Is this a Successful System? | Page 2 | Forex Factory

Options

Go to
Permalinks
Print Thread

Similar Threads

Successful Trading & Determining the Probability of a Trading System 10 replies

What are the components of a successful trading system? 12 replies

successful system needs coding 0 replies

Successful Manual System - EA Needed Please 19 replies

Successful system needs MT4 programmer 6 replies

Attachments: Is this a Successful System?
Exit Attachments

Is this a Successful System?

  • #21
  • Oct 3, 2006 2:22pm Oct 3, 2006 2:22pm
It seems that this book answers all those questions that I have in mind. I'll order the book very soon.

Your first post could potentially save me thousands of dollars. Welcome to FF.

Thanks very much,


Quoting trader77
My First post.<?xml:namespace prefix = o ns = "urn:schemas-microsoft-com:office:office" /><o></o>

<o></o>

I personally never touch a system that did not have at least profit ratio of 2.1 because forward testing usually deteriorates the results some what. I read a really good book a few years ago which you might consider reading, Its "Trading Systems that Work by Thomas Stridman". <o></o>

<o></o>

Trader77<o></o>

<o></o>
  • #22
  • Oct 3, 2006 3:08pm Oct 3, 2006 3:08pm
Quoting aparsai
Could you explain what does "curve fitting" refer to? Is there any good resource that explain it?

Thanks,
"Curve fitting" as I call it is optimizing based on previous data that will in no way optimize the future. For example, say take one trade a day and you have 12 months of data on record. By a miracle of the gods, it has been a sheer coincidence that the 3rd wednesday of every month has been a winning trade. One can't "curve fit" their money managment to trade higher leverage on future 3rd wednesdays, just because it happened in the past.

There is a degree of optimization that can be done to a system, but anything over, well, is just overboard.
  • #23
  • Oct 4, 2006 10:01am Oct 4, 2006 10:01am
Quoting twinchell
"Curve fitting" as I call it is optimizing based on previous data that will in no way optimize the future. .
Thanks Twinchell,

That is a good summary of what I was trying to say.

Quote
Hi aparsai

The issue here is that you have "optimised" the parameters for a set of data over a (previous) period of time....

It does NOT follow that the parameters for the previous period will be the best parameters for the future period.....

It sounds logical in theory but it is a fallacy.
  • #24
  • Oct 4, 2006 10:50am Oct 4, 2006 10:50am
I wouldn't put any faith at all in the Meta Trader strategy tester, it is a joke in my opinion.

I have an EA that if I use the strategy tester it shows me doing nothing but losing money the past three months but the past three months of live trading have gained around 250 pips a month.
  • #25
  • Oct 4, 2006 12:31pm Oct 4, 2006 12:31pm
You might have a problem in setting up the tester. I have used my EA live and it acts exactly the same as what my tester says. The fact that you can or cannot rely on backtesting has nothing to do with the quality of MT4 tester.

I had the same idea as you do about MT4 strategy tester but then found out that I couldn't set it up properly.


Quoting hidethereal
I wouldn't put any faith at all in the Meta Trader strategy tester, it is a joke in my opinion.

I have an EA that if I use the strategy tester it shows me doing nothing but losing money the past three months but the past three months of live trading have gained around 250 pips a month.
  • #26
  • Oct 4, 2006 12:45pm Oct 4, 2006 12:45pm
Quoting aparsai
You might have a problem in setting up the tester. I have used my EA live and it acts exactly the same as what my tester says. The fact that you can or cannot rely on backtesting has nothing to do with the quality of MT4 tester.

