Haasonline Simple Trade Bot For BTCe and Bitstamp[Main topic]

[Stephan224]

Stephan, can you explain the difference between the EMA and EMA_Alternative signals?

EMA uses the last value as the last reference point, EMA_Alternative uses the current value as the last reference point. (easy explanation)
The mathematics can be found here: http://en.wikipedia.org/wiki/Moving_average

I have only really been using high-low and that can work. It is also really good with the new jump out function since that works as a stop loss. You do have to reset the high-low indicator to reflect current prices.  I am still trying to understand the new indicators.

I keep on pushing to use the benchmark tool. It is very easy to find out how the indicators responses and when the given thresholds are met. (under which condition)

Think about this:
Setup a trade bot and do not activate it yet. Add the indicators you want to try. Wait to collect data or click the benchmark tool and Load a currency file. As long as you have some data to run a benchmark you are good. Now, go to the Indicators tab and browse through the available Graphs. You can reflect each individual indicators response here. Reflect it to the graph displayed above it.

If you see one of more that is giving you good buy/sell momentum (read this from both graphs) you can repeat the steps you made before but with variants indicators on the few you like. (different settings!) You can repeating this process as many times as you want. But be sure to test the indicators you find best to more currency data. This way your logic will work always or most of the times.

After this, well... activation and happy trading.

As soon as you have used the software for a while, click the benchmark button and save your currency data. This way you use it later on again to refine your trading strategy using the steps described before.

[San1ty]

@San1ty:
I simply avoid it to happen at all.

If you use technical market indicators you will notice they act before the lowest or highest point is reached. Of course this can cut you off some better momentum, but that is why i created version 0.9. Using more indicators an better momentum can be found. And again, there will be overhead inside it.

In the rare case a order be at the situation you have described (0,1%) the overall best solution is chosen: Wait it out. A currency goes up and down in time and the price will be reached most of the times. In absolute rare case it does not, you need to cancel the order yourself.

Hmm, my opinion is that a bot should be completely self sufficient.
I would like to suggest you add a clean-up procedure that checks your open orders that are no longer relevant and cancels them.

Otherwise if you leave the bot unattended for a longer period, you might have an unpleasant surprise.

Stephan, any reaction on this?

[Stephan224]

Stephan, any reaction on this?

Sorry, i have not read the your response.

The Simple Trade Bot has been build for both the short therm and long therm. So any order that will get stocked will simply be forfilled in time. A currency evolves in waves of going up and down. And as long this is stable you do not have to expect orders to be stocked. You can just wait it out. The only case this could happen is if the overall currency is dropping or raising.

There is not really a solution for this problem, because like you say i can cancel an order. But an order can be filled for a part at the exchange, so in case i cancel it you will only be getting a part of the coins back. (the other coins are exchanged already)  In this case your trade-amount will be wrong and everything will go wrong from this point on. I could re-stabilize it using logic and All-in, but a second problem will come. That problem is that i have no way of controlling the minimum trade amounts, so this could also result in getting stocked too. (trade amount to lot to trade -> no trade made, coins stocked) So in the end there is no good solution to solve this issue, the only way is manually control this issue if it happens.

[San1ty]

Stephan, any reaction on this?

Sorry, i have not read the your response.

The Simple Trade Bot has been build for both the short therm and long therm. So any order that will get stocked will simply be forfilled in time. A currency evolves in waves of going up and down. And as long this is stable you do not have to expect orders to be stocked. You can just wait it out. The only case this could happen is if the overall currency is dropping or raising.

There is not really a solution for this problem, because like you say i can cancel an order. But an order can be filled for a part at the exchange, so in case i cancel it you will only be getting a part of the coins back. (the other coins are exchanged already)  In this case your trade-amount will be wrong and everything will go wrong from this point on. I could re-stabilize it using logic and All-in, but a second problem will come. That problem is that i have no way of controlling the minimum trade amounts, so this could also result in getting stocked too. (trade amount to lot to trade -> no trade made, coins stocked) So in the end there is no good solution to solve this issue, the only way is manually control this issue if it happens.

The only logical solution I see is to allow slippage. For example:

- The bot identifies a rally and would like to buy 10 BTC (The current market price is 100 USD)
- The bot is able to buy 5 BTC at 100 USD and another 5 BTC at 101 USD.
- The rally ends and the bot would like to sell the 10 BTC bought. (The current market price is 150 USD)
- The bot is able to sell 5 BTC at 150 USD and another 5 BTC at 149 USD.

