You are all probably perfectly right when you say one needs to be able to trade in order to to program a profitable bot. However, I do not enjoy trading manually but I do enjoy programming - that’s why I directly jumped into the programming part. I tried manual trading for a couple of weeks but I am quite sure I can never muster the patience for it.
Right now I have like three tracks in this project: the first one is to develop some statistical helper software which allows me to assess a given entry logic with regard to its potential, similar to what the Mechanical Trading guy suggests. In the second track I grab any entry logic I can get hold of and feed it into my helper software, so I can debug and improve the software while getting ideas of what could help in building a profitable system and what is nonsense. So this answers HoG’s question of how I can possibly know which signals to act on - I simply try them all! The last track is “general forex education”, like following this thread, always hunting for new bits of information. When you guys post your setups, I often go to the chart and try to understand what you are doing ;-).
I have not even started demoing any bot, only backtesting. My best try so far trades the hourly chart of GBPUSD for 2007-2010 with a profit of 850$ and a maximal drawdown of 245$. Unfortunately, out of 181 trades, 32 are breakeven and only 9(!) are winners, the rest are small losers. With on average only about 2 winners a year this nothing to be put into practice :-). Anyway, as said before, I do not expect to have a useful bot soon. But still, maybe some day…
Buckscoder, you are my (until now secret) role model ;-). You seem so realistic about the whole bot issue that I really absorb whatever you have to say in this regard.
Have a nice trading day,
solohayuna
PS: Ah, forgot one thing. Buckscoder, you say “The key to trading is imho to learn statistics.”. Well, I did work long time in Academia on a topic dealing with probability & statistics, did teach statistics to students. So I am quite confident I get this part right!
ST, then I hope you got the Kristallweizen. It’s in my opinion the best tasting beer around here.
Hello solohayuna.
Well, I look like this now: :8: after your compliment, lol. Thanks!
However, I’d suggest to find yourself in trading. I am in no way a good role model for trading. My trading itself is terrible, ha ha. And that is the gentle description.
So, if you try to find an edge with the brute force style, that’s fine! However, it needs also patience. Patience is one of the core parts of successful trading imho.
Anyways, you will probably still have to learn how to trade. At least the “mechanics”. To find out how that clockwork gives you an edge. Just the ability to code is imho a great skill, but in trading you need also to know a lot about some other things.
Just two hints for backtesting: Forget demo trading. That just wastes time and is not better than backtesting. Backtest with reliable data on any timeframe h1 and up and if you find something then go live with it with very low leverage and check it out. You can compare then the live trades with the backtest (running the backtest after live trades) and if that is >90% the same (trades have same outcome in backtest and live) then it shows that your backtest is at least telling you most of the truth. After a few months you will also see then if the expectancy of the system is the same as in backtesting. Most important values to look for is drawdown. Take the max. dd of backtesting, multiplied by 2 and then let it run until it would exceed this dd. If your work was successful you will at least see the bot trade for some months without going over the max dd.
After all the most important rule: Kisss. Keep it super super simple! So, don’t look after every statistic book or like people work as academics. You need imho to find a healthy balance between theory and practice. Go live asap. If just with a pico account that’s fine.
Just try to go through your reply a bit at a time. If anything here sounds a bit harsh or blunt, I don’t mean it, I’m only asking a question. Also I should stress right now that I know very little about trading, and a whole lot less about programming, so if something I say or ask doesn’t make sense, again I apologise Solohayuna
First question I’d like to ask I guess is that if you don’t like trading, what is your purpose in doing this. You said you are quite sure you could never muster the patience to learn to trade manually. But to be honest, and again I apologise ( doing a lot of that today ) if this sounds blunt, that to me sounds like, “I can’t be bothered to trade, I’d rather just invent a bot to make money for me while I do something more interesting.”
We all would, but I don’t see how you can if you don’t learn to trade.
To test all signals is a task I rather feel you could still be at in the next life as well as this one. Signals are generated, according to who you listen to and what tools they swear by, in a huge variety of ways. One thing that did occur to me though, and I reckon Buckscoder would be better placed to tell you if this idea would work, is taken from something I saw on the lovely clean chart ( …ahem ) Buckscoder posted not long ago.
At the bottom of his chart you’ll see, wrote in green, ‘Bull Eng’. ( which as you may or may not know stands for ‘Bullish Engulfing’.) Unless I’m mistaken, this is software that gives prompts when it recognises certain known candlestick patterns that usually signal a reversal, continuation etc etc. So that sort of ‘pattern recognition’ software already exists and therefore if you DO know how to programme, and also know the patterns TO programme, you could programme your own bot to act on these signals.
I don’t know how successful that programme would be as signals are wrong as well as right, it would be interesting to find statistics on the success rate of such a thing. Recently I downloaded a candlestick seminar run by Steve Nison, ( the guy credited with bringing candlesticks to the west ). This teaches candlestick recognition in 4 different videos of about 1hr 30 mins each. There are also some PDF’s with it. I would send it to you if I knew how but the file is approx 2Gb in size and that’s too big for an email attachment.
I know there are ways to compress files but I don’t know how to do that. Let me know if you are interested in that and we’ll try to work something out. Might involve you giving me a computer lesson but everything has it’s price LOL !!
