Forex Factory (https://www.forexfactory.com/forum.php)
-   Trading Systems (https://www.forexfactory.com/forumdisplay.php?f=71)
-   -   2 Trades per Week (Weekly scalping&First Strike modification) (https://www.forexfactory.com/showthread.php?t=144542)

mades Jan 12, 2009 4:44am | Post# 1

2 Trades per Week (Weekly scalping&First Strike modification)
 
Hi everyone,

Probably all of you know the simple strategy which has been here presented in Weekly Scalping thread for over 2 years now. The basis is simple, you open one of 2 positions set at certain time, and which one gets hit, will be traded. Joel Rensink does have an advanced strategy called First Strike plus, where he sets orders according to last week's volatility. Because year 2008 has been very kind but also very harsh to this strategy and weekly scalping thread went to silence, i took a look and did some strategy modification + paper trading for year 2008, so i'd like to present you my results. Please note, i am just human and i might have made mistakes. I used IBFX chart, because i have good experience with accuracy of their charts.

The basic numbers (30 pip offset, 45SL, 135TP) were chosen after examination and statistics i did with hand (i also tried 50/100/300, 50/75/225, 40/80/240, 40/60/180 and 30/60/180).

Allright, here are the rules:
1) I did only EUR/USD
2) starting line is every Monday 06.00GMT open
3) both buy and sell orders are: 06.00GMT open +- 30 PIPS
4) If one order gets hit, the SL will be 45PIPS.
5) SL for the first order is also reverse order = if SL gets hit, we reverse position. The new SL will be again 45pips (= original first order). We do this only ONCE (so maximum 2 trades per week).
6) TP is ALWAYS 3:1 = 135 pips (3*45).
7) Money management. For the first order we use 5% of equity. For the second order we use 7% of equity. So all in all, we risk maximum per week 12% of equity. If we reach TP with first order, we have +15%. If we reach TP with second order we have +16% (7*3-5 = 16%). This is rather aggresive MM, some might consider.

  1. I didn't take into consideration, that if target wasn't reached, position could still be alive and in positive (or less negative). Any not-reached target was considered as full loss.
  2. First week starts at 7th of january 2008.
  3. I did round the equity results.
  4. The equity results are approximative, there are no commissions for trades included
  5. I also used exact numbers considering zero spread (which every good ECN should get you during busy hours anyway).
  6. They were alltogether 51 trading weeks (i didnt include the last one startin 22th of dec. 2008)
  7. I didn't check more back into 2007. The reason is simple, i believe that market conditions are changing rapidly and the old data don't have that much relevance anymore. 2008 was perfect, lot of ranging weeks, lot of strong trending weeks.

Results (30/45/135):

Out of 50 Weeks: 30 positive, 20 negative (60%). Out of 30 positive: 14 first order hit (47%), 16 second order hit (53%). Total number of trades: 78.
(p.s. i did some mistakes before, these are the fixed results, sorry for that)

When using DST (using 5.00GMT instead of 6.00GMT during DST):

Out of 50 Weeks: 32 positive, 18 negative (64%)



This was all done by hand and my goal here is just to show you, that basic idea of doing one or 2 trades per week starting on monday can be very profitable, if the parameters are set right. And they need to be adjusted every year. It would be interesting to code an algorithm, which would find the most profitable combination offset, SL, TP and you could also modify the open price (don't forget, i used 06.00GMT monday open). Also checking other pairs would be very intriguing and i'll try to do it in the next days. For now, i hope i gave you something to think about.

Well, life is never ideal , is it? After 2nd manual check, i did find quite a lot errors i did with 30/45/135 and the results dropped significantly. I will do some more examination also for higher offset (which basically made most positive-to-negative turns). Please excuse me, i am just human and pretty tired human for now


et_phonehome_2 Jan 12, 2009 4:54am | Post# 2

All I need is an EA to automatically place the trades. I had also been trading it manually with only the GBPUSD, its very impressive except I do not wait until Friday to close the positions as outlined in the document.

mades Jan 12, 2009 4:56am | Post# 3

All I need is an EA to automatically place the trades. I had also been trading it manually with only the GBPUSD, its very impressive except I do not wait until Friday to close the positions as outlined in the document.
Me neither. In fact, ALL positive trades were closed much sooner due to fixed TP, some of them even on Monday. Waiting until Friday can earn you huge amount of pips but the % of positive trades falls rapidly. And as we all know, money is not being made by number of pips but number of pips X position size.

hollandpippin Jan 12, 2009 10:31am | Post# 4

would like to see a chart with your explanasion on it just to be sure what u mean

but 2000% is more then great!!!!

mades Jan 12, 2009 12:20pm | Post# 5

would like to see a chart with your explanasion on it just to be sure what u mean

but 2000% is more then great!!!!
What part of my explanation don't you understand?

split_unit Jan 12, 2009 12:40pm | Post# 6

Thank you for sharing this system with us,I am seriously looking forward to back testing your system as i believe that the opening of the week is the most crucial time of the week, but i never found a system that supported this or took advantage of this, besides Joel's system who i intend to learn his system.

