Topic: (Closed) Butter Bot!: Premier Bitstamp, and BTC-E EMA Trading Platform (Closed) - page 73. (Read 274808 times)

Hi thanks for the reply! It's close but it's not exactly what I mean:

The system you currently have can abruptly stop trading when there is no reason to.

I suggest a trailing stop loss per trade:

I buy 10 BTC at 200, and have a stop loss set at 10 %:
the market climbs to 250 and then drops to 225 (a drop of 250 = 10% of the high) therefor the bot executes a sell order.
After this trade the stop loss should be reset untill you buy the next time based on the indicator.

Also I notice you have to set the trading fee manually (this should be an api call to the exchange every X minutes to check what the trading fee is).
Bitstamp has a trading fee based on your volume.

Ok, I just saw your last post saying how smaller intervals help, so I feel better since I use a 30 minute window.

Thanks!

Hi NginUS,
    There is no suggested TSL setting because TSL has to be closely watched because if it is not it can cause serious losses. We largely discourage its use, it was a feature included because a small group of expert traders wanted to test some strategy, that's it.

If you do use it, I would suggest a very high number, but keep in mind that when TSL is triggered it shuts off the bot so you could miss a flash rally. I would personally just trust EMA to protect me.

I hope that helps, please let me know if you need anything at all :=).

Pablo.

Hi Saan1ty ,
   Interesting question. The bots Trailing Stop Loss (TSL) feature, does exactly that, it takes the highest value since you turned on the bot and sells all your BTC for fiat if price falls by the percentage specified.

I say this every couple of pages:  It is an easy feature to understand in principle, but it is very hard to use in combination with EMA and can result in losses if improperly set. The argument on this can go either way, if it were me, I would have my bot on a 30 or 15 minute interval and not use TSL; I would just rely on the bot to act correctly.

Now a lot can go wrong in a flash crash/rocket recuperation, there is no magic setting that will protect you against all cases. A shorter time frame will make the bot more reactive, but it could follow false signals and incur some losses. In the end, you can't look at this in single trades, even if we are discussing large market movements, you have to look at it in aggregate over time or you are really just looking at noise.

Please let me know if you need anything at all .

Pablo.

Pablo,

I'm currently using the recommended default settings you posted that we should use for BTC-e. Those settings didn't include any mention of a sensible TSL value to protect us from a flash crash. What would be a sensible setting there to protect against a crash like in April, or would one not cause losses anyway?

Of course, I don't want to lose a bunch of money if we have a crash, & would prefer to be protected.

Hi ,
  Sure, happy to explain:

Your client just talks to the server to get the current data points, API keys
are not transmitted to our server.

That might change in the future, when we release hosting or remote
controls, but it will be optional and explicitly stated.

Please keep in mind that even if we had bad faith there is very little we could do as your keys should only have "Get Info" and "Trade" permissions.

Please let me know if you need anything at all .

Pablo.

Imho this is a weakness in the bot. Every selfrespecting bot should have a stop-loss percentage set based on the high it achieved while being bought. The bot should pull the latest ticker information every minute in between it's regular intervals to act upon the stop-loss.

YES.

I asked this question earlier and the only feature that should (in theory) protect you is the TSL.

Though as pablo mentioned earlier in this thread it would first sell as it detects the crash and then turn the bot off. So I am not sure if TSL is useful.

Would it do it (sell) after the next update period of 30 minutes? It would seem it would. (Pablo can clarify)

Someone posted a formula that addressed the [potential] issue a few pages ago. But Pablo asked them if they were a mathematician and the discussion did not seem to evolve from there. [unfortunately]

=======[Discount Code for Butter (10% off)]=========

https://butter-bot.com/?r=f5a17f5b

Use my referal code cause' im sooo cute!  



You'll get an extra month free, it's all on me!

Test your settings with the 15 and 5 minute trade frequencies to see how backtesting might go. I'd say look back in each time frame: 1 week, 1 month, 3 month, etc, all the way back to the crash. It may help.

Obviously if it folds within 10-15 minutes and you're on a 30 minute trade cycle, you'll miss the sell, but the bot will buy BTC at rock bottom pricing.

In all honesty a lot of situations can occur: Even if you have 15 minute trading, what happens if something occurs outside of that window?

Don't forget, there is the tick offset feature. This may help, as a lot of automated traders simply work at the top of the hour. You might want to do 30 minute trading with a 5 or 10 minute tick offset. Try playing with it in backtesting, just like above.