Slippage also has it's disadvantage that you will always pay a premium for what you want but at least your order size will be consistent and always filled.

What do you think?

[Stephan224]

Well i think Yes and No at the same time.

Yes because it should solve the issue, in principle. But no because you think to know what the currency is doing, while in fact nobody knows. You could make a buy at both 100 Usd and 101 Usd, but the currency can drop after this to 99 Usd and then you may have missed a better momentum. The same applies sell moment, you can sell it for 150 Usd and 149 Usd. But the currency can go up again to 151 Usd and there will be the momentum you will be losing.

Do not get me wrong because i like your suggestion and i will have an look if i could support this. But i can not promise anything yet, it will not be easy. I have to think out a better and more stable solution then you are suggesting. I must remain focused on 24/7 trading of course and protecting everybody from losses or wrong trades.

-------------------------

Edit: I am searching for somebody who want to help the German market. It means you have to translate the software for me to German and after this you can do affiliation to help to get the software sold. Affiliation is rewarded per sell.

[San1ty]

Well i think Yes and No at the same time.

Yes because it should solve the issue, in principle. But no because you think to know what the currency is doing, while in fact nobody knows. You could make a buy at both 100 Usd and 101 Usd, but the currency can drop after this to 99 Usd and then you may have missed a better momentum. The same applies sell moment, you can sell it for 150 Usd and 149 Usd. But the currency can go up again to 151 Usd and there will be the momentum you will be losing.

Do not get me wrong because i like your suggestion and i will have an look if i could support this. But i can not promise anything yet, it will not be easy. I have to think out a better and more stable solution then you are suggesting. I must remain focused on 24/7 trading of course and protecting everybody from losses or wrong trades.

-------------------------

Edit: I am searching for somebody who want to help the German market. It means you have to translate the software for me to German and after this you can do affiliation to help to get the software sold. Affiliation is rewarded per sell.

Hi, thanks for taking the time in explaining your thought process and being open to suggestions.
What you just wrote is indeed the issue with allowing slippage, losses might be minor or significant and it's hard to determine.

It might be an option (Checkbox: Allow Slippage) that you could configure in the application.

As far as trading strategies go:

Imho a weakness in your application is that users can not configure their own strategies to a wide extend. For example a cool feature would be to have trading strategies as plug-ins and people could write their own plug-ins for their own needs (for example write a small script in VB or .NET).

Also for a default strategy the most effective I have tried out recently on NASDAQ data was a combination of MACD and RSI (Only trade if both indicators signal a buy and sell if both indicators signal a sell). In the BTC rally of 2013 RSI would have indicated overbought but MACD would have confirmed the rally trend and you wouldn't have missed out on that epic rally. Then when MACD discovered a turn around, RSI would still be overbought and you would sell nearly at the top.

[Stephan224]

It might be an option (Checkbox: Allow Slippage) that you could configure in the application.

This is not as simple as you write it down. Adding an checkbox is easy, but making it work is a different story. As i said, i will have a look into this. Right now my focus is on version 0.9.

Imho a weakness in your application is that users can not configure their own strategies to a wide extend. For example a cool feature would be to have trading strategies as plug-ins and people could write their own plug-ins for their own needs (for example write a small script in VB or .NET).

Also for a default strategy the most effective I have tried out recently on NASDAQ data was a combination of MACD and RSI (Only trade if both indicators signal a buy and sell if both indicators signal a sell). In the BTC rally of 2013 RSI would have indicated overbought but MACD would have confirmed the rally trend and you wouldn't have missed out on that epic rally. Then when MACD discovered a turn around, RSI would still be overbought and you would sell nearly at the top.

Just wait till you get your hand on version 0.9. With that version you are able to build up a trading strategy yourself. You will be able to combine more indicators resulting into 1 trade signal. Benchmarking can be done, currency data files can be shared. Results can be shared and verified and strategies can be shared and discussed.

And i will remain active and i will keep on adding indicators, exchanges and new coins.

But if you really want something to program yourself you could use the Command Line Trade Bot for this. With this software you can setup the entire logic in the development environment you want. That piece of software offers a interface over the Command Prompt or using a local webserver.

