scooby-doo BOJ JPY pairs trading robot | Page 385 | Forex Factory

Options

Go to
Permalinks
Print Thread

Similar Threads

Trading correlation pairs by using the other pairs 843 replies

Export All pairs or selected pairs to .csv with script 3 replies

Pairs of Currency Pairs 4 replies

Trending Pairs / Ranging Pairs 0 replies

Robot trading 120 replies

Attachments: scooby-doo BOJ JPY pairs trading robot
Exit Attachments

scooby-doo BOJ JPY pairs trading robot

  • #7,681
  • Mar 20, 2010 6:41am Mar 20, 2010 6:41am
Quoting mbkennel
Just so people remember: THM 1.6x does NOT include any news filter.
Many thanks for your hard work on the makrel!
I was just wondering why your version does not include the news filter from Scoobs THG 1.6?
Would you anticipate any adverse reactions by including it in the Makrel code?
The best trade I ever made was my SIN for Jesus' RIGHTEOUSNESS.
  • #7,682
  • Mar 20, 2010 6:53am Mar 20, 2010 6:53am
Quoting camvcvoo
Hi Dryclean,

Have been following the link of yours:

http://www.myfxbook.com/members/dryc...ean-fxdd/18028

Here's mine:

https://www.myfxbook.com/portfolio/young/20610

As you can see, we have quite a different result. Of course, mine is worse. Your result is encouraging.

I have done a trade activity comparison between yours and mine in excel.

Do you think you can provide your set file for the EA? I really like to use your setting for my live account to achieve similar result as yours.

I'm using version 2.4, no...
Various past posts have shown there are large differences in broker feeds, even differences with the same broker but different servers.

So short of using the same broker AND the same server you are never going to get the similar results ... IMHO
  • #7,683
  • Mar 20, 2010 8:32am Mar 20, 2010 8:32am
Quoting macman
Various past posts have shown there are large differences in broker feeds, even differences with the same broker but different servers.

So short of using the same broker AND the same server you are never going to get the similar results ... IMHO
macman,

Thank you for reminding. I'm aware of what you are saying. I just want to find out what Dryclean is using that are different from mine.

福 子 Victor V
  • #7,684
  • Mar 20, 2010 11:46am Mar 20, 2010 11:46am
Quoting jimmynoodles
Many thanks for your hard work on the makrel!
I was just wondering why your version does not include the news filter from Scoobs THG 1.6?
Would you anticipate any adverse reactions by including it in the Makrel code?
a) I had some plan of a rewrite of the news filter which now is complex & heavy inside an external custom indicator. And i'm not sure that other thing even works right.

b) at some point scoobs took off news filter in his secret workshop robots
  • #7,685
  • Mar 20, 2010 3:08pm Mar 20, 2010 3:08pm
Quoting mbkennel
a) I had some plan of a rewrite of the news filter which now is complex & heavy inside an external custom indicator. And i'm not sure that other thing even works right.

b) at some point scoobs took off news filter in his secret workshop robots
Ok thanks for the reply.
The best trade I ever made was my SIN for Jesus' RIGHTEOUSNESS.
  • #7,686
  • Mar 20, 2010 9:20pm Mar 20, 2010 9:20pm
Not sure what you all will think about this, but I will throw this out for discussion:

I was thinking further about MiddleEastern's comment about how a number of trades just dive straight to the SL. So what I wondered is this: If the trade is already halfway to the SL and the RSI is still extreme, then rather than waiting for it to crash right into the hard SL, why not close it right then and there?

For example, RSI on H1 has to be < 20 to trigger a buy trade. The SL is 4 x M15 ATR. So we put in a buy trade. But the price just keeps on going dow strongly. The price is then halfway to the SL (2 x M15 ATR) and now is < 10, so rather than hope that it might turn around, we assume that the trade is doomed and just getting worse, so we close it there.

 

  1. For buy trade: If price is more than 2x ATR from original buy towards the SL, and the H1 RSI < 10, then close trade.
  2. For sell trade: If price is more than 2x ATR from original buy towards the SL, and the H1 RSI > 90, then close the trade.

I hope this makes sense. What do you think?

  • #7,687
  • Mar 20, 2010 9:31pm Mar 20, 2010 9:31pm
I played with this in backtester a bit the other day (obviously it will only trade the pair you backtest on, not the others), and I noticed the 1:2 RR in the code as well. Didn't like that, so i modified it to be 3:1 instead.
But that got me thinking on another note, having a trailing stop based on atr instead. So stop could be 1 or 2x ATR, but then instead of a hard target, its a trailing stop of 3x atr.
Dunno if it would work out or not, I couldnt play with it much to find out because there is no trailing stop function coded into the ea as is.
I think I am over stops being larger than targets though, but that's just me
  • #7,688
  • Edited 10:10pm Mar 20, 2010 10:08pm | Edited 10:10pm
Quoting Invisible
Not sure what you all will think about this, but I will throw this out for discussion:

I was thinking further about MiddleEastern's comment about how a number of trades just dive straight to the SL. So what I wondered is this: If the trade is already halfway to the SL and the RSI is still extreme, then rather than waiting for it to crash right into the hard SL, why not close it right then and there?