At the risk of repeating myself and also sounding like a complete pain in the you know what, I don’t think you’ll have a successful bot until you learn to trade yourself.
You could also try going to the FXstreet website, then choosing ‘Education’ from the tabs at the top, and you’ll find videos from a guy called Sam Seidan. One of the earlier ones is the use of supply and demand levels as areas of suppost and resistance. I wonder if it would be possible to programme a bot incorporating these?
Buckscoder, you are my (until now secret) role model ;-). You seem so realistic about the whole bot issue that I really absorb whatever you have to say in this regard.
Just thought I’d put this in so some-one out there can enlighten me as to what we are ALL doing wrong.
My trading platform runs a monthly competition for members who trade a micro account which they automatically enter you into ( free of charge ) if you have $500 or more in your account. The first prize of $25,000, goes to the person who grows there account by the biggest % each month.
Just read the results of last months winner who grew his account by, are you all ready for this, just over 12,000% !! Yes you did read that correctly, that would be 12 THOUSAND %. Trashing the person who came in second who could only manage to grow his/her account by a miserable 4 THOUSAND %.
WTF are we all doing wrong guys ??? Ever feel like a total under achiever ??
As Nikita said, don’t assume that the strategy being used to make that profit is something that a career trader would be comfortable with. Indeed, the strategy in that case might be to win the $25,000 - use a tiny Stop, like two pips, risk 50% of the account and set a crazy target, like hundreds of pips. Keep repeating (and taking a lot of tiny losses) until one of them plays out.
So it gives the platform a great headline, but does not represent genuine, consistent trading over the long term. Hold the same competition but for people trading a $10,000+ account and it would be interesting to compare the winner’s statistics!
So I’m not knocking it, it serves a purpose, but personally I don’t think that we are doing anything wrong. My best trade ever is under 20% and I’m quite happy with that lol!
Can’t help but feel tempted to go long Euro Dollar right now. I know we have a strong resistance around the 1-3869’s, and I know we’ve failed around the 1-3850 region a couple of times recently, but.
Still hovering around mid 1-3830’s, futures are all pointing higher right now, Mr Ber-fill my pockets first, worry about Italy later-lusconi is sitting in the ejector seat right now. Just tempted.
I think China has PPI figures out overnight and market in such a fragile state that any headline ( good or bad ) could send the market either way.
Managed to scalp 20 pips in the last two hours.
Wish I had went fishing with Buckscoder’s bots!!
EDIT; FXCM dollar index looking as though it could fall so pushing EU up, but all in all I might just content myself with the +20 tonight and read through some of the threads instead.
Really enjoying the VSA thread by petefader right now. Wish I could test it out but my trading platform only provides volumes indicator in form of a line and not bars.
Looked at both charts RC and in the end decided just to stay out. Can’t honestly say I see anything that’s convincing me to trade either way. So I suppose it’s the old, “If in doudt, stay out.” way of thinking for me tonight.
THX to RC for bar info. I do have a couple of questions ( told you ) which I will pester you with later tonight. At work just now, difficult to get time to write
Bingo Bango Bongo!! Nice work, what percentage gain does that give you on your account (I’m a percentage man rather than a lot man)? I looked at Cable, but figured three trades was enough. Of course, now you have me thinking that four is the magic number. Oh well.
Bit busier here - I am currently short USD/JPY, NZD/USD and EUR/CHF. All going fine so far but it is early days…
At work just now ST so figures may be a few pennies out. Started this week around £136 or £137. All totalled this week’s trading has me at £156 just now. Did find a new app for iPhone that gives me better charts but still won’t trade through phone nowadays, so no more trades today.
Fancy USD/JPY drifting lower over the piece but nothing doing now
Nice work ST! Currently out… ran my GU short to 1.6016 for my daily target of 1% (if were talking percentages lol). Normally I would remain out now until NY close and recompute but EU looks to have pretty much bottomed out so I’ll risk a long off 1.3644 until NY close. Nothing ventured and all that! Hehe
Anything around 150 is fantastic from where you started out, better than 10% in a week. BINGO BANGO BONGO!!
Thank you, all three trades are 3:1 R:R so looks to be a decent week so far. Important lesson reinforced for me: you might recall that I backed a losing NZD/USD trade last week, well my basic idea was sound, even if my timing was off, and this week it looks like working out - sometimes we have to take a loss or two on the chin getting the Entry right. With a decent R:R it works overall. Always good to get these reminders of trading basics!!
Yes, big drop on EUR/USD - in a way I’m glad I’m tied up with three other trades as I don’t have to involve myself too hard in wondering how to time the Long! Haven’t been below the current level since early October, though, so certainly 1.3600 is a level to watch for a reaction.
Nice work on bagging your daily target, feels to me as though the market is a little calmer overall. I’m looking to beat 10% this week which would be a marked improvement on the last few lol.
Just add a quick ( probably pointless ) correction to my last post. My account balance figures should have been in $ instead of £. I know there was no need really to add that but I just like things kept as accurate as possible.
You’re a stand up guy, HoG. Although why you want to trade in that colonial currency is beyond me…!! I’ll bet if I wandered into the Trader’s Arms with a clip of greenbacks you’d make me put it on my tab and settle it later when I had some real money in my pocket…