I had the same idea as you do about MT4 strategy tester but then found out that I couldn't set it up properly.
I know how to use it correctly. It's just not accurate.
  • #27
  • Oct 4, 2006 1:07pm Oct 4, 2006 1:07pm
Quoting aparsai
My perception of those numbers is that they are VARs. Anyway I expanded the time from of the backtesting. It now tests the system from June 16, 04 to Oct 3, 06. I couldn't go further back as I have no access to 1M chart data. Anyway the following table shows the profit system makes for different VARs. I posted the detailed reports to my website www.fxnewstrader.com. You can click the links to view them all. BTW, it seems that maximum profit lies somewhere between 15.20% to 20% VAR as 20% VAR shows less profit than 15.20%

<table class="MsoTableGrid" style="border: medium none ; border-collapse: collapse;" border="1" cellpadding="0" cellspacing="0"><tbody><tr style=""><td style="border: 1pt solid windowtext; padding: 0in 5.4pt; background: rgb(230, 230, 230) none repeat scroll 0% 50%; -moz-background-clip: -moz-initial; -moz-background-origin: -moz-initial; -moz-background-inline-policy: -moz-initial; width: 49.05pt;" valign="top" width="65">

VAR<o></o>

</td><td style="border-style: solid solid solid none; border-color: windowtext windowtext windowtext rgb(212, 208, 200); border-width: 1pt 1pt 1pt medium; padding: 0in 5.4pt; background: rgb(230, 230, 230) none repeat scroll 0% 50%; -moz-background-clip: -moz-initial; -moz-background-origin: -moz-initial; -moz-background-inline-policy: -moz-initial; width: 91.35pt;" valign="top" width="122">

Net Profit<o></o>

</td><td style="border-style: solid solid solid none; border-color: windowtext windowtext windowtext rgb(212, 208, 200); border-width: 1pt 1pt 1pt medium; padding: 0in 5.4pt; background: rgb(230, 230, 230) none repeat scroll 0% 50%; -moz-background-clip: -moz-initial; -moz-background-origin: -moz-initial; -moz-background-inline-policy: -moz-initial; width: 1.5in;" valign="top" width="144">

Detailed Report<o></o>

</td></tr><tr style=""><td style="border-style: none solid solid; border-color: rgb(212, 208, 200) windowtext windowtext; border-width: medium 1pt 1pt; padding: 0in 5.4pt; width: 49.05pt; background-color: transparent;" valign="top" width="65">

0.50%<o></o>

</td><td style="border-style: none solid solid none; border-color: rgb(212, 208, 200) windowtext windowtext rgb(212, 208, 200); border-width: medium 1pt 1pt medium; padding: 0in 5.4pt; width: 91.35pt; background-color: transparent;" valign="top" width="122">

$3,147.79<o></o>

</td><td style="border-style: none solid solid none; border-color: rgb(212, 208, 200) windowtext windowtext rgb(212, 208, 200); border-width: medium 1pt 1pt medium; padding: 0in 5.4pt; width: 1.5in; background-color: transparent;" valign="top" width="144">

Click Here<o></o>

</td></tr><tr style=""><td style="border-style: none solid solid; border-color: rgb(212, 208, 200) windowtext windowtext; border-width: medium 1pt 1pt; padding: 0in 5.4pt; width: 49.05pt; background-color: transparent;" valign="top" width="65">

1.00%<o></o>

</td><td style="border-style: none solid solid none; border-color: rgb(212, 208, 200) windowtext windowtext rgb(212, 208, 200); border-width: medium 1pt 1pt medium; padding: 0in 5.4pt; width: 91.35pt; background-color: transparent;" valign="top" width="122">

$8,633.57<o></o>

</td><td style="border-style: none solid solid none; border-color: rgb(212, 208, 200) windowtext windowtext rgb(212, 208, 200); border-width: medium 1pt 1pt medium; padding: 0in 5.4pt; width: 1.5in; background-color: transparent;" valign="top" width="144">

Click Here<o></o>

</td></tr><tr style=""><td style="border-style: none solid solid; border-color: rgb(212, 208, 200) windowtext windowtext; border-width: medium 1pt 1pt; padding: 0in 5.4pt; background: rgb(204, 255, 204) none repeat scroll 0% 50%; -moz-background-clip: -moz-initial; -moz-background-origin: -moz-initial; -moz-background-inline-policy: -moz-initial; width: 49.05pt;" valign="top" width="65">