For example, RSI on H1 has to be < 20 to trigger a buy trade. The SL is 4 x M15 ATR. So we put in a buy trade. But the price just keeps on going...
I like this. But if you still have the trend and rsi signal ( <20) to trigger a Buy wouldn't the code just open a new Buy order for the pair? So you'd have to kill the trade and prevent new buy trades for that pair for say an hour?

Sandy
  • #7,689
  • Mar 21, 2010 2:08am Mar 21, 2010 2:08am
actually, Invisible is suggesting something that is similar to my idea mentioned before. and many others mentioned from different angles in fact.

we need a function or subroutine to "Inspect Price Action".

so price jumping off cliff? we shorten SL, and hold on trading.
if price flying like a rocket, we move tp further away.

however, these are very easily said, but very difficult to code. again, very difficult to code, at least many many conditions to consider, even when our human eyes only see price move rapidly or slowly.

coding such a routine will definitely help ANY EA, not only this one...... just it is something as great/time-consuming as Steve's MPTM.

but before we having it, we'd better focus on if the basic HG v1.6~2.x 's idea is truly profitable. 1 or 2 more months, we'll see...


Quoting nubchai
I like this. But if you still have the trend and rsi signal ( <20) to trigger a Buy wouldn't the code just open a new Buy order for the pair? So you'd have to kill the trade and prevent new buy trades for that pair for say an hour?

Sandy
  • #7,690
  • Mar 21, 2010 4:56am Mar 21, 2010 4:56am
The main reason why < 20 > 80 systems tend to fail is because you are actually buying weakness and selling strength.

Just because RSI is > 80 does not mean it wont go on to 99.99 and stay there for long period of time.

This is why sometimes it works and sometimes it fails.

Scoobs.
  • #7,691
  • Mar 21, 2010 6:31am Mar 21, 2010 6:31am
Quoting octomanoak
Thanks Macman. I would like to give it a try, but once I started looking at the mass of settings for MPTM I realized I don't know how to set it up with what you posted. I need to study the user guide for sure. But if you would consider posting your MPTM set file you use with THG (or PMing to me if you prefer) it will greatly facilitate my coming to understand not only your settings but MPTM.

JR
I understand your reasoning behind asking for a pre-set file, but there are several reasons why I think it is not a good idea.

1. The settings are totally different for 4 & 5 digit brokers
2. The partial close setting refers to the initial lot size, also different for every user

So to avoid confusion, I think it best you play with mptm on demo until you get that light bulb moment.

Just to recap, the settings I am using which are purely a guess are:

BE @ +15 with +2 profit buffer
Jump stop @ +17,5 with +2 profit buffer
Jump stop after BE set to 'false'
Partial close enabled (1/2 position closed at +15)

It would be nice if others can maybe test with different exit settings

Best of luck -
  • #7,692
  • Mar 21, 2010 6:44am Mar 21, 2010 6:44am
Quoting scooby-doo
The main reason why < 20 > 80 systems tend to fail is because you are actually buying weakness and selling strength.

Just because RSI is > 80 does not mean it wont go on to 99.99 and stay there for long period of time.

This is why sometimes it works and sometimes it fails.

Scoobs.
Then will something like the KG robot setup not work.. The primer is for the RSI to go below X (20) then the trigger is for the RSI to go above Y (25) for a minimum of 5 mins.
  • #7,693
  • Mar 21, 2010 7:05am Mar 21, 2010 7:05am
Quoting sundog
Then will something like the KG robot setup not work.. The primer is for the RSI to go below X (20) then the trigger is for the RSI to go above Y (25) for a minimum of 5 mins.
It will work, indeed does work, so long as you have the lot sizes and patience to ride out the periods when the trades go against you. I have come to the conclusion that this happens too often and will be replacing the KG robot as soon as I have something that proves to be better. There are a number of candidates here and in the BS thread, so this time is not far off.