I am a lurker in FF and have found very few systems that didnt rely on indicators and i believe yours is one of them as yours rely on price action, your generousity is beyond words, usually when someone has something like this they keep it to themselfs, but you have shared it with the rest of the world i truely applaud you.

Theres a point that i didnt understand and it would be great if you had the time to clearify it if possible:
5) SL for the first order is also reverse order = if SL gets hit, we reverse position. The new SL will be again 45pips (= original first order). We do this only ONCE (so maximum 2 trades per week).

thanks

PipAdder Jan 12, 2009 12:50pm | Post# 7

Theres a point that i didnt understand and it would be great if you had the time to clearify it if possible:
5) SL for the first order is also reverse order = if SL gets hit, we reverse position. The new SL will be again 45pips (= original first order). We do this only ONCE (so maximum 2 trades per week).

thanks
If I may chime in...

I believe he means that once the first order of the straddle is triggered (let's say it is a BUY STOP order), instead of cancelling the second order (a SELL STOP), he modifies it to match the price of the stop loss of the first order. That way, if after triggering the first order price goes against you and hits the SL it will automatically trigger the SELL order. Do that only once (no more triggering of additional orders if the second is also stooped out).

This bit is an addition to the basic First Strike method that I have seen at least once, proposed by Tkimble (and probably others before him). It is somewhere in a thread in this forum.

split_unit Jan 12, 2009 12:55pm | Post# 8

If I may chime in...

I believe he means that once the first order of the straddle is triggered (let's say it is a BUY STOP order), instead of cancelling the second order (a SELL STOP), he modifies it to match the price of the stop loss of the first order. That way, if after triggering the first order price goes against you and hits the SL it will automatically trigger the SELL order. Do that only once (no more triggering of additional orders if the second is also stooped out).

This bit is an addition to the basic First Strike method that I have...
Thank you Pipadder for your quick explanation, i understand now.

mades Jan 12, 2009 1:10pm | Post# 9

If I may chime in...

I believe he means that once the first order of the straddle is triggered (let's say it is a BUY STOP order), instead of cancelling the second order (a SELL STOP), he modifies it to match the price of the stop loss of the first order. That way, if after triggering the first order price goes against you and hits the SL it will automatically trigger the SELL order. Do that only once (no more triggering of additional orders if the second is also stooped out).

This bit is an addition to the basic First Strike method that I have...
Yes, exactly. I don't think i invented anything new, i just made some effort to find the working combination offset/SL/TP. The reverse order is very essential in my modification because 57% of successful trades were reversal orders. Also, if BOTH the orders get stopped, it usually indicates a very ranging day and range Monday indicates that the whole week will be ranging. It is not 100% valid, but with high probability. Look at today's EUR/USD chart, the 2nd order got stopped around 12GMT and it kept ranging since then. I believe we will see more ranging days and maybe one-two days trending(and contra trending). This is very typical after strong trending week. The advantage of this strategy with my parameters is that it wins during trendy weeks and has some chance to win during ranging weeks due to low TP (pipwise).

I'll do some more examination on how the price behaves on next days of the week after both orders were stopped.

EAprogrammer Jan 12, 2009 1:45pm | Post# 10

Worth to take a look. Thanks.

karun Jan 12, 2009 2:39pm | Post# 11


both buy and sell orders are: 06.00GMT open +- 30 PIPS
Could you explain this with an example? how do you decide on the offset whether its +30 or -30? Or is it Buy price = open + 30 ; Sell price = open - 30

Thanks
Karun

FrankTheTank Jan 12, 2009 2:58pm | Post# 12

Wow, thatnks for backtesting by hand. I know what a pain that is so its well appreciated.

I think I have an EA that can backtest this for you. I am sure there are a bunch of free "breakout" type EAs out there that we can find as well.

I will post results if I can find it.

FrankTheTank Jan 12, 2009 3:05pm | Post# 13

Further to my last post, the problem with these breakout systems and backtesting is that you can pretty much optomize it to be profitable for any backtesting time period (in your example, the year of 2008).

A few years back I had a two year period I was looking at and figured out (like you did) that if you have a stop loss of X and profit target of Y and took a trade at time Z then you could make $$.

