Weekly 80/20 Setups | Page 23 | Forex Factory

Options

Go to
Permalinks
Print Thread

Similar Threads

Trading Systems or Trading Setups 9 replies

Brainstorming Session: OctaveJay's Weekly n Daily Trade Setups 61 replies

Intraday setups with Weekly TF 30 replies

High Probability Trading Setups for the Currency Market 6 replies

Current Weekly Price Action Setups 1 reply

Attachments: Weekly 80/20 Setups
Exit Attachments

Weekly 80/20 Setups

  • #441
  • Aug 15, 2008 11:18pm Aug 15, 2008 11:18pm
Fj,

I am not convinced that a hedge would be the answer, but I only brought it up for discussion, because others might have a lot of experience handling these things.

on the other hand it complicates the system considerably and in the end the system might do very well without the hedge.

was just wondering if the system does very well in normal times, but might give back (almost) everything in adverse conditions.

Fj, having the most experience, could probably have a clue as to what is possible after the USD saga has been played out.
  • #442
  • Aug 16, 2008 7:51am Aug 16, 2008 7:51am
Quoting yonnie
Fj,

I am not convinced that a hedge would be the answer, but I only brought it up for discussion, because others might have a lot of experience handling these things.

on the other hand it complicates the system considerably and in the end the system might do very well without the hedge.

was just wondering if the system does very well in normal times, but might give back (almost) everything in adverse conditions.

Fj, having the most experience, could probably have a clue as to what is possible after the USD saga has been played out.
I wouldn't say I have the most experience. I've only been trading live since February. I have learned two things from the "USD saga". One is to not get greedy with return expectations by clamping down on the amount of capital being risked per trade. This system needs a lot of "wiggle room". The next is to not be overly exposed to one single currency. The USD was involved in 13 of the 26 currencies I traded. I got caught in a situation where half my open trades were long the USD. This was the first time I manually intervened in the system to prevent further capital loss. I've adjusted how I trade the system (less risk and less single currency exposure), but have not adjusted the system. The lessons have been learned and I won't repeat those mistakes again.
  • #443
  • Aug 16, 2008 8:40am Aug 16, 2008 8:40am
Hi all,

Yonnie, sorry I didn’t reply yet, I’ve been busy. I’m using GMT time.
For backtesting I’m using five years of hourly data, freely available at Dukascopy, for the same ten pairs FlyJetz last choose, from that data I grab the high and low for that day and the time they happened.
To make backtesting easyer I decided to use daily timeframe and close all trades at the end of the day, this way the system is totally mechanical.
About 2 months of testing show that this kind of trading is profitable with any kind of settings, some perform better than others and some work better on some pairs than others, but the differences aren’t very significant.
After that I get more or less the same results that Fj, Math and you, about 65% success, then I group the trades by the hour that the pending order is executed, and the results show that on about 80% of the winning trades the pending order is executed in the first half of the day (12:00 GMT), after that time the loosing trades are greater then the winning trades.
After implement that new condition in every test, the success rate increased to about 80% and the average winning trade is greater than the average loosing trade. On some pairs I used ten years of hourly data with same results. I did one test using weekly timeframe and get almost the same results. I didn’t try to get the best performing settings to avoid curve fitting. It seems that this system performs well in all market conditions.
On five years test using data from the same source for the ten pairs I get more than 190K pips, there are some days with huge drawdown’s (more then 1K pips), but they are recovered in the next 2 or 3 days, that happened about 3 to 4 times in five years.
Right now I’m looking to get an EA for backtesting on MT platform to confirm my findings and forwardtesting, because I’m not very disciplined and I hate to loose, so when I see a loosing trade I usually don’t close it, and hijack all the testing. Any help is welcome.

