[CLOSED] MtGox Volatility Trading Bot [GMVT-BOT]

[GeoRW]

Original post:
Since there's too many mining and pirate related securities I'm curious if there would be a demand for something like this.
I'm considering creating a security for my MtGox volatility trading bot I created recently. Bot is very secure, never makes a loss, makes money on volatility. The bigger the market volatility, the bigger the returns. But according to my calculations ROI is pretty low. On a high volatile market (like the one this week and last week), it would yield up to 0.6-0.8% per week (I already substracted 10% management fees from that), but usually less than 0.5%. Profits should improve if there's bigger trade volume, currently bot pays 0.6% fee per trade as the trade volume is less than 100 BTC. For trade volume > 100 BTC, MtGox charges only 0.55% etc.
Bot is working on EUR/BTC market and has orders in the range between 3 EUR/BTC to 10 EUR/BTC (current price is around 7 EUR/BTC). If the price is swinging in this range bot is making money. If the price will go upwards of 10 EUR/BTC that shouldn't be a problem because all the profit is always on the BTC side (bot is always buying first if there's available EUR balance) and each order is valued at the EUR price, so I can move the range upwards without losing money. If price will go downwards of 3 EUR/BTC there will be a decision to make if to move the range downwards and lose some profit or close down the operation and return the money.
MtGox account will of course be protected by 2-factor authentication and withdrawals through API will be blocked.
I will setup this as a company with voting rights. I plan to issue 1000 2000 shares at 0.1 BTC, total 100 200 BTC, for starters.
Any questions, ideas?



Bots funds:

Bot	Approximate value of funds	Current Trade Fee
-----------------------	---------------------------------	-----------------------
MtGox USD	150 BTC	0.53%
MtGox EUR	50 BTC	0.55%
BitcoinCentral EUR	70 BTC	0%

Bots performance:

Week	Start Date	End Date	Profit (BTC)	Management Fee (10%)	Dividend per share	ROI per week
---------	------------	------------	----------------	---------------------	---------------------	-----------------
1	13.7.2012	19.7.2012	0.201616	0.020162	0.000726	0.73%
2	20.7.2012	26.7.2012	0.173569	0.060423	0.000625	0.62%
3	27.7.2012	2.8.2012	0.029527	0.002953	0.000106	0.11%
4	3.8.2012	9.8.2012	0.076204	0.007620	0.000274	0.27%
5	10.8.2012	16.8.2012	0.231258	0.023126	0.000139	0.14%	<= added bot working on MtGox USD market
6	17.8.2012	23.8.2012	13.974017	1.397402	0.006288	6.29%	<= market crash and very high volatility
7	24.8.2012	30.8.2012	1.523698	0.152370	0.000686	0.69%
8	1.9.2012	6.9.2012	0.779619	0.077962	0.000351	0.35%
9	7.9.2012	13.9.2012	0.892514	0.089251	0.000298	0.30%	<= added bot working on Bitcoin-Central
10	14.9.2012	20.9.2012	0.605584	0.060558	0.000202	0.20%
10	21.9.2012	27.9.2012	0.40708	0.040708	0.000136	0.14%

Note that some weeks were very volatile. I expect the average dividend to be below 0.5 % per week (maybe below 0.3% unless there'll be higher trade volumes).


Contract:
GMVT Trading Bot asset (GMVT-BOT) was created with the intent to give investors the opportunity to participate in operation of a trading bot. GMVT Trading Bot is operating on bitcoin exchanges (currently at MtGox only but may expand to other in the future). Each share in this asset represent share in the ownership of funds the bot is operating with and share on the revenue. 10% of the profits bot makes will go to the asset operator who will cover all operating costs (electricity, management etc.). Operator will also cover any loss that may result from unexpected bot behaviour. On the other hand, operator is not responsible for funds that may be lost due to particular BTC exchange default and funds blockage.
The operator reserves the rights, regardless of the amount of shares owned, to raise a motion, and close down the operation. At least 60% of the shareholders will need to agree for a motion to be valid. In the event of liquidation, all available funds the bot is operating with will be converted to BTC and will be distributed to shareholders.
Dividends will be paid at least once per month.

[1713468029]

1713468029

[1713468029]

1713468029

[1713468029]

1713468029

[1713468029]

1713468029

[1713468029]

1713468029

[1713468029]

1713468029

[pieppiep]

I like the idea, I've thought about making something like that myself before, but never got the time for it.
For when the range have to be moved down I thought about only using a part of the total money for each range.
That way you won't lose money when the range goes down, but you can use less money to trade.

[Factory]

Bot is very secure, never makes a loss, makes money on volatility.

People claiming "never makes a loss" make me very nervous.....

Just saying.

[ianspain]

interesting, I am looking to invest in a bot, I have extensive financial experience and wouldn't mind both investing and helping you out in a project like this.

[FreeMoney]

What would the algo do if the price went straight up?

[GeoRW]

Bot is very secure, never makes a loss, makes money on volatility.

People claiming "never makes a loss" make me very nervous.....

Just saying.