3.04%<o></o>

</td><td style="border-style: none solid solid none; border-color: rgb(212, 208, 200) windowtext windowtext rgb(212, 208, 200); border-width: medium 1pt 1pt medium; padding: 0in 5.4pt; background: rgb(204, 255, 204) none repeat scroll 0% 50%; -moz-background-clip: -moz-initial; -moz-background-origin: -moz-initial; -moz-background-inline-policy: -moz-initial; width: 91.35pt;" valign="top" width="122">

$45,792.26<o></o>

</td><td style="border-style: none solid solid none; border-color: rgb(212, 208, 200) windowtext windowtext rgb(212, 208, 200); border-width: medium 1pt 1pt medium; padding: 0in 5.4pt; background: rgb(204, 255, 204) none repeat scroll 0% 50%; -moz-background-clip: -moz-initial; -moz-background-origin: -moz-initial; -moz-background-inline-policy: -moz-initial; width: 1.5in;" valign="top" width="144">

Click Here<o></o>

</td></tr><tr style=""><td style="border-style: none solid solid; border-color: rgb(212, 208, 200) windowtext windowtext; border-width: medium 1pt 1pt; padding: 0in 5.4pt; background: rgb(204, 255, 204) none repeat scroll 0% 50%; -moz-background-clip: -moz-initial; -moz-background-origin: -moz-initial; -moz-background-inline-policy: -moz-initial; width: 49.05pt;" valign="top" width="65">

3.80%<o></o>

</td><td style="border-style: none solid solid none; border-color: rgb(212, 208, 200) windowtext windowtext rgb(212, 208, 200); border-width: medium 1pt 1pt medium; padding: 0in 5.4pt; background: rgb(204, 255, 204) none repeat scroll 0% 50%; -moz-background-clip: -moz-initial; -moz-background-origin: -moz-initial; -moz-background-inline-policy: -moz-initial; width: 91.35pt;" valign="top" width="122">

$70,147.98<o></o>

</td><td style="border-style: none solid solid none; border-color: rgb(212, 208, 200) windowtext windowtext rgb(212, 208, 200); border-width: medium 1pt 1pt medium; padding: 0in 5.4pt; background: rgb(204, 255, 204) none repeat scroll 0% 50%; -moz-background-clip: -moz-initial; -moz-background-origin: -moz-initial; -moz-background-inline-policy: -moz-initial; width: 1.5in;" valign="top" width="144">

Click Here<o></o>

</td></tr><tr style=""><td style="border-style: none solid solid; border-color: rgb(212, 208, 200) windowtext windowtext; border-width: medium 1pt 1pt; padding: 0in 5.4pt; background: rgb(204, 255, 204) none repeat scroll 0% 50%; -moz-background-clip: -moz-initial; -moz-background-origin: -moz-initial; -moz-background-inline-policy: -moz-initial; width: 49.05pt;" valign="top" width="65">

5.07%<o></o>

</td><td style="border-style: none solid solid none; border-color: rgb(212, 208, 200) windowtext windowtext rgb(212, 208, 200); border-width: medium 1pt 1pt medium; padding: 0in 5.4pt; background: rgb(204, 255, 204) none repeat scroll 0% 50%; -moz-background-clip: -moz-initial; -moz-background-origin: -moz-initial; -moz-background-inline-policy: -moz-initial; width: 91.35pt;" valign="top" width="122">

$129,119.12<o></o>

</td><td style="border-style: none solid solid none; border-color: rgb(212, 208, 200) windowtext windowtext rgb(212, 208, 200); border-width: medium 1pt 1pt medium; padding: 0in 5.4pt; background: rgb(204, 255, 204) none repeat scroll 0% 50%; -moz-background-clip: -moz-initial; -moz-background-origin: -moz-initial; -moz-background-inline-policy: -moz-initial; width: 1.5in;" valign="top" width="144">