Regards SJ
  • #444
  • Aug 16, 2008 2:46pm Aug 16, 2008 2:46pm
Quoting sjbelchior
Hi all,

Yonnie, sorry I didn’t reply yet, I’ve been busy. I’m using GMT time.
For backtesting I’m using five years of hourly data, freely available at Dukascopy, for the same ten pairs FlyJetz last choose, from that data I grab the high and low for that day and the time they happened.
To make backtesting easyer I decided to use daily timeframe and close all trades at the end of the day, this way the system is totally mechanical.
About 2 months of testing show that this kind of trading is profitable with any kind of settings, some perform better than others and some work better on some pairs than others, but the differences aren’t very significant.
After that I get more or less the same results that Fj, Math and you, about 65% success, then I group the trades by the hour that the pending order is executed, and the results show that on about 80% of the winning trades the pending order is executed in the first half of the day (12:00 GMT), after that time the loosing trades are greater then the winning trades.
After implement that new condition in every test, the success rate increased to about 80% and the average winning trade is greater than the average loosing trade. On some pairs I used ten years of hourly data with same results. I did one test using weekly timeframe and get almost the same results. I didn’t try to get the best performing settings to avoid curve fitting. It seems that this system performs well in all market conditions.
On five years test using data from the same source for the ten pairs I get more than 190K pips, there are some days with huge drawdown’s (more then 1K pips), but they are recovered in the next 2 or 3 days, that happened about 3 to 4 times in five years.
Right now I’m looking to get an EA for backtesting on MT platform to confirm my findings and forwardtesting, because I’m not very disciplined and I hate to loose, so when I see a loosing trade I usually don’t close it, and hijack all the testing. Any help is welcome.

Regards SJ
SJ

Very impressive work!! I've always wanted to experiment with this system on a daily interval, but knew the manual work would exceed my time available. Can you give me an idea of how many backtested trades your version of this system looked at? If you do develop an EA, let me know. I've never used EA's before, but if what you say is true, it may be worth exploring. I know there were some earlier posters on this thread that looked at developing EAs. You'll have to go back and do some research to find them though.

FJ
  • #445
  • Aug 16, 2008 7:11pm Aug 16, 2008 7:11pm
Sj,

thank you for your contribution.

some questions:
how do you calculate the midpoint MP for the daily system?

direction was different as well: what is different?

did you backtest trading with the trend or going long and short the same pair?

what settings did you use for your backtests and what were the results?

thanks
  • #446
  • Aug 16, 2008 9:52pm Aug 16, 2008 9:52pm
Great system with lots of strengths. minimal screen time and the effect of spread is negligable. I love the way it always closes losing trades after a favourable move, so never at the bottom or top of a move against you.
Quoting Flyjetz
Closed Trade #242
22-Jul-08 Short EUR/CZK 23.28 12-Aug-08 23.927 -6470
This huge loss was an unusual trend reversal which caught your system out. Have you thought of anyways to avoid such a disaster in the future? mayby a maximum loss cut off or reducing the weekly target to 50% after a trade moves into a loss. I would just like to see this loss reduced to keep your stats better and your system more able to absorb a serious of trades like this should that ever occur.
BTC Return This Year: na
  • #447
  • Aug 17, 2008 1:16am Aug 17, 2008 1:16am
Quoting M.A.C.Doug
Great system with lots of strengths. minimal screen time and the effect of spread is negligable. I love the way it always closes losing trades after a favourable move, so never at the bottom or top of a move against you.

This huge loss was an unusual trend reversal which caught your system out. Have you thought of anyways to avoid such a disaster in the future? mayby a maximum loss cut off or reducing the weekly target to 50% after a trade moves into a loss. I would just like to see this loss reduced to keep your stats better and your system more able to absorb a serious of trades like this should that ever occur.
It's not as big of a loss as it appears. 23.28 to 23.927 is a loss of 2.78%. If this same movement were to happen on a Long trade on the Euro from 1.50, you would close out at 1.4583. Or a loss of 417 pips. Pips (for comparison's sake) really are meaningless when you get into the Exotics. That is why I separate out the statistics by Majors, Exotics, and Metals.
  • #448
  • Aug 17, 2008 2:00am Aug 17, 2008 2:00am
Quoting sjbelchior
Hi all,

