Trumpcard: a candlestick closing price breakout EA by SteveHopwood

SERIOUS WARNING

 

1. Most Forex traders lose all their money.
2. Using the robot posted here in trading Forex does not guarantee success.
3. Trading this robot could lead to serious financial loss.
4. Trading this robot without understanding its underlying trading strategies guarantees traders will lose their money.
5. This is not a set-and-forget ea; there is no such thing and anyone who tries to claim there is, is either stupid or lying. This ea requires frequent manual intervention.
6. At best, a trading robot is only 90% as good as the manual strategy it trades. At best. At worst, it can be much less effective. If the strategy is rubbish, so is the robot.
7. To trade this robot, you have to understand:
   1. How to use EA's.

DO NOT USE THIS EA ON A LIVE ACCOUNT. IT IS EXPERIMENTAL.

Here is a digest of a pm I received from Trumpcard a couple of weeks ago:
"Recently I've been dabbling in 15 minute closing price breakouts among other things. I've been tinkering around with the relationship of the last closing price relative to the previous three closes and basing entries on that.

I've come up with rules for a 15 min system which on some very limited cable backtesting showed quite some promise. However, adapting this to the 5 min timeframe (something I've never really traded/examined before) results seem even better.

I would lke to know if the below trading system has some merit. Do you think it's worthwhile sharing it publicly with the idea of forward testing and improving it?

Here is what I've come up with:

Trading from 11am until 10pm. Thi is when to take all trades and is something that could be altered)

I've looked at this on EU and E/J on random dates.

EJ 13th Sep had 24 wins, 8 losses.
EU 23rd Sep had 21 wins, 12 losses. 19Sep had 11wins, 13 losses.


Buy setup (opposite for shorts)

From 11am, wait for a bar whose close is higher than the 3 previous closes. This will be the trigger bar.

Place a pending buy order 1 pip above the high of the trigger bar. TP and SL are 10 pips.

If the pending order isn't triggered on the next bar, cancel the order and wait for the next trigger bar.

HOWEVER, IF the bar(s) immediately following the trigger bar are Inside Bars, the pending order remains valid and then triggered on the subsequent bar or cancelled if no breakout.

Only take 1 trade at a time. Once a trade closes, immediately look for the next trigger bar. This could be on the current bar of which the previous trade exited.

A suitable R:R is 1:1"



I slightly misunderstood Dan's pm. I thought he was asking me to code an EA, not whether I thought it worthwhile sharing the system here at FF. I gave my standard response: loosely translated it is, "Come back when you have traded this manually and successfully for a couple of weeks, then I will think about it."


Last week, Dan came back to me with this:
"hi Steve! Well I've been forward testing my strategy on EJ and EU since the 23rd September and have just finished back-testing for the rest of September. It has taken me many many long hours examining each chart. I have saved each chart annotated with every trade indicating whether it was a win or loss. On a few occasions when I couldn't zoom in any further with some trades, I simply marked them as a loss.

I am very pleased with the results. Also the first week of October has been great as well. I can email you a spreadsheet with my results if you like! I am still going to forward test until the end of October.

cheers
Dan


Sep Summary E/J E/U

Total Trades 237 285
Total Wins 143 170
Total Losses 94 115
Net 49 55"



Anyone who puts this much effort into something deserves to be taken seriously, hence the new EA. Dan, maybe you could post your spreadsheet here?



In a subsequent pm, Dan stated that he trades from a couple of hours into the Frankie session; the default trading hours are correct for me here in the UK. You will need to input your own local time that corresponds with Frankie start + a couple of hours.


In strategy tester from 6th July (EU M15), I found Dan's tp/sl of 10 was murdered; I tried the defaults out of curiosity; they make a killing. I haven't a clue what the best settings should be, is what I am saying here. Weekend strategy testers, remember to widen your MaxSpread input.



The inputs are the usual suspects; those of you new to my work will have to read around my threads and pick up one of the user guides that explains what they are and how they work - Alex at http://www.forexfactory.com/showthread.php?t=316936 is probably the best one to use. You can always stick the bot on a few charts and just watch for a while until the inputs start to make sense to you.


New inputs for TC:

 

1. LookBackCandles: allows you to customise the number of candles TC will check to ensure the previous candle had the highest close.
2. PendingPipsDistance: the distance from the trigger candle's hi-lo to set the pending trade.

TC makes full use of Stealth; turn it off with a HiddenPips input of 0.

TC will generate plenty of trades, so it will not be long before we know whether he has fit legs or not. Have fun.

Hanover
The Hanover module allows us to trade in the direction indicated by hanover's Recent Strength indi. I consider this to be the most significant development in my time at FF. Details in the Hanover thread athttp://www.forexfactory.com/showthread.php?t=319709. The Hanover check comes at trade-send time; so pending orders are set regardless of RS.