The problem is that this data was all optomized based on the past. When you try to trade into the future, it often does not work out to be profitable.

Ronald Raygun Jan 12, 2009 3:29pm | Post# 14

Frank,

What you say is true. But remember that all mechanical trading is based off of the theory that the market repeats itself.

With that in mind, you just try to find a robust or self-adapting method for breakout-trading.

Iced Earth Jan 12, 2009 3:33pm | Post# 15

What part of my explanation don't you understand?
what I understand from your system is that you set up 2 enter points for the begining of next week, am I right? If it is so, 1- how did you decide what first trade should be sell or buy??
And 2- when you set up your pending orders?, because at the end of the trading weeks pending orders does'nt work ?

Thanks Beforehand

mades Jan 12, 2009 3:51pm | Post# 16

Further to my last post, the problem with these breakout systems and backtesting is that you can pretty much optomize it to be profitable for any backtesting time period (in your example, the year of 2008).

A few years back I had a two year period I was looking at and figured out (like you did) that if you have a stop loss of X and profit target of Y and took a trade at time Z then you could make $$.

The problem is that this data was all optimized based on the past....
Hi,

yes you are right. But past is the only thing we can use to backtest. I used year 2008 because 2009 will be closest to 2008 (as 08 was to 07 and so on) regarding price behavior (because markets evolve). I also explained that 2008 was very good because it had both types of weeks: ranging and trending and it was like 50:50. Please let me explain the time and parameters:

The 06.00GMT time Monday was chosen because of Joel Rensink's First Strike system (which he claims professional traders have been using at least 20 years). So the time i did for setting the offsets definitely has an edge of being proven. I wasn't looking for any other time and this one has a lot of common sense inside (big traders placing their orders at the beginning of the week).

The 30/45/135 were chosen based on my observation of price. I tried different (higher) offset, SL and TP and i realized that once price started to trend it didn't matter if i had 30 or 50 pips offset. But with 30 and 45SL, i can have smaller TP which basically means higher chance to reach. Once price started to range, it again didn't really matter if the SL was 45 or 100. Of course this is far from being fully optimized for 2008, but its easier to count when i backtest manually
In fact, we could call it "optimized" for EUR/USD based on average weekly movements. The goal obviously is to reach higher than 50% of successful TP.
I think, that you have to re-calculate the values every year based on the last year. So for 2009 maybe 30/45/135 will not work that well, but if it can keep above 50% it will be still positive (2008 was over 68%). Do we think the market changes that much within one year?

And please note, i would not call this a typical breakout system, because then you could tell that about any system using buy limit and sell stop orders. I wait for price to reach a certain level , which is measured based on open price on certain time and i believe (based on what Joel Rensink said) that most professional traders use one or another trading system which we could clasify as a breakout (DIBS for example).

mades Jan 12, 2009 3:54pm | Post# 17

what I understand from your system is that you set up 2 enter points for the begining of next week, am I right? If it is so, 1- how did you decide what first trade should be sell or buy??
And 2- when you set up your pending orders?, because at the end of the trading weeks pending orders does'nt work ?

Thanks Beforehand
1) It will depend on which order gets hit as first one. I don't decide, market does (either it goes up = buy order , or down = sell order).
2) 06.00GMT on Monday. I don't wait till the end of the week, i wait till it reaches my Take Profit target (135pips). If it won't reach the target and won't be stopped out, i will close it at the end of the week.

mades Jan 12, 2009 3:56pm | Post# 18

Could you explain this with an example? how do you decide on the offset whether its +30 or -30? Or is it Buy price = open + 30 ; Sell price = open - 30

Thanks
Karun
yes. buy = buy limit, sell = sell stop.

Iced Earth Jan 13, 2009 12:05am | Post# 19

1) It will depend on which order gets hit as first one. I don't decide, market does (either it goes up = buy order , or down = sell order).
2) 06.00GMT on Monday. I don't wait till the end of the week, i wait till it reaches my Take Profit target (135pips). If it won't reach the target and won't be stopped out, i will close it at the end of the week.
Thanks a lot for your explanation

about 2) : I want to know when you set up your trade, the previous friday or at the begining of the week ( monday)? because if you set it up on monday you would not be able to take the opening gap opportunities

WedWed Jan 13, 2009 1:51am | Post# 20

Thanks a lot for your explanation

about 2) : I want to know when you set up your trade, the previous friday or at the begining of the week ( monday)? because if you set it up on monday you would not be able to take the opening gap opportunities

Hi my friend Iced Earth , regarding to this quastion
2- when you set up your pending orders?

He answer it

2) 06.00GMT on Monday.


Thanx for you mades for sharing your system.


© Forex Factory