Yonnie, sorry I didn’t reply yet, I’ve been busy. I’m using GMT time.
For backtesting I’m using five years of hourly data, freely available at Dukascopy, for the same ten pairs FlyJetz last choose, from that data I grab the high and low for that day and the time they happened.
To make backtesting easyer I decided to use daily timeframe and close all trades at the end of the day, this way the system is totally mechanical.
About 2 months of testing show that this kind of trading is profitable with any kind of settings, some perform better than others and some work better on some pairs than others, but the differences aren’t very significant.
After that I get more or less the same results that Fj, Math and you, about 65% success, then I group the trades by the hour that the pending order is executed, and the results show that on about 80% of the winning trades the pending order is executed in the first half of the day (12:00 GMT), after that time the loosing trades are greater then the winning trades.Regards SJ
SJ,

In Fj`s system triggered trades are not closed at the end of the week, but stay open into the next week.
In that case there`s no need to cancel all open untriggered orders by Wednesday.

Or are you saying that it is more profitable to close all triggered trades at the close of the week and remove all untriggered orders on Wednesday?

thanks for your info
  • #449
  • Aug 17, 2008 2:22am Aug 17, 2008 2:22am
Quoting sjbelchior
Hi all,

Yonnie, sorry I didn’t reply yet, I’ve been busy. I’m using GMT time.
For backtesting I’m using five years of hourly data, freely available at Dukascopy, for the same ten pairs FlyJetz last choose, from that data I grab the high and low for that day and the time they happened.
To make backtesting easyer I decided to use daily timeframe and close all trades at the end of the day, this way the system is totally mechanical.
About 2 months of testing show that this kind of trading is profitable with any kind of settings, some perform better than others and some work better on some pairs than others, but the differences aren’t very significant.
After that I get more or less the same results that Fj, Math and you, about 65% success, then I group the trades by the hour that the pending order is executed, and the results show that on about 80% of the winning trades the pending order is executed in the first half of the day (12:00 GMT), after that time the loosing trades are greater then the winning trades.
After implement that new condition in every test, the success rate increased to about 80% and the average winning trade is greater than the average loosing trade. On some pairs I used ten years of hourly data with same results. I did one test using weekly timeframe and get almost the same results. I didn’t try to get the best performing settings to avoid curve fitting. It seems that this system performs well in all market conditions.
On five years test using data from the same source for the ten pairs I get more than 190K pips, there are some days with huge drawdown’s (more then 1K pips), but they are recovered in the next 2 or 3 days, that happened about 3 to 4 times in five years.
Right now I’m looking to get an EA for backtesting on MT platform to confirm my findings and forwardtesting, because I’m not very disciplined and I hate to loose, so when I see a loosing trade I usually don’t close it, and hijack all the testing. Any help is welcome.

Regards SJ
Thanks SJ for your new method of 8020. Could you post the rule of your new ideas?

I am thinking method to hedge the DD of current 8020 too. Some breakout strategies e.g. Newstrike or AshFX Daily might help.
  • #450
  • Aug 17, 2008 2:25am Aug 17, 2008 2:25am
Yes Fj, you are the one with the most experience trading this system since Feb., or did you get this system from somewhere else?

by trading LONG + SHORT in a trending pair you run more than twice the risk of being caught in adverse conditions.

you might be long high in the swing like @ 1.5800 in EUR/USD and the pair crashes through the swing low and bang you`re more than 1,000 pips in the red.
this wont happen very often.

on the other hand you might be short @ 1.5500 and the pair motors on through 1.6100 and beyond.
this might happen quite regularly.

by trading LONG only the adverse risks are limited to a change of the trend or a slight reversal.

so in theory the LONG only system will be a lot safer.