Click Here<o></o>

</td></tr><tr style=""><td style="border-style: none solid solid; border-color: rgb(212, 208, 200) windowtext windowtext; border-width: medium 1pt 1pt; padding: 0in 5.4pt; background: rgb(204, 255, 204) none repeat scroll 0% 50%; -moz-background-clip: -moz-initial; -moz-background-origin: -moz-initial; -moz-background-inline-policy: -moz-initial; width: 49.05pt;" valign="top" width="65">

7.60%<o></o>

</td><td style="border-style: none solid solid none; border-color: rgb(212, 208, 200) windowtext windowtext rgb(212, 208, 200); border-width: medium 1pt 1pt medium; padding: 0in 5.4pt; background: rgb(204, 255, 204) none repeat scroll 0% 50%; -moz-background-clip: -moz-initial; -moz-background-origin: -moz-initial; -moz-background-inline-policy: -moz-initial; width: 91.35pt;" valign="top" width="122">

$337,342.23<o></o>

</td><td style="border-style: none solid solid none; border-color: rgb(212, 208, 200) windowtext windowtext rgb(212, 208, 200); border-width: medium 1pt 1pt medium; padding: 0in 5.4pt; background: rgb(204, 255, 204) none repeat scroll 0% 50%; -moz-background-clip: -moz-initial; -moz-background-origin: -moz-initial; -moz-background-inline-policy: -moz-initial; width: 1.5in;" valign="top" width="144">

Click Here<o></o>

</td></tr><tr style=""><td style="border-style: none solid solid; border-color: rgb(212, 208, 200) windowtext windowtext; border-width: medium 1pt 1pt; padding: 0in 5.4pt; width: 49.05pt; background-color: transparent;" valign="top" width="65">

10.00%<o></o>

</td><td style="border-style: none solid solid none; border-color: rgb(212, 208, 200) windowtext windowtext rgb(212, 208, 200); border-width: medium 1pt 1pt medium; padding: 0in 5.4pt; width: 91.35pt; background-color: transparent;" valign="top" width="122">

$628,131.20<o></o>

</td><td style="border-style: none solid solid none; border-color: rgb(212, 208, 200) windowtext windowtext rgb(212, 208, 200); border-width: medium 1pt 1pt medium; padding: 0in 5.4pt; width: 1.5in; background-color: transparent;" valign="top" width="144">

Click Here<o></o>

</td></tr><tr style=""><td style="border-style: none solid solid; border-color: rgb(212, 208, 200) windowtext windowtext; border-width: medium 1pt 1pt; padding: 0in 5.4pt; background: rgb(204, 255, 204) none repeat scroll 0% 50%; -moz-background-clip: -moz-initial; -moz-background-origin: -moz-initial; -moz-background-inline-policy: -moz-initial; width: 49.05pt;" valign="top" width="65">

15.20%<o></o>

</td><td style="border-style: none solid solid none; border-color: rgb(212, 208, 200) windowtext windowtext rgb(212, 208, 200); border-width: medium 1pt 1pt medium; padding: 0in 5.4pt; background: rgb(204, 255, 204) none repeat scroll 0% 50%; -moz-background-clip: -moz-initial; -moz-background-origin: -moz-initial; -moz-background-inline-policy: -moz-initial; width: 91.35pt;" valign="top" width="122">

$3,846,616.36<o></o>

</td><td style="border-style: none solid solid none; border-color: rgb(212, 208, 200) windowtext windowtext rgb(212, 208, 200); border-width: medium 1pt 1pt medium; padding: 0in 5.4pt; background: rgb(204, 255, 204) none repeat scroll 0% 50%; -moz-background-clip: -moz-initial; -moz-background-origin: -moz-initial; -moz-background-inline-policy: -moz-initial; width: 1.5in;" valign="top" width="144">

Click Here<o></o>

</td></tr><tr style=""><td style="border-style: none solid solid; border-color: rgb(212, 208, 200) windowtext windowtext; border-width: medium 1pt 1pt; padding: 0in 5.4pt; width: 49.05pt; background-color: transparent;" valign="top" width="65">