  • #7,694
  • Mar 21, 2010 8:35am Mar 21, 2010 8:35am
Quoting charvo
actually, Invisible is suggesting something that is similar to my idea mentioned before. and many others mentioned from different angles in fact.

we need a function or subroutine to "Inspect Price Action".

so price jumping off cliff? we shorten SL, and hold on trading.
if price flying like a rocket, we move tp further away.

however, these are very easily said, but very difficult to code. again, very difficult to code, at least many many conditions to consider, even when our human eyes only see price move rapidly or slowly.

coding...
What I like about Invisible's idea is that when the PA continues going north or south after the RSI level has been hit, if you shorten the SL and prevent trades in the same direction for a certain number of minutes then this is protection against run away trades, extreme trend reversals due to hedge fund or bank actions etc etc. It does sound hard to program.

Sandy
  • #7,695
  • Mar 21, 2010 8:38am Mar 21, 2010 8:38am
While there is no scheduled MAJOR news releases on Monday becareful trading USD pairs.

If Obama gets his health reforms through then the Dollar will rise. If he doesnt then it will fall.

Scoobs.
  • #7,696
  • Mar 21, 2010 8:51am Mar 21, 2010 8:51am
I'm using version 2.4, no hedging, both buy and sell close upon expiry 300 minutes later, Wave 2 turn off.

My email is [email protected]

Thank you in advance.
[/quote]

Thanks for the inquire
I'm using FXDD.. The time stamp is so variable... it's crazy. There is my local time GMT-5 DST there is my broker time GMT+2 but they are based in GMT-5 DST???. Then myfxbook has it's own time. So when the bot places a trade lets say at 00:00 broker time who has the time to calculate what time its gonna put where.....the only thing important is the correct time in the bot.

Like I said I'm GMT-5 DST so now in the bot I'm 18:00 and 2:00 was 17:00 1:00 before DST time

I also have the profit multiplier set from 200 to 250
I also set short trade true and long trade false for the 1st trade
and long trade true short trade false for the 2nd trade and I have no time
limits, also Hedging off. My 2nd wave longs have been hitting TP way faster then the 1st wave short.

I'm tempted to adjust the lot size and multiplier for the 2nd wave but for now I'll just sit back and be a nice little pig and collect pips

My only 2 losing baskets have been from the time limit being on and a long basket for the 1st wave. The time limit one was a small loss but with in 30 min more it would have hit TP. The long basket one was the killer it hit 10% SL

The only thing now with the bot is I have to manually go in and change the first wave to .....short true and long false.... then just before I go to bed and hopefully the 1st wave has hit TP I have to reverse it and say
.......short false and long true.

Wish there was a way to make that automatic

But last week was a very nice trading week
  • #7,697
  • Mar 21, 2010 9:03am Mar 21, 2010 9:03am
Quoting scooby-doo
While there is no scheduled MAJOR news releases on Monday becareful trading USD pairs.

If Obama gets his health reforms through then the Dollar will rise. If he doesnt then it will fall.

Scoobs.
I would expect it the other way.(Dollar fall if passed) 70% of the public and probably 100% of businesses is against this Govt take over of our health care. Yes we need an overhaul but this bill is an abortion..... So why would the market approve of it..... The only reason I could see the Dollar rise from this is some type of stability being seen here in the States
although in the long run we are F-ed.
  • #7,698
  • Mar 21, 2010 9:12am Mar 21, 2010 9:12am
However, per your entry conditions--"trend" is identified first by the MAs, therefore, you enter the major trend direction, unless you come across a "trend" reversal, the trade would be ok.

usually, trend reversal happen less, so SL probability is less than TP probability (assume sl:tp 1:1).

my dumb guess...

Quoting scooby-doo
The main reason why < 20 > 80 systems tend to fail is because you are actually buying weakness and selling strength.

Just because RSI is > 80 does not mean it wont go on to 99.99 and stay there for long period of time.

This is why sometimes it works and sometimes it fails.
  • #7,699
  • Mar 21, 2010 9:23am Mar 21, 2010 9:23am
Quoting dryclean
I would expect it the other way.(Dollar fall if passed) 70% of the public and probably 100% of businesses is against this Govt take over of our health care. Yes we need an overhaul but this bill is an abortion..... So why would the market approve of it..... The only reason I could see the Dollar rise from this is some type of stability being seen here in the States
although in the long run we are F-ed.
The markets/money men don't care about PUBLIC sentiments.

You are correct that they want stability opposed to will it/wont it pass regardless of its actual content/merits.

They are looking for a clear direction.

Scoobs.
  • #7,700
  • Mar 21, 2010 10:13am Mar 21, 2010 10:13am
Hi all

For 3+ weeks I have run this EA without making any modifications and have not had a losing week yet.

Attached are the specifics for analysis. More later...

T...
Attached Image (click to enlarge)
Click to Enlarge Name: The Holy Grail DetailedStatement thru 0319.gif Size: 6 KB
7 traders viewing now