Matters of general interest



Read this post:
http://www.forexfactory.com/showthre...11#post4923611 Newbie traders, read this for inspiration and warning. Those of you who have suffered losses, read it for inspiration.
Masterly summary of hedging: http://www.forexfactory.com/showthre...79#post4977179

EA coding
I receive may requests to code EA's. Traders, if I code your EA for free, then I will share it here at FF. If you want me to code your EA without sharing, then my fee is $100 payable into my paypal account. I explain why at http://www.forexfactory.com/showthre...94#post4431794, at the bottom of the post.


EA coding
If I code your EA for free, then I will share it here at FF. If you want me to code your EA without sharing, then my fee is $100 payable into my paypal account.



Useful utilities/EA's/scripts

 

1. MaryJane has posted a fabulous utility to allow us to resize EA input windows: http://www.forexfactory.com/showthread.php?t=299017
2. Multi purpose trade manager: http://www.forexfactory.com/showthread.php?t=89371
3. Account equity analyser: http://www.forexfactory.com/showthre...16#post4693716 Also at http://www.forexfactory.com/showthre...98#post4709098, which includes a template for both USD and the Euro.
4. Past-trades display indi by squalou; shows a graphical display of the success/failure of trades in your History tab. Breathtaking piece of work. http://www.forexfactory.com/showthre...10#post3830610
5. My shell EA – the code I use to produce EA's so quickly. http://www.forexfactory.com/showthre...94#post4431794

Most promising trading EA's

 

1. Spider. I am trading this live. http://www.forexfactory.com/showthread.php?t=313680
2. BigBob: an implementation of Nanningbob V10. http://www.forexfactory.com/showthread.php?t=309980. The strategy is fantastic and I am having a ball trading a variation of it manually.
3. Aurora, a PwR-based EU scalper at http://www.forexfactory.com/showthread.php?t=312936 made a brilliant start.
4. Trendy: http://www.forexfactory.com/showthre...82#post4840182, but should be left behind by BigBob.
5. Reversi: http://www.forexfactory.com/showthread.php?t=294255 – I have traded this live.
6. Rip: http://www.forexfactory.com/showthre...98#post4777998
7. The Beast: http://www.forexfactory.com/showthre...=1#post4135863 A number of traders are still trading this live.

_________________________________________________________________________________
Posting restrictions
I have restricted the ability to post in this thread to: traders with at least 1 voucher; and traders on my 'buddy' list. I have done this because I do not want the thread cluttered with 'newbie' questions about the basics. New/inexperienced traders, this is not intended as a direct insult. What I want here is discussion with traders who fully understand what it is they are doing; when you newbs have been around a while, you will come to understand what a difference experience makes to the posts a member writes and shudder at the recollection of some of the questions you asked. I do. http://cdn.forexfactory.com/images/s..._the_floor.gif Go to your profile page; there is a link to the FF page that describes the voucher process.

Do not attempt to get around this by sending me pm's asking for help with your basic problems. Doing so will earn you an automatic ban from all of my threads.

I have set up a partner to my restricted threads at http://www.forexfactory.com/showthre...66#post3948666
There, you can post the questions that you cannot post here, about anything you need to know. Someone with the knowledge and experience you need will answer.

Traders/coders with experience who have not yet been vouched for, can get around this restriction be being added to my 'buddy' list; all you have to do is pm me to ask me to do this.

Brilliant! Thanks Steve for doing this!

For anyone who's interested I'll post a background to this system along with concise rules and a bunch of charts in the next day or so.

I intend to continue testing for the rest of October and then will dabble in the EA side of things. FWIW, EJ has had a net 7 wins and EU has had a net 17 wins for the first week of October

and here's the spreadsheet. Ive nicknamed the system Running Man but open to new suggestions.

Thanks Steve, You should have a web site such as Steve`s EA FACTORY.com. I gave up counting the number of EAs created by you long time ago.

Trumpcard, please post the info whenever you have the time, I am sure many will be interested to know more about this approach.

Kind regards

Keeping people up to date is the best way of engaging the FF Brains Trust. Thr BT includes some amazingly imaginative minds, so attracting their attention is well worth while doing.

For example, it is no coincidence that my two most recent posts at FF have followed those posted by astral. Guess why?

Sweet, more toys to play with....

Limited time as a new Term begins today in North-East China and another craptacular course has to be taught to lazy students...Nonetheless Forex Factory appears not to be blocked on my office computer nor in my apartment so away we go!

Here's a background to how I came up with the closing price strategy.

I am quite interested in breakouts especially on the 15 min timeframe. Over my time I've blown up heaps of pissy-sized accounts through being undisciplined, impatient and too greedy. I guess the keyy/dream for me has always been to have a little robot do all the trading for me with minimal supervision and slowly build wealth as pain-free as possible. I used to get excited at the prospect of compounding $200 200 times at a rate of 4%. I've learned the hard way that 4% is a little too high especially when one stops sticking to rules and uses a gambler's mentality!

My broker is GoMarkets Australia and thir charts are GMT +2. With all my dabbling in breakouts, trying this method and that, I've tended to focus on the 11am bar (9amGMT) as I observed that trends often would commence around this time. Not sure why this is, perhaps first few hours suggest indecision as London comes on board.