what do Fj and the backtesters think?
  • #451
  • Aug 17, 2008 6:17am Aug 17, 2008 6:17am
Fj,

are you going to trade the 10 pairs long and short each pair?

thanks
  • #452
  • Aug 17, 2008 6:09pm Aug 17, 2008 6:09pm
Quoting yonnie
Fj,

are you going to trade the 10 pairs long and short each pair?

thanks

Yes, I am. I am also going to start keeping track of the results with the trend and against the trend.
  • #453
  • Aug 18, 2008 6:47am Aug 18, 2008 6:47am
Quoting Flyjetz
It's not as big of a loss as it appears. 23.28 to 23.927 is a loss of 2.78%. If this same movement were to happen on a Long trade on the Euro from 1.50, you would close out at 1.4583. Or a loss of 417 pips. Pips (for comparison's sake) really are meaningless when you get into the Exotics. That is why I separate out the statistics by Majors, Exotics, and Metals.
Ahh - got it - see I said it was a great system, sorry for the small doubt, carry on
BTC Return This Year: na
  • #454
  • Aug 18, 2008 6:47pm Aug 18, 2008 6:47pm
(FJ, You wrote this in post #1 when introducing the original system).

I still don't understand why you'd want to abandon your fundamental friend.
Quoting Flyjetz
Yes, I am. I am also going to start keeping track of the results with the trend and against the trend.
Last week, mathematician also demonstrated the edge that going with the trend offers.

I hate to ask again, but what am I missing here? It just seems like low-probability overtrading to me. I am really just trying to understand this at this point...

Thanks and good luck!
6pack
  • #455
  • Aug 18, 2008 8:38pm Aug 18, 2008 8:38pm
Quoting 6pack
(FJ, You wrote this in post #1 when introducing the original system).

I still don't understand why you'd want to abandon your fundamental friend.

Last week, mathematician also demonstrated the edge that going with the trend offers.

I hate to ask again, but what am I missing here? It just seems like low-probability overtrading to me. I am really just trying to understand this at this point...

Thanks and good luck!
6pack
Yes I did say that. And I struggled with the decision to trade both directions. I have no doubt that trading with the trend will outperform trading against the trend. What I am interested in is whether or not trading against the trend outperforms letting your capital sit idle in your account while you are waiting for retracement. In other words, is it consistently profitable? I feel that it has been. But I want to verify that fact in my future trades. Besides, it's obvious from this thread's posts that there are already more than a dozen "offshoot" systems that trade this system with the basic fundamentals, but with a tweak or two here and there. Feel free to trade this system however it is most comfortable to you. If I can prove to myself that trading this system against the trend is not profitable, I will be the first to change it.
  • #456
  • Aug 19, 2008 11:36pm Aug 19, 2008 11:36pm
Quoting Flyjetz
Yes I did say that. And I struggled with the decision to trade both directions. I have no doubt that trading with the trend will outperform trading against the trend. What I am interested in is whether or not trading against the trend outperforms letting your capital sit idle in your account while you are waiting for retracement. In other words, is it consistently profitable? I feel that it has been. But I want to verify that fact in my future trades.

in my opinion you have been lucky that sofar this year it has been a ranging market mostly.
for me I would rather raise my bets on trend trading than trading against the trend.
a slow steady trend is alright, but watch out if you`re short and the pair shoots up and never comes back.
it will be the USD/AUD saga all over again.

