Jollyroger: an inside bar auto-trader by SteveHopwood | Page 7 | Forex Factory

Options

Go to
Permalinks
Print Thread

Similar Threads

3 Bar Fractal with mid-bar and bar on each side 11 replies

Inside bar or outside bar identifier 7 replies

SteveHopwood EAs 4 replies

Pin bar, pin bar, pin bar 0 replies

What's Wrong with this Inside Bar Indicator 1 reply

Attachments: Jollyroger: an inside bar auto-trader by SteveHopwood
Exit Attachments

Jollyroger: an inside bar auto-trader by SteveHopwood

  • #121
  • Jun 7, 2011 11:12am Jun 7, 2011 11:12am
Quoting hankeys
i am thinking about the INSIDE BARS and trading this system:
1) wait for the inside bar on H1
2) count the difference between high and low of the inside bar and divide be 2 - in other words get the middle of the inside bar
3) if the close of the inside bar is above the middle - put buy stop on high of inside bar
4) if the close of the inside bar is under the middle - put sell stop on low of inside bar
5) if the close of the inside bar is on the middle (+- 5 pips) put both buy and sell stop on high, low
6) if the stop orders is not hit on close of...
before somebody puts effort to write EA, can you convince him/her why this system would be profitable ? how you came up with these rules ? i suggest you take daily chart of EU, GU and do a study of this and log the trades in excel and post the results. if its convincing i will write the EA. clearly specify the exit rules also in the excel
  • #122
  • Jun 8, 2011 1:00am Jun 8, 2011 1:00am
dont want to interupt with the thread theme.
i've tried to manipulate rsi multipairs system, into 1h momentum system.
would some one mind to take a look at the code and make it work?
thanks in advance.
Attached File(s)
File Type: mq4 1h_multipairs .mq4   10 KB | 321 downloads
File Type: ex4 1h_multipairs .ex4   8 KB | 283 downloads
File Type: ex4 rsi multipairs.ex4.ex4   17 KB | 314 downloads
  • #123
  • Jun 23, 2011 8:57pm Jun 23, 2011 8:57pm
Good method. Will give it a try for few weeks
  • #124
  • Jul 4, 2011 5:15pm Jul 4, 2011 5:15pm
Not much is happening on this thread but I am putting up a few numbers for those of interest. During the month of June I netted 256 pips trading the latest Dibs EA. The one prior to the latest generated a loss 340 pips. I trade on 4 pairs, EUR/USD, USD/CHF, GBP/USD and USD/CHF. Obviously the latest version is the one I am currently trading.

I had been trading both EA's using the daily open at 0 GMT rather than the recommended 5 GMT. For the month of July I have switched to the 5 GMT start of trading. I am using the jumping stop loss with a .06 initial order.

There is a problem with the EA in that it is very selective on what IB's it chooses to recognize and place pending orders on. I would estimate it picks up less than 50% of the IB's generated by PA. If this could be corrected by Steve or another competent coder I am sure the results would improve dramatically. If this is enabled a maximum order limit tab would be advisable to avoid over extending the account (read "blowing out"). Also it would be desirable to enable the EA to be used on other timeframes (higher).

I believe that this is a viable trading method and intend to continue to forward test on a demo account. Unfortunately I do not code.
  • #125
  • Jul 4, 2011 6:08pm Jul 4, 2011 6:08pm
Hi Jollyroger,

Thanks for reviving the thread, I had forgotten all about it.

As for the 50% of IB's this EA does not pick up, can you please give me an example as soon as you see one next?

Also, are you using the EA in this post?
  • #126
  • Jul 4, 2011 7:05pm Jul 4, 2011 7:05pm
Quoting the-game
Hi Jollyroger,

Quote
Thanks for reviving the thread, I had forgotten all about it.

As for the 50% of IB's this EA does not pick up, can you please give me an example as soon as you see one next?
Also, are you using the EA in this post?
Hi Game:
I will try to upload a current chart that illustrates the point on the IB's. Two IB's are shown and the magenta lines in the previous day are also IB's.