20.00%<o></o>

</td><td style="border-style: none solid solid none; border-color: rgb(212, 208, 200) windowtext windowtext rgb(212, 208, 200); border-width: medium 1pt 1pt medium; padding: 0in 5.4pt; width: 91.35pt; background-color: transparent;" valign="top" width="122">

$2,939,788.57<o></o>

</td><td style="border-style: none solid solid none; border-color: rgb(212, 208, 200) windowtext windowtext rgb(212, 208, 200); border-width: medium 1pt 1pt medium; padding: 0in 5.4pt; width: 1.5in; background-color: transparent;" valign="top" width="144">

Click Here<o></o>

</td></tr></tbody></table>

The shaded rows are those numbers you suggested.

Thanks,
Al
Hi I have two questions to these reports and graphs

1. for different VAR there is nonequal amount of market entries - this may mislead.

2. look at the peak at the right side of the graphs, this sugegsts that this system works good in certain conditions on the market (whatever this means) and in other - looses money - see the deppression at trades 440-540 and sharp decline of equity for trades after 570. imho far too long, far too sharp. and the question is - do you know the nature of "certain conditions"?

answer to this question can point out when to trade in order to increase equity and when not to.

This is a first impression not mature opinion - this system is not successful - too risky (at least 4 me). I'll explain this later.

Hope this helps.

D.
Quidquid latine dictum sit altum viditur
  • #28
  • Oct 4, 2006 1:34pm Oct 4, 2006 1:34pm
Quoting dottore
Hi I have two questions to these reports and graphs

1. for different VAR there is nonequal amount of market entries - this may mislead.

2. look at the peak at the right side of the graphs, this sugegsts that this system works good in certain conditions on the market (whatever this means) and in other - looses money - see the deppression at trades 440-540 and sharp decline of equity for trades after 570. imho far too long, far too sharp. and the question is - do you know the nature of "certain conditions"?
....
D.
Thanks very much for your feedback.

1. I didn't understand this question very well. You may need to rephrase it. However, based on my understanding I need to explain that the lot size is calculated based on three parameters, "the current account free margin", "the VAR", and "the stop loss". The S/L is a variable value in the system and it's based on the several parameters that's why for the same free margin and VAR you may see different lot sizes to be traded.

2. Apparently this system keeps loosing for a while and then rises for a while. I'm not quite sure what's the rational behind it. It's a box trader and for me it's really hard to say why it acts like that. I'll publish the code very soon so you all have the chance to look into it. Somebody may come up with an idea to enhance it.

I'm not quite sure if this system is successful or not that's why I raised the question in first place.

Thanks again for your feedback.
  • #29
  • Oct 4, 2006 2:12pm Oct 4, 2006 2:12pm
Quoting trader77
My First post.<?xml:namespace prefix = o ns = "urn:schemas-microsoft-com:office:office" /><o></o>

<o></o>

I personally never touch a system that did not have at least profit ratio of 2.1 because forward testing usually deteriorates the results some what. I read a really good book a few years ago which you might consider reading, Its "Trading Systems that Work by Thomas Stridman". <o></o>

<o></o>

Trader77<o></o>

<o></o>
I would be more concerned about touching a system with a profit factor of over 2 than one under 2. A high profit factor is one of the first warnings of an overfit system.

I have Stridman's book in front of me. Can you tell me where he says it's necesary to have a profit factor of 2? Here's what he said on page 8: "Many system vendors and trading experts believe that you should not trade a system with a hypothetically back-tested profit factor of below three, because they know from experience that the profit factors for all their systems will decrease considerably when the system is traded live on unseen data. Probably the only reason for this recommendation is that they do not understand how to build a robust trading strategy in the first place."

By the way, welcome to FF!
  • #30
  • Oct 4, 2006 2:15pm Oct 4, 2006 2:15pm
Quoting aparsai
I develped an EA for MetaTrader 4 and tested it on USDCHF. The result for a 2 year period from Sep 20,04 to Sep 20, 06 with 1% VAR (Value At Risk) is as follows.
aparsai, what did you use for commissions and slippage in your testing?
  • #31
  • Oct 4, 2006 2:32pm Oct 4, 2006 2:32pm
Quoting jacko
Most people, when they backtest a system, unconsciously and subconsciously tend to inadvertently "optimise" the system to find the "best fit".