Besides, it's obvious from this thread's posts that there are already more than a dozen "offshoot" systems that trade this system with the basic fundamentals, but with a tweak or two here and there.
wonder if people could post how they`re doing and what changes have been made so that we can all profit from it.
those systems can possibly be improved even more with all the bright brains over here.

this thread is very quiet, so wonder where those more than a dozen people are???
  • #457
  • Aug 20, 2008 12:51am Aug 20, 2008 12:51am
Quoting yonnie
wonder if people could post how they`re doing and what changes have been made so that we can all profit from it.
those systems can possibly be improved even more with all the bright brains over here.

this thread is very quiet, so wonder where those more than a dozen people are???
I am waiting for SJ for his method..
Currently I am thinking of using breakout and trending method to hedge
  • #458
  • Aug 20, 2008 3:17am Aug 20, 2008 3:17am
had a look at the uptrend of EUR/USD from Dec 2005 till now.

it had a high of 1.2320, went down to 1.1825 and shot up 1,153 pips to 1.2978
eventually it went down to 1.2458, before going up again.
suppose you were short between 1.2320 and 1.1825

it had a high of 1.3851, went down to 1.3359 and shot up 1,608 pips to 1.4967.
eventually it went down to 1.4364, before going up again.
suppose you were short between 1.3851 and 1.3359

it had a high of 1.4967, went down to 1.4438 and shot up 1,580 pips to 1.6018.
eventually it went down to 1.5283, before going up again.
suppose you were short between 1.4967 and 1.4438

even during the large run-ups you could have been caught short.

on the other hand there were never any problems with trend changes/large reversals so this pair was very reliable until recently.

so from December 2005 till now with this pair you could have had at least 4 potential disasters depending on having been long or short.

with going with the trend, you would have had 1 emergency only with the recent large reversal.
  • #459
  • Aug 20, 2008 11:29am Aug 20, 2008 11:29am
Quoting yonnie
wonder if people could post how they`re doing and what changes have been made so that we can all profit from it.
those systems can possibly be improved even more with all the bright brains over here.

this thread is very quiet, so wonder where those more than a dozen people are???
I am not trading this system, but I hang around because I like the discussions here. Some other threads here in FF attract... well, let's just say not the most intelligent posts, at times.
I'm trading extremely low-key now, almost not trading. You asked, Yonnie, so I'll talk about what I plan on doing. When I have some more time, I will build a discretionary system based on the following principles:

1. The trend is your friend. Use triple screen principles: go with long-term trend, but when going against some midterm trend, make sure a short term trend is with you!

2. Combine tested S/R + Fib levels (see No Brainer trades thread) to decide on entries and exits.

3. Let your profits run, but cut losses short. Don't let a position move against you. If the price for this is multiple re-entries - consider paying that price.

4. In Forex, not all currencies were created equal. Has anyone ever felt that US$ is king... [be careful of correlations, multiple "bets" on dollar, etc.]

5. Time is money, so don't "baby-sit" your positions. This implies trading on long timeframes.

Easier said than done...

FJ, I want to stress that this is only in response to Yonnie's question - I'm not trying to take anything or anyone away from your excellent system.

Cheers!
6pack
  • #460
  • Aug 21, 2008 12:31pm Aug 21, 2008 12:31pm
Quoting yonnie
had a look at the uptrend of EUR/USD from Dec 2005 till now.

it had a high of 1.2320, went down to 1.1825 and shot up 1,153 pips to 1.2978
eventually it went down to 1.2458, before going up again.
suppose you were short between 1.2320 and 1.1825

it had a high of 1.3851, went down to 1.3359 and shot up 1,608 pips to 1.4967.
eventually it went down to 1.4364, before going up again.
suppose you were short between 1.3851 and 1.3359

it had a high of 1.4967, went down to 1.4438 and shot up 1,580 pips to 1.6018.
eventually it went down to 1.5283, before going up again.
suppose you were short between 1.4967 and 1.4438

even during the large run-ups you could have been caught short.

on the other hand there were never any problems with trend changes/large reversals so this pair was very reliable until recently.

so from December 2005 till now with this pair you could have had at least 4 potential disasters depending on having been long or short.

with going with the trend, you would have had 1 emergency only with the recent large reversal.
What time frames do these movements occur over? Are these one-week moves? At the end of the week, this system adjusts its TP's, so there is a good chance, this system could have ridden through those extremes without sustaining the full drawdown.
0 traders viewing now