 Show Posts
|
|
Pages: « 1 2 [3] 4 »
|
You're right. That's just my emotions with the bot - I don't like to loose BTC  I would prefer loosing fiat ... So I'm going to do something about it. I m going to make an emotional bet: "BTC will raise in long therm" - and in this case - I would prefer to stay on BTC than sell it too low. So I think to myself: I'm going to implement an algo which will prevent my stupid bot from selling bTC lower than they were bought How do You think about it ? Well, I know the feeling about wanting to improve the strategy But my experience is that it often results in less profit in the long run... There are some very smart people who have developed these strategies, and at least I am not good enough to improve them yet. I often think my bot is stupid, and sometimes it DOES make wrong decisions of course. But more often it's right in the end, and it turns out that it made a BETTER decision than I would have. The thing is that you would like to buy at the exact bottom, and sell at the exact top of every swing, but no bot (or human) can ever predict that (and if you try to achieve that, you would have to take too high risk and end up losing in the long run). The problem with your argument that "it would be better to stay on BTC than selling too low" is that if you (or your bot) truly believes that the price will drop even lower in a moment, it's always smarter to sell now and buy again after some time. Your decision should always be based on "where the price is going" rather than "what the price is". And I think you're always better off trying to "forget" what you actually paid for your BTC when you bought it. The only thing that could make a diferece is the actual trading fees But of course, there can be better strategies out there to discover and develop, so go for it! |
|
|
|
Here's a (very brutal, indeed) fix for everyone having issues with v0 gox apis: v1 work very well indeed!
You need to change the following lines in background.js:
Thanks, I'll have a look at it! The v0 API is unreliable indeed... |
|
|
|
It seems to be "unproof" from minimal price fluctuations (like those from 2013-05-09 09:00).
I think it could be improved adding more smart algo to decide if to buy/sell (ie concerning also wallet history)
Did first step in this direction - just pushed mod which gets wallet history and displays it under chart.
Now we can think how to use the data deciding on current trade ... maybe some additional "wait" mode ? (as extention of currrent tradeEnabled flag)
When the market is calm, like yesterday, it's probably better not to trade at all. No bot (or person), regardless how smart they are can make money on a market that stands still. You need a swing (that is larger than the fees). Also, taking wallet history into account when making buy/sell-decisions sounds wrong for a bot. Because that's exactly what people do when they trade, and that's why we need a bot to keep our feelings AWAY from the decision making After all, your personal wallet history or trading history never ever affects the future of the market, and if you think the price will go down you should sell - no matter what you paid, and if you think the market will go up, you should buy, no matter how you have traded before... But for people it's very difficult to keep the emotions away - at least for me |
|
|
|
I want to first thank you for your hard work on your free bot. I am a little confused how it trades. According to your description, it buys and sells based on the "absolute difference between EMA %" and that statement worries me. I thought you would initiate a buy when it's positive difference and a sell when the difference is negative. How does it determine when to buy and sell based off absolute values?
That description is a leftover from the original bot (that didn't have separate buy/sell thresholds). I'll commit a new version soon with a more correct description. |
|
|
|
Wich threshold do you suggest? Are 0.25% (Buy/Sell T) by default, suitable?
Well, I'm no expert in trading, so other people are probably better to help you with that... I think the optimal threshold setting depends on the sample interval and how "long trends" you try to catch with the bot, and it also depends on how volatile the market is at the moment... The default 0.25% thresholds and 1 hour sample interval comes from the original bot. Many seem to prefer to lower the sample interval and lower the thresholds to allow "faster trading". I can't judge if that's a wise strategy, and I personally think that the biggest problem when trading manually (without a bot) is that you react too quickly and too strong, so the whole point for me using the bot is that it will take it more easy Yesterday when the market was very calm without a distinct trend, my bot lost money because the thresholds where too low and the fees where higher than the profit. On a very calm market, it's probably better not to trade at all (by setting the thresholds higher). But as I said, I'm no expert in trading - I just modified the bot, allowing more settings... |
|
|
|
- Keep an amount of BTC away from trading
- Can you do the same for fiat ? (usefull to test) I started implementing that, but it turned out to be a bit tricky. When making a "buy" with the MtGox API, the bot calls "buyBTC" and specify how many BTC to buy. Or in fact 1000 BTC right now, so in practice it will buy as much as you can afford (if your not filthy rich! The bot doesn't know what the exact price will be in fiat before the trade, so it doesn't know how much BTC to buy to keep a certain amount of fiat. It would however be possible to make an estimate based on last price, but it would not be exact. But maybe that's ok... |
|
|
|
Tobbe
Your modified extension is Ba Dass!