I predict we'll be at $300/btc soon, and bubble will pop at $400. But it will rise again I look forward to the next bubble pop. I will be on a buying frenzy.

I began trading crypto earlier this year and found it difficult to track the movements and trends (especially overnight), so I've downloaded the Butter Bot demo and am intrigued by the moving averages, tweaking settings, etc.

I'm currently running the 30 minute trade frequency setting, but I'm wondering what would happen if there is to be another crash in the BTC market similar to April (which I suspect is coming soon). I was sitting in front of my computer when it happened in April, and as I recall, the market went from 260 USD down to 50 USD or so in the matter of 15-30 minutes. If my trade frequency is set to 30 minutes, isn't it possible that the bot could entirely miss the start and finish of the crash and I'd end up selling at 50, if this were to happen again?

An ideal quality of a bot certainly is to remove the element of human emotions, as a poker might put it, "going on tilt" and making bad calls, but in the event of a crash such as the one described above, would it be best to take manual control, or should Butter Bot on a 30 minute trade setting be able to cover something like this? I'm not afraid to take control if I have to (and have the opportunity), but I'm just wondering if one could depend on a bot to ride through something like that?

Thanks in advance for any feedback, and I have been enjoying the trial so far 

When the client is running locally you seem to be contacting your own server a lot (over SSL).
I think people have a right to know what exactly you are sending back to your own server.

Care to explain? Even when saving API keys, butter-bot.com is contacted...

Ah.. i still had the old btcbots ref-url in my signature - thanks for the hint!
Now, it should work again.
And thanks for using my link kamiN ))

Sounds good! I´m gonna think about it 

Thanks!

Hey PapaMiner ,
  Yep, because EMA is a trend algorithm, the bot will only trade when the market reverses itself. Right now it is holding BTC and BTC is going up, so we are on the right track .

If you feel you need more time to evaluate the bot and see more trades, feel free to download another demos at the end of your week, you are no limited to one .

We have thought about putting a High Frequency Trading (HFT) bot out there, it's a lot more "showy" and customers tend to like it, but they don't back test very well. We will keep looking into it if we find a profitable algorithm however.

Please let me know if you need anything at all .

Pablo.

Have you tried Coinbase? It's an online wallet, but will allow you to integrate credit cards/banks/etc. You can buy/sell BTC fairly quickly, and verification is super quick also. You can also use it as an online wallet just like blockchain. I use it to cash out cheaply/safely/quickly. You can sell your BTC today and it wires to your bank within 3 business days, and the fees are really cheap!

Check it out, and if you want to use it, use my referral link in my sig below. It'll give you $5USD in BTC.


The default settings will work, and are safe. Right now you won't see much trading, but you might see some buying if you fund your account with USD. This is because the past few days it's been going up slowly. Once it drops beyond your sell threshold, you will see a sale.

Use the backtesting feature to test your settings. I don't want to suggest particular settings to make trading happen quicker/more often, but you can try it out yourself in backtesting. We can all provide setting suggestions if you ask, but everyone has their own preferences.

I've made money with it, but it took a few weeks and it's best with more than just a few $ worth of btc. Use my referral link if you want to buy

I have the same issue too... I setup everything and started it up... and waiting for some "ACTION"... but nothing happens...

I was hoping the bot to trade 10+ times daily.... to see some action... I currently have the DEMO...

and it is not a demo if it is not TRADING right?

anyway... it seems like the default/given setup for EMA/threshold is for the SAFEST way possible...

Hey Pablo,

yes that makes sense..I wonder why i didn´t think about that   
I already startet to get funds to btce over OKpay, i took me 1 month to get verified there...fuckin sucks when you want to do profit   

Thanks alot!

You'll see more profits with more money--but better yet, when there is a significant difference between the selling and buying price.

If your bot bought .20btc at $120 ($24 usd) and sold .20btc at $130, you will only see $26usd, and that's not counting any fees.

If you had .20btc @ $120usd but sold it at $180, then you would have $36usd, which is more noticeable.

Hey WantaDuff ,
   It looks to me like you're on the right track there. The issue is that with so few funds your profit will fly under the radar as it will be in cents. For example, if you made 5% in a month that would be only 1.50 USD which you could think was part of the normal trading range.

Once you are more comfortable with the bot, I would suggest adding some trading funds, 150-250 USD is the recommended starting base.

Please let me know if you need anything .

Pablo.

Hi Nwfella ,
   If you look a couple of posts back I said I would make sure both you guys got your free month. Please PM me the email you used to purchase from and I will take care of it. I already PM'd Grinny, I'm waiting for his response .

Pablo.