Yes I am using the latest Dibs EA as referenced in your post. Hopefully you can both see the problem and come up with a solution Another problem that we are all aware of is the refusal of the criminals to place stops and/or pending orders too close to the current price but I don't know if anything can be done about that. Not being able to code and barely computer literate is a royal pain in the butt when you are trying to develop effective Forex trading strategies particularly scripts and EAs. If it wasn't for people like yourself, Steve and others I would be stuck behind a computer using manual trading for a lot longer than I want to be.
Roger
Attached Image (click to enlarge)
Click to Enlarge Name: chart3.gif Size: 56 KB
  • #127
  • Jul 4, 2011 10:00pm Jul 4, 2011 10:00pm
Can you please make it more obvious which candles were IB's, but the EA had ignored?
  • #128
  • Jul 5, 2011 1:30am Jul 5, 2011 1:30am
Quoting the-game
Can you please make it more obvious which candles were IB's, but the EA had ignored?
Attached is a chart that is clearer. You will see one IB with a pending order. There were no pending orders from the preceding day and the IB's are clearly marked by the bot. The magenta bars are an indicator I use to delineate IB's in case the bot overlooks them. Hope this helps.
Roger
Attached Image (click to enlarge)
Click to Enlarge Name: dib.4.gif Size: 16 KB
  • #129
  • Jul 5, 2011 3:52am Jul 5, 2011 3:52am
Jollyroger, i have been using a similar bot for some time but haven't really been able to come out of breakeven with it. Two questions:

- Is the starting time more important than the general trend? I have been filtering the trades with a bigger timeframe ma instead of the day open. I noticed quite some days where a day goes 100 pips down with general uptrend, forms an ib at the very bottom, and then goes north again. Your missing al these trades then.
- How do you trail it? Seems like one of the bigger problems here. Most IB's seems to go good, around 50-100 pips. After that price retraces, a lot most of the time. If i trail to agressive i wont be left with a lot of pips, if i trail conservative then the sl will be hit too many times.

Wondering your experience on this one
  • #130
  • Jul 5, 2011 4:02am Jul 5, 2011 4:02am
i trade similar system, its giving me good results. but its discretionary system, thats why i dont use EA

1. trade only in the direction of 200 MA. TF is H1

2. trade only during EU session and US morning session

3. on breakout of Inside Bar

4. the Inside Bar must also be a narrow range bar. i use NR7. That means narrowest in last 7 candles.

5. breakout must occur in next bar
  • #131
  • Jul 5, 2011 9:05am Jul 5, 2011 9:05am
Quote
- Is the starting time more important than the general trend? I have been filtering the trades with a bigger timeframe ma instead of the day open. I noticed quite some days where a day goes 100 pips down with general uptrend, forms an ib at the very bottom, and then goes north again. Your missing al these trades then.
- How do you trail it? Seems like one of the bigger problems here. Most IB's seems to go good, around 50-100 pips. After that price retraces, a lot most of the time. If i trail to agressive i wont be left with a lot of pips, if i trail...

You have raised two very key questions to which I unfortunately have no definitive answer. Currently I am trading according to the rules outlined on page 1 of this thread using a jump stop as built into the bot.

I am beginning to have doubts about whether this system can be automated using an EA for exactly the reasons you outlined in your questions. The rules of the system call for taking the trade based on the market direction at 0500 GMT REGARDLESS OF THE GENERAL DIRECTION OF THE EXISTING LONG TERM TREND. This often violates the axiom of trading with the long term trend but complies with a short term trend that has been established for that trading day. (Perhaps a solution would be to code into the bot a requirement that the trade should be taken only if the daily trend established matches the long term trend). In manual trading this would be a judgement call.

The other point mentioned regarding the SL is also open to discussion for the reasons mentioned in your question. Steve mentions in the thread that when the bot is tested satisfactorily his intention is to open a new thread to discuss this exact subject. Right now I am thinking of changing my SL method to a MA SL.

If you haven't already done so I suggest that you read the thread and definately use the links provided by Murphyman in Post 7 where you will find positive comments on both the DIBS system and trading in general.

Sorry for the long post but hope it helps.