A couple of thoughts:
You should add a space to line 22 of popup.js for the ticks element.
document.getElementById("ticks").innerHTML=bp.tickCount+ " samples"
And I added a new Ema s/l % metric to the graph tooltip (You won't want to add it like this, it's just a hack):
function formatEMALongTooltip(sp, options, fields){
var trend='?';
if (lastEmaTime==fields.x)
trend=(fields.y<lastEmaShort?'<span style="font-weight:bold;color:#6F6">UP</span>':(fields.y>lastEmaShort?'<span style="font-weight:bold;color:#F66">DOWN</span>':'none'));
//
// Display EMA S/L %. Helpful for gauging on the graph when trades execute on new trend directions.
// Round to 3 decimal places.
//
var per = ((lastEmaShort-fields.y) / ((lastEmaShort+fields.y)/2)) * 100;
trend = trend + ' ' + Math.abs(Math.round(per*1000)/1000) + '%';
return '<span style="color: '+fields.color+'">●</span> EMA'+bp.EmaLongPar+': '+formatChartNumbers(fields.y)+'</td></tr></table>Trend: '+trend;
}
I wanted to see from the chart when trades would be placed having different thresholds....
Cheers.
Jon
Thanks, I'll fix the typo and check your addition! I've also thought about trying to add some kind of indication in the chart where the trend switches and the bot would trigger a trade. But today I'll celebrate my birthday |
|
|
|
Yeah, but where? I just upgraded and i don't see it... There's an icon in the popup-window (in the upper right corner)! Sorry but i don't see it... are you sure you've commit the right files? i only see the sparkline & jquery js, and 2 png in your last commit, but no other files such as popup.html or popup.js Also, small typo in the popup.html: Buy threshold: Sell threshold = instead of Buy threshold: Sell threshold :hope this helps, and thanks for all your work! Ohhh, you're right! I didn't get all the files in the commit! It's fixed now - thank's for telling me! |
|
|
|
Yeah, but where? I just upgraded and i don't see it... There's an icon in the popup-window (in the upper right corner)! |
|
|
|
If you want to turn it into a standalone app, Node.js would be the way to go...
Yes, someting like that... Maybe in the future! |
|
|
|
Yes, probably was some problem delivering the data... Now it is working great! After a time of testing, I think you deserve a little donation ... Thanks for your work! Do you think that would be easy turning the extension into some independent html/javascript elements? In order to, for example, host the files on a local webserver and run it from any device, without the need of having chrome and this pluging on every computer/device! I tried with the original extension, but this seems to be a little more tricky (with progress fetch log, etc)... I'm not sure how hard that would be to make it into a server version. I can definitely see the advantage, and I agree it would be nice! But since extensions are javascript-code, it's normally not suited for running on a server right away. I'll have think about that, and how it could be done without too much hassle... |
|
|
|
I've just downloaded it, and I love the way it loads the data... congrats  ! (well, thank everyone modding and coding this extension, not only you, of course!). However, I'm having problems with trading interval; when selecting some on the options panel, if different from 1hour, it automatically changes itself to 1hour! It's a strange bug! Anyone has checked it? EDIT: Now seems to be ok.. bizarre :/ Thanks, I haven't seen any problem changing interval (but it may ju be that I don't do that very often once it's set I'll see if other experience the same... (and right now it seems that MtGox have a bit problem to deliver any data...) |
|
|
|
You could just fork - mine version has now chart and other fixes - You could easilly decide what to get from it ...
Thanks, but there are a lot of internal fixes in my code (for stability and "trading safety") that I'd like to keep. (I just took a quick look at your code, and I have quite a lot of differences that seem a bit tricky to merge... |
|
|
|
Hello! I decided to make my own modification of the original trading bot extension. Feel free to test it: https://github.com/TobbeLino/GoxTradingBotTobliIt's free, and it has the following enhancements: - Set sample intervals for the EMA calculations and trading (5 min, 10 min, 15 min, 30 min, 1 hour, 2 hours, 3 hours) - Decide when to trade after trend switch (1, 2 or 3 samples above thresholds) - Separate buy and sell thresholds - Trade in fiat currencies other than USD - Keep an amount of BTC away from trading - Detect and eliminate peaks/invalid samples from MtGox - Possibility to disable actual trading (the bot does everything, except the actual trading). Good for testing if e.g. a changed sample interval will cause a trend switch and would trigger an immediate trade. - Plus a bit nicer GUI and Options-page... - There are also some extra safety-checks to avoid erroneous trading (check that user info/funds has been properly loaded, and that all data has been fetched before calculating EMAs and trading. With the original extension, trades COULD happen based on old data under certain circumstances, e.g. when the user changed the settings while data was loading) Of course, I have to add the usual disclaimer: Use it at your own risk! ...but I'm using it myself, and I think it's working It's free, but I did put a lot of hours in it, so if you like it feel free to make a small donation: 1LUqdAXvH9gbYemZKeiMrVJ5njhm6ZvKmF |
|
|
|
|
hopefylly not a newbie for very much longer...
|
|
|
|
Doesn't quite make sense to me, but whatever... |
|
|
|
|
Ok, so this is how you prove your not a spamming newbie troll on this forum?
...by posting meaningless spam posts in the newbie section...
|
|
|
|
|