A few months ago I began focusing on closing prices a lot more and through in various indicators but could never really rely on the indicators and subsequently removed them.

A few weeks ago, I intensely focused on inside bar breakouts and came up with a method that had a terrible reward to risk (1:10) but was quite successful. However, when the inevitable 2 losses appear, so does a chunk of capital! This led me to take another approach and go back to closing prices. I focused on Cable and had a method on 5 min timeframe with a risk to reward of 1:2. I wanted something more robust so shifted to the 15 min with a risk to reward of 1:1 and chose E/J as the main pair.

After E/J's backtests proved successful, I examined E/U and found it to be even better. Manual backtesting has taken a long time (I examined each chart for September, annotating where each trade would have occured and the outcome). Thesystem works on 5 min as well BUT I don't have the resources to test it - takes forever to examine each chart.

Each chart I've looked at has been saved. I will continue doing thsi for the rest of October before even touching any EA's I guess it won't be a pure forward test as I have no intention or desire to sit for 12 hours staring at the screen waiting for each trade, rather, I'll examine the charts the next day. So perhaps it's a forward backtest.

Here is my system I have never seen this written down anywhere but it's so simple that I'm sure someone must trade similar to this.

In a nutshell, the system involves waiting for a trigger bar to form then placing a pending order based on that trigger bar.


Trigger Bar Identification (buy setup)

From 11am (that is the 15 min candle which begins at 11am on my chart), wait for a bar whose close is at least 1 pip higher than the 3 previous bars' closes.

If such a bar occurs, it will be the trigger bar.

The 11am bar's close can therefore be a trigger bar. If the current bar's close is lower by at least 1 pip than the bar before it but say 0.8 pips away form the close 2 bars ago, then it is not valid.

No more trigger bars are scanned for once the 11.45 bar closes. ie, the 11.45 bar is the last bar which can be a trigger bar for the day.

(pics to come)


Order Placement, Take Profit and Stop Loss

Place a pending buy order 1 pip above the high of the trigger bar. TP and SL are 10 pips.

Trade Management

IF the pending order isn't triggered on the next bar, it is cancelled and we wait for the next trigger bar to form.

HOWEVER, IF the bar(s) immediately following the trigger bar are inside bars, the pending order remains valid and then triggered on the subsequent bar or cancelled if no breakout occurs. If the inside bar (following the trigger bar) is followed immediately by another inside bar and so on, then the pending order remains until the breakout occurs on the next bar, otherwise the order is cancelled.

Only take 1 trade at a time. Once a trade closes, immediately look for the next trigger bar. This could be on the current bar of which the previous trade exited.

A suitable R:R is 1:1


Reverse everything for sells.

Here's a chart showing all the trades that would have occurred. The trigger bars are the ones with the purple line under them and the immediate tick or cross represents whether it was a win or loss.

I won't be able to upload any more pics until later today as they're all on my computer back home!

the beauty of this system is that it doesn't need any indicators to act as filters. the entry criteria and 1 pip breakout act as in-built filters. Therefore, it aims to capture small bags of pips while the market trends while automatically trying to stay out of choppiness.

Thanks for sharing your system Dan, and a big thanks to Steve for automating it.
Been testing it out today and so far it looks very promising....

Great to hear bobasker! Still remains a work in progress Good luck!

Update: Trades taken so far during London open,

3 BUYS

10 TP 10 SL = - 30 PIPS

15 TP 15 SL = + 30 PIPS -15 PIPS

20 TP 20 SL = + 60 PIPS

http://my.jetscreenshot.com/5637/201...9vxx-140kb.jpg

EU, Thu and Fri last week

here's a pic showing a trigger bar (bar 1) which happens to be an inside bar. It's close is lower than the 3 previous closes by at least 1 pip and therefore is valid. A sell stop would have been placed 1 pip below the low waiting for a breakout on the next bar. It did and 10 pips take profit was achieved. This bar (bar 2) then also closed lower than the 3 previous closes by at least 1 pip and therefore it became a valid trigger bar and a sell stop would have been placed 1 pip below the low. However, the sell stop was not activated but the order isn't cancelled because the bar immediately following the trigger bar formed an inside bar. The next bar in the sequence then broke bar 2's low but got stopped out at the bar with the thumbs down above it.

before I go and eat dinner, I just spotted a current live trigger bar on E/U. Let's see if it breaks out. If not then we would wait for the next trigger bar to form and any order would meanwhile be cancelled.

last trigger bar was followed by an inside bar. price broke down and actually formed a close lower tthan the 3 previous closes so a new Trigger Bar is in play.

the low of the current Trigger Bar is 1.35696 so following the system rules, a sell stop would be placed at 1.35686 with take profit 1.35586 and stop loss of 1.35786

breakout didn't occur so order would be cancelled and wait for next Trigger Bar

I welcome all feedback!

I see some folks using variable TP and SL values based on the session and the current volatility.

It'd be interesting to try to automate this using ATR with the Missing SL/TP function of mptm. I'll elaborate a setfile for this and set up a demo for this.