understand why you mentioned it...

 

[rodmarb2b]

Very good thread.
I´m interested in cryptobots.
Following.

 

[Vox0ut]

Is your bot doing automated trades or are you approving manually?
If your bot is working I why not just trade up and make money with it?

 

[Ngo]

A lot of unrelated comments in this thread: let's focus on helping the OP here.

@mattzoing, what you've presented is a classic problem for any high-growth startup, i.e., lack of capital. What you _can_ do is run your bot as a SaaS (software as a service) webapp, where you accept monthly subscriptions + percentage fee from your users. You've done a lot of work already, but there needs to a bunch of more programming to be able to market/sell this bot -- trust me I hate this step too

After this, your goal is to get a large client who wishes to run your bot at high volumes. To be able to convince a whale to use your bot, you'll need to show smaller users and evidence of actual profit. This would be an 'enterprise'/'over the counter' deal where you provide additional support, maintenance, private hosting, and anything else they need. Your goal would be to acquire a handful of whales who consistently use your bot -- 5 whales doing 50 BTC volume/month = 250 btc vol/month; charge 5% fees = 12.5 BTC per month, sounds like a sweet income

As you reach that stage, you'll find a lot more options to be able to use your bot. If you've made something that can consistently get profitable returns, you are on a goldmine. Work hard and work smarter and you're going be very, very successful.

 

[Juis money]

Hello, I created this forum account because of your post am new here can you send me a PM have an offer for you

 

[mattzoing]

Hello, I created this forum account because of your post am new here can you send me a PM have an offer for you

I'll send you a PM in a moment.

Is your bot doing automated trades or are you approving manually?

It's capable of executing orders by itself, but right now it's in alpha or early beta so I'm not doing so.

If your bot is working I why not just trade up and make money with it?

I explained why in my original post. You need an interesting amount of money invested in different currencies waiting for the opportunity to arbitrage them, which I don't yet have.

 

[Deleted member 969102]

Selling this bot to multiple people is a bad idea. It means that the opportunities it displays aren’t actually real, because x other people are thinking of buying and everyone can’t have the whole thing.

The patent thing seems like absolute bs to me

 

[mattzoing]

A lot of unrelated comments in this thread: let's focus on helping the OP here.

@mattzoing, what you've presented is a classic problem for any high-growth startup, i.e., lack of capital. What you _can_ do is run your bot as a SaaS (software as a service) webapp, where you accept monthly subscriptions + percentage fee from your users. You've done a lot of work already, but there needs to a bunch of more programming to be able to market/sell this bot -- trust me I hate this step too

After this, your goal is to get a large client who wishes to run your bot at high volumes. To be able to convince a whale to use your bot, you'll need to show smaller users and evidence of actual profit. This would be an 'enterprise'/'over the counter' deal where you provide additional support, maintenance, private hosting, and anything else they need. Your goal would be to acquire a handful of whales who consistently use your bot -- 5 whales doing 50 BTC volume/month = 250 btc vol/month; charge 5% fees = 12.5 BTC per month, sounds like a sweet income

As you reach that stage, you'll find a lot more options to be able to use your bot. If you've made something that can consistently get profitable returns, you are on a goldmine. Work hard and work smarter and you're going be very, very successful.

Thank you for returning the thread to its original course.

I understand what you're talking about pretty well, that's exactly the problem right now. In order for this to work at scale I need a few big investors and to get them I need many small investors to prove it works. However, as you mentioned I would need to create a whole web/platform/something around it, which would take me a few more months and a lot of resources. That is why I'm looking for alternatives.

 

[mattzoing]

Selling this bot to multiple people is a bad idea. It means that the opportunities it displays aren’t actually real, because x other people are thinking of buying and everyone can’t have the whole thing.

I see your point but that's partially correct. There are some opportunities to invest over 100k usd, unless they all want to put that amount there shouldn't be any problem. For instance, 10 investors can put 10k and all of them would take some profit out. For smaller opportunities however what you said is correct, first come first served. But even in that case there are a lot of currencies to diversify your portfolio so that shouldn't be a problem either.

I'm getting dozens of reports per minute, way more than what I would ever invest in those currencies.

The patent thing seems like absolute bs to me

Completely agree. I asked him for more details and still no response. Makes no sense to me.