(However there are some real turkeys here too... LOL).


Jacko
Jacko,

What's the alternative? Deliberately design a system that doesn't work or one that performs poorly?

The selection of a currency pair, the time frame, the indicators, the time of day traded, the position sizing model......are all forms of optimization.

Signed,
One of the turkeys
  • #32
  • Oct 4, 2006 3:31pm Oct 4, 2006 3:31pm
Quoting aparsai
Thanks very much for your feedback.

1. I didn't understand this question very well. You may need to rephrase it. However, based on my understanding I need to explain that the lot size is calculated based on three parameters, "the current account free margin", "the VAR", and "the stop loss". The S/L is a variable value in the system and it's based on the several parameters that's why for the same free margin and VAR you may see different lot sizes to be traded.

2. Apparently this system keeps loosing for a while and then rises for a while. I'm not quite sure what's the rational behind it. It's a box trader and for me it's really hard to say why it acts like that. I'll publish the code very soon so you all have the chance to look into it. Somebody may come up with an idea to enhance it.

I'm not quite sure if this system is successful or not that's why I raised the question in first place.

Thanks again for your feedback.
Sorry for my English.

1. number of trades - on different graphs there is different number of trades, is that OK?

2. On the attached graph I see five different periods.

1 - all seems ok, stable growth of equity, at 364 - drawdown but short and I believe acceptable.

2 - 40 trades - first sharp decline - I believe 60 % of capital.

3. - that I believe is acceptable - small but steady growth.

4 - we are in heaven...

5. - looong decline - 60 loosing trades.

This system likes long loosing series - probably it is caused by "certain conditions of the market."

The overall outcome looks good, BUT - what if we start with 100 k and "certain conditions" will be like in periods 2 or 5? 60% down? What if next "loosing period" will last 80 trades?

Thats why i suggest to study differences between 1,2,4 and 5. maybe there are strong trends present? For example up, down, up down.. This will explain where this system performs good and where it fails.

And if we add certain filter (stop entering the martket if .....) - this will improve the outcome.

Hope this helps...
Attached File(s)
File Type: zip five periods.zip   44 KB | 271 downloads
Quidquid latine dictum sit altum viditur
  • #33
  • Oct 4, 2006 3:42pm Oct 4, 2006 3:42pm
Quoting dottore
Sorry for my English.

1. number of trades - on different graphs there is different number of trades, is that OK?

2. On the attached graph I see five different periods.

1 - all seems ok, stable growth of equity, at 364 - drawdown but short and I believe acceptable.

2 - 40 trades - first sharp decline - I believe 60 % of capital.

3. - that I believe is acceptable - small but steady growth.

4 - we are in heaven...

5. - looong decline - 60 loosing trades.

This system likes long loosing series - probably it is caused by "certain conditions of the market."

The overall outcome looks good, BUT - what if we start with 100 k and "certain conditions" will be like in periods 2 or 5? 60% down? What if next "loosing period" will last 80 trades?

Thats why i suggest to study differences between 1,2,4 and 5. maybe there are strong trends present? For example up, down, up down.. This will explain where this system performs good and where it fails.

And if we add certain filter (stop entering the martket if .....) - this will improve the outcome.

Hope this helps...
I looked at USD/CHF trying to spot the differences between periods. It seems that system performs "well" in bullish trends, wery well in bearish and fails if the market is undecided - last 3,5 - 4 months (60-80 trading days).

try this system on shorter periods with clear characteristics 1. only bullish (half april - end of july 2005), 2 only bearish (half april - half may this year), 3. sideways trend (half may - today). This testing maybe explain where and why it fails.