The code itself is very simple. Bot is just inserting orders when he finds out that the gap between maximum EUR and minimum BTC order is bigger than it should be. He's always buying low and selling high, he cannot make any loss by design. There are no complicated calculations for orders placing. I can put into contract, that I will personally cover any losses that will be caused by unexpected bot behaviour. Of course there is a chance that MtGox itself will close down and block the funds, that case I cannot cover.
I will run the bot for a couple of weeks more to see how it performs. Then I plan to issue shares for just 100 BTC for starters, then maybe expand to other markets (USD, GBP) or maybe to other exchanges. Depends on decision of shareholders, if it will come to that.

[GeoRW]

What would the algo do if the price went straight up?

It will just place buy orders from money he got at selling when price went up. Currently bot is always keeping 300 active orders covering the whole range from 3 to 10 EUR/BTC, when some order is matched, he either places buy order or sell order depending on available balance.

[GeoRW]

I have written a btc trading bot, it takes a lot of time to tune it. Everyday you wake you will check log files and see what it is doing. I did about a month of testing and just gathering information before it even had control of my money. So good luck and never claim "never make a loss" Bots are only as good as there programmers and programmers loss sometimes.

This bot is very simple by design. There's nothing to tune except for the EUR amount of one order and desired profit. Downside of this is that it is not earning a lot. There are different trading strategies, that can yield more but are much more risky and you can lose money. This bot will "never make a loss" in BTC by design. The only possibility he could behave unexpectedly is when he gets wrong information from MtGox about his active orders which is not very likely and there are checks implemented like for example if there's too few active orders (less than 250, which can actually happen when there'll be crazy levels of volatility), bot will stop operation and I will have to inspect and restart it. Also he can place maximum of 1 order per 10 seconds and maximum of 30 orders per hour. I'm sure bot will never lose money and I can cover this possibility in contract. Also I don't expect that bot will ever work with more than 300 BTC amount under one account on one market, so the risk of loss for me is very low.

[GeoRW]

I added some performance statistics for MtGox bot working on EUR market for first two weeks of operation. I payed 8 BTC for asset creation (was 2 days ago, still not on IPO page though). IPO is planned for 10-Aug-2012. I will issue 2000 shares for 0.1 BTC each. Around 250-300 will I buy myself, rest will go for public sale. I will run 2 bots at starters, one at MtGox EUR market with 50 BTC and one at MtGox USD market with approximately 150 BTC.
I already created 3rd bot that is working on Bitcoin Central currently, but that one needs more testing and performance statistics gathering before I can release it. There's no trading API at Bitcoin Central and I have to parse HTML pages, so it needs more testing for safe operation. Also I didn't figure out yet how to secure that account with 2-factor authentication and let the bot work on that exchange at the same time. On the other hand, Bitcoin Central charges no fees, so profits are potentially higher. I will definitely not release this bot before the IPO. There will need to be a motion if we will issue more shares to raise funds for this bot or not.

[unclescrooge]

I would be very interested in such an asset. As you said, glbse is all about mining and pirate asset right now, we could use a little diversification.

[nimda]

Did you see what happened to TYGRR-BOT? Fees ate most of the profits.

[dust]

Bot is very secure, never makes a loss, makes money on volatility.

People claiming "never makes a loss" make me very nervous.....

Just saying.

The code itself is very simple. Bot is just inserting orders when he finds out that the gap between maximum EUR and minimum BTC order is bigger than it should be. He's always buying low and selling high, he cannot make any loss by design. There are no complicated calculations for orders placing. I can put into contract, that I will personally cover any losses that will be caused by unexpected bot behaviour. Of course there is a chance that MtGox itself will close down and block the funds, that case I cannot cover.
I will run the bot for a couple of weeks more to see how it performs. Then I plan to issue shares for just 100 BTC for starters, then maybe expand to other markets (USD, GBP) or maybe to other exchanges. Depends on decision of shareholders, if it will come to that.

Your bot can lose BTC if the price goes up a large amount and never returns to the price level you initially sold at.
Example:
Bid is $8.50 and ask is $9.00.  This is a nice spread so your bot places and ask at $8.99 and a bid at $8.51.  Your ask gets hit, selling for $8.99, and then the price continues to increase, never returning to <$9. You how hold some dollars from selling BTC for cheap and the value of your account in BTC is lower overall.

[btharper]

So this bot basically waits for a wide gap and tries to be the winning bid on both sides of the spread?

[ColdHardMetal]

Did you see what happened to TYGRR-BOT? Fees ate most of the profits.

That was an arbitrage bot that needed to move fiat between exchanges to function, entailing all the fees etc that go with that. I don't think the cost of the trading commissions were really the problem.

[GeoRW]

Bot is very secure, never makes a loss, makes money on volatility.

People claiming "never makes a loss" make me very nervous.....

Just saying.