Roger
  • #132
  • Jul 5, 2011 1:20pm Jul 5, 2011 1:20pm
Thanks for the awnser jollyroger. I did read the most important posts of the thread starter, but couldn't bother to read the 600 pages in the thread.

What did get me most was the message from the engineer who intensivly looked at this system and came to the conclusions that:

- 6 GTM did work as the trend direction, without any other instrument (there was also another hour which seem to do the job, but he didn't tell)
- More or less any trailing method would lead to the same results
- Profit factor increased from 1.21 to 1.27 if the close half @ r:r 1:1 wasn't used.

When someone spend a lot of money investigating these kind of things statistically , i tend to believe it quite fast. So basically, leaving this standard idea running for a couple of months in some corner of a vps should lead to approx. 1.2 profit factor.

Didn't get those results yet, but you did seem to get it. Could become interesting but i'd like to leave it running for another couple of weeks and see what it does.
  • #133
  • Jul 5, 2011 3:12pm Jul 5, 2011 3:12pm
-

Quote
Didn't get those results yet, but you did seem to get it. Could become interesting but i'd like to leave it running for another couple of weeks and see what it does.
[/quote]

Sounds like a plan. Please let us know of your conclusions.
  • #134
  • Jul 5, 2011 6:54pm Jul 5, 2011 6:54pm
I tested this EA for about three weeks and it never missed any trades for me. My broker doesn't have any restrictions on placing pending orders though, so that may have something to do with it.

I tried the 21 period moving average for a trailing stop and it worked to some degree, but only about break even over the three weeks. I suspect the original method of closing half of the position at break even, not moving the stop loss and closing out winners manually would work best in the long term. There would be lots of small losses before a huge winner. Personally I'm not patient enough for that style of trading, and prefer not to have any discretion involved, so have retired the EA.
  • #135
  • Jul 5, 2011 7:17pm Jul 5, 2011 7:17pm
[quote]
Quoting skoda2008
I tested this EA for about three weeks and it never missed any trades for me. My broker doesn't have any restrictions on placing pending orders though, so that may have something to do with it.
Thanks Skoda for your input. I would be most interested in the name of your broker if you care to reveal it. Thanks.
Roger
  • #136
  • Jul 5, 2011 7:53pm Jul 5, 2011 7:53pm
Jolly Roger, have a look at page 10 of the Reversi thread.
  • #137
  • Jul 21, 2011 1:18pm Jul 21, 2011 1:18pm
Quote
Sounds like a plan. Please let us know of your conclusions.

So far so good. The equity line of the bot is going really south, if you want to become profitable with this bot you have to close the trades manually. Entries are good, just dont know about the exits yet. For now i just let the bot run and every now and then i close the big profits. Not really scientific method but works for now.

Trading 0.01 on an euro account (demo), first went down 60 euros, then started closing some and went up to + 40. Bot retraced it back to about 0, just closed some trades now and i'm up 60 euro's.

Pretty messed up zigzag balance line, but if i can manage to make the tops higher then that's all we need

Dunno if it's a bug by the way but i'm sitting on 2 very negative USDCHF trades, clearly went past their sl's.

How's yours going jollyroger?
  • #138
  • Jul 21, 2011 6:51pm Jul 21, 2011 6:51pm
How's yours going jollyroger?[/quote]

I'm not really using this anymore. There are bugs in the code and I don't code so I am mostly trading manually. The DIBS code works but the way it is currently set up has not been profitable for me. Mostly a break even or small loss. When I have more time I will play with it manually but right now I am working on something else that requires most of my attention.

This is not my thread to abandon but I am unsubscribing. Thanks to Steve and all the others who have contributed toward trying to make this a profitable EA.

Roger
  • #139
  • Jul 21, 2011 7:02pm Jul 21, 2011 7:02pm
Back tested it on GPB/USD on H1 from 2009 to 2011 with bad results?

So what do??
  • #140
  • Sep 5, 2011 9:36am Sep 5, 2011 9:36am
Latest update in post 1 guys. See the 5th Sept edit for details.

0 traders viewing now