[seanrarey]

_______________snip____________
"Also for a default strategy the most effective I have tried out recently on NASDAQ data was a combination of MACD and RSI (Only trade if both indicators signal a buy and sell if both indicators signal a sell). In the BTC rally of 2013 RSI would have indicated overbought but MACD would have confirmed the rally trend and you wouldn't have missed out on that epic rally. Then when MACD discovered a turn around, RSI would still be overbought and you would sell nearly at the top."
_______________snip_____________

Interesting you should post this, as I have been trying to do exactly this with the alpha .9 version.  In theory .9 should be able to do this, but I am still learning it's ropes & do not have this "stacked" strategy functioning the way I want yet... will keep all posted...

[San1ty]

It might be an option (Checkbox: Allow Slippage) that you could configure in the application.

This is not as simple as you write it down. Adding an checkbox is easy, but making it work is a different story. As i said, i will have a look into this. Right now my focus is on version 0.9.

I fully appreciate how hard it is to write a trading bot.
With what I said, I definitely wasn't trying to say that it would be easy, just that it should be optional :-).

I look forward to trying out 0.9.

[Stephan224]

To all Alpha testers: Can you please report all the bugs you may have found before friday the 18th? I am want to release the English beta of version 0.9.0 on monday the 21th.

[Tad3j]

I have seen some reviews on the net about this software and so far I am really impressed how easy it is. I am going to buy it in the near future, but there's one question commin to my mind after all of this. Is it possible to built in arbitrage? Since the software is working with two different exchange sites?

[Leehoya]

Seen some good reviews and looking forward to buying it. One question, will there be support for cryptsy?

[Stephan224]

The simple trade bot will not support arbitrage, for this you need a different kind of trade bot. The only thing you can expect is inter-market trade signals. A signal from one exchange to responce to it on another. That is planned for the future as new feature to the indicators.

At the current moment my trade bot supports:
- Bitstamp
- BTC-e
- Camp BX

These will be available at the launch of version 0.9. In the near future you will see that i will add both Cryptsy and mcxNOW. So those to will come after the initial release. We are delayed on development of Cryptsy due slow communication to the exchange. But i already have approval of both Cryptsy and mcxNOW, so yes it will come but a little later then i hoped.

[Newar]

[..] In the near future you will see that i will add both Cryptsy and mcxNOW. So those to will come after the initial release. We are delayed on development of Cryptsy due slow communication to the exchange. But i already have approval of both Cryptsy and mcxNOW, so yes it will come but a little later then i hoped.

Any plans for Kraken.com? I know volume is low at the moment, since they just started out, but it looks like a promising platform.

[Stephan224]

At this moment i have no plans yet to support Kraken.com but i keep a eye on them. If the volume increases and i get a lot of requests for it i will have a look. But at this moment i rather go for Bitcoin.de then Kraken.com, because of the volume.

[dddbtc]

At this moment i have no plans yet to support Kraken.com but i keep a eye on them. If the volume increases and i get a lot of requests for it i will have a look. But at this moment i rather go for Bitcoin.de then Kraken.com, because of the volume.

Can you add 1hr to the time update interval, I see the current max is 30mins.  1hr works better when "stacking" certain strategies.

[Stephan224]

I do not see how this can be usefull. Settings the update timer so high will only result is missing good trade momentums. Or short: You will miss profits/earnings.

This does not mean the lower the better, but is means the more updates the better.

[San1ty]

When you are trading for a very long time (1 Year + of automated trading) it's often good to filter out all that noise. A lot of people trade on 15 minute intervals and even 1 day isn't surprising to me. It really depends on what you want to do with your trading strategy, do you want to scalp or do you want to ride the huge waves?

Candlesticks also offer a lot of advantages besides the chart being more readable:

- You aggregate data on a certain period (X Seconds up to X Days).
- You can let your indicators work on the Open, High, Low or Close of a candlestick instead of just a snapshot in time that you have now with the refresh interval
- A candlestick contains ALL data that happened in that period

[Stephan224]

I have never used my software is this way, but logic tells me it can be done. So i will update the amount of seconds to support longer periods.

A candle stick diagram can be added, however it is not as good you tell here. Of course it shows information like the high/low/open/close price but is does not show the current sell price. And to understand my software correctly you need to understand there are 2 prices at work on each currency. Otherwise mistakes can happen and i get a lot of people complaining because they to not understand it. Thats what i want to avoid. So initially a line diagram is used. But if a lot of people are asking me to use a candle diagram i am willing to add it. For me it is just important that every user really understand what is happening inside the software.

[San1ty]

Candlesticks are always based on trades, not on bids and asks. But I definitely would really enjoy both to be available :-).

Also how do the bid and ask prices currently interact with your indicators?

When you calculate RSI is it on Bid or Ask or on the Midpoint?