The code itself is very simple. Bot is just inserting orders when he finds out that the gap between maximum EUR and minimum BTC order is bigger than it should be. He's always buying low and selling high, he cannot make any loss by design. There are no complicated calculations for orders placing. I can put into contract, that I will personally cover any losses that will be caused by unexpected bot behaviour. Of course there is a chance that MtGox itself will close down and block the funds, that case I cannot cover.
I will run the bot for a couple of weeks more to see how it performs. Then I plan to issue shares for just 100 BTC for starters, then maybe expand to other markets (USD, GBP) or maybe to other exchanges. Depends on decision of shareholders, if it will come to that.

Your bot can lose BTC if the price goes up a large amount and never returns to the price level you initially sold at.
Example:
Bid is $8.50 and ask is $9.00.  This is a nice spread so your bot places and ask at $8.99 and a bid at $8.51.  Your ask gets hit, selling for $8.99, and then the price continues to increase, never returning to <$9. You how hold some dollars from selling BTC for cheap and the value of your account in BTC is lower overall.

This would be true in case we decided to close down the operation and spent all dollars to buy BTC and pay them to shareholders. But if the bot is working and price swings in the range he is covering, he is making money, no matter what is the price of BTC. Also when I move the range upwards, no money is lost, cause the dollar value of each order is constant and bot is keeping revenue on BTC side (always spending available dollar balance first).

[GeoRW]

So this bot basically waits for a wide gap and tries to be the winning bid on both sides of the spread?

He's making money on volatility. Always having gap between orders on the USD side and the BTC side, buying low and selling high. The higher the volatility (up and down swings), the higher the returns.

[GeoRW]

I added stats for the third week of running bot at MtGox EUR market. In this period price went just straight up, there were only a few swings back, so the dividend is quite low. Tomorrow I will add weekly stats for bot running at BitcoinCentral, so far it has earned almost 0.4%.

[GeoRW]

Added stats for the forth week of running bot at MtGox EUR market. First 2 days were pretty volatile, price went down, so bot made some coins when buying lower then he sold last week. The rest of this week was calm, except for today, when MtGox had some problems with disappearing orders on ask side (bot made some advantage of the price swings, though his orders were also vanishing most of the time).
Tomorrow is IPO so you may grab some shares if you want. Money from IPO will primarily go to bot that will work on USD market. If enough shares are sold than bot working on EUR market will get additional funds to work with. BitcoinCentral bot is still being tested, I will post his stats tomorrow, so far he's made 0.89%. After all IPO shares are sold and bot will be ready for some real trading I will raise a motion to sell additional shares to get some coins for him to work with.

[Shadow383]

This would be true in case we decided to close down the operation and spent all dollars to buy BTC and pay them to shareholders. But if the bot is working and price swingsin the range he is covering, he is making money, no matter what is the price of BTC. Also when I move the range upwards, no money is lost, cause the dollar value of each order is constant and bot is keeping revenue on BTC side (always spending available dollar balance first).

Absolutely key.

Quick suggestion for a feature to add to your bot that I've seen be very VERY successful on MtGox bots:

Look at bid/ask spread - if looking to make a buy order place 0.01BTC sell orders very close to the buy price.
There are a surprising number of people on MtGox who will look at a spread like that, see the most recent sale price and place an order at that, because they're making a regular exchange rather than speculating and they just want it to go through quickly.

Then, as soon as sell volume appears at that price, you buy into it.

You can do the exact same thing in reverse when making sales. Taking advantage of the lack of "depth perception" of "casual" traders can make you a fairly tidy profit on the exchange, whereas a bot as simple as yours seems more likely to see steady growth eventually followed by catastrophic loss.

[GeoRW]

This would be true in case we decided to close down the operation and spent all dollars to buy BTC and pay them to shareholders. But if the bot is working and price swingsin the range he is covering, he is making money, no matter what is the price of BTC. Also when I move the range upwards, no money is lost, cause the dollar value of each order is constant and bot is keeping revenue on BTC side (always spending available dollar balance first).

Absolutely key.

Quick suggestion for a feature to add to your bot that I've seen be very VERY successful on MtGox bots:

Look at bid/ask spread - if looking to make a buy order place 0.01BTC sell orders very close to the buy price.
There are a surprising number of people on MtGox who will look at a spread like that, see the most recent sale price and place an order at that, because they're making a regular exchange rather than speculating and they just want it to go through quickly.

Then, as soon as sell volume appears at that price, you buy into it.

You can do the exact same thing in reverse when making sales. Taking advantage of the lack of "depth perception" of "casual" traders can make you a fairly tidy profit on the exchange, whereas a bot as simple as yours seems more likely to see steady growth eventually followed by catastrophic loss.

What do you mean by catastrophic loss?  All the revenue bot makes is kept on the BTC side, so he will not make a loss until after operation is closed down at the high price and I will need to convert fiat to BTC. Your  proposed strategy is risky and on MtGox your bait order will be consumed faster than anyone will notice. BTW, I have currently around 1500 very small orders (around 0.01 BTC each) covering the whole range, so I don't see how I could even implement your suggestion. It's not easy to fool the market. This is volatility bot, taking advantage of price movements and it will stay so. You're free to create your own bot and compete with mine