Quidquid latine dictum sit altum viditur
  • #34
  • Oct 4, 2006 5:58pm Oct 4, 2006 5:58pm
Quoting hidethereal
I know how to use it correctly. It's just not accurate.
I apologize if I made you upset. I didn't mean that you don't know how to use it. I meant you probably didn't set it up properly. For example my system is based on M15 charts but you need to have access to M1 charts in order to simulate the tick price action. Withou use of an M1 chart it results in a much more optimistic outcome which is not true.
  • #35
  • Oct 4, 2006 6:05pm Oct 4, 2006 6:05pm
Quoting philmcgrew
aparsai, what did you use for commissions and slippage in your testing?
Spread=3
Swap type=in points
Swap long=1.2
Swap short=-1.5
Slippage=0
Commission=0
Contract size=100000

These are pretty much the same as what my broker offers.
  • #36
  • Oct 4, 2006 6:11pm Oct 4, 2006 6:11pm
Quoting dottore
Sorry for my English.

1. number of trades - on different graphs there is different number of trades, is that OK?

2. On the attached graph I see five different periods.

1 - all seems ok, stable growth of equity, at 364 - drawdown but short and I believe acceptable.

2 - 40 trades - first sharp decline - I believe 60 % of capital.

3. - that I believe is acceptable - small but steady growth.

4 - we are in heaven...

5. - looong decline - 60 loosing trades.

This system likes long loosing series - probably it is caused by "certain conditions of the market."

The overall outcome looks good, BUT - what if we start with 100 k and "certain conditions" will be like in periods 2 or 5? 60% down? What if next "loosing period" will last 80 trades?

Thats why i suggest to study differences between 1,2,4 and 5. maybe there are strong trends present? For example up, down, up down.. This will explain where this system performs good and where it fails.

And if we add certain filter (stop entering the martket if .....) - this will improve the outcome.

Hope this helps...
The tester shows 621 trades for all VARs but 15.20% and 20%. I'll look into this to see what went wrong.

Thanks for analyzing the chart. It makes it more clear how this system works. I will publish the system tonight and then you can test it yourself. It did quite well on Demo account within the last few days and all trades ended up in profit. However, as you see on images there are periods that it keeps losing.
  • #37
  • Oct 4, 2006 6:12pm Oct 4, 2006 6:12pm
Quoting aparsai
Spread=3
Swap type=in points
Swap long=1.2
Swap short=-1.5
Slippage=0
Commission=0
Contract size=100000

These are pretty much the same as what my broker offers.
Good for you! I dont know many brokers who offer 3 pips on the CHF without commissions but I believe you. I have no interest in your system but when I look at your results I would say it has potential and is one of the most realistic systems I've seen on FF. It's refreshing to see that you don't claim to make a gazillion pips in 15 days.

One word of advice that I would offer is that when doing your initial evaluation you do not take advantage of position sizing. It is easier to judge a system on a fixed contract basis. Once you get the system where you want it then add in the position sizing code. Position sizing can cover up some bad spots, which is good, but misleading when doing your initial analysis.

Don't forget, "optimization" is not a dirty word.
  • #38
  • Oct 4, 2006 6:13pm Oct 4, 2006 6:13pm
Quoting dottore
I looked at USD/CHF trying to spot the differences between periods. It seems that system performs "well" in bullish trends, wery well in bearish and fails if the market is undecided - last 3,5 - 4 months (60-80 trading days).

try this system on shorter periods with clear characteristics 1. only bullish (half april - end of july 2005), 2 only bearish (half april - half may this year), 3. sideways trend (half may - today). This testing maybe explain where and why it fails.

I'll pubish the system in a couple of hours and then you and others will hopefully help me to optimize it. I'm very optimistic about this system but it needs some modifications to become a reliable one.
  • #39
  • Oct 4, 2006 7:12pm Oct 4, 2006 7:12pm
I posted the EA in a separate thread. Please continue the discussion over there:

http://www.forexfactory.com/forexfor...ead.php?t=8692
  • #40
  • Oct 5, 2006 8:26am Oct 5, 2006 8:26am
Quoting aparsai
I apologize if I made you upset. I didn't mean that you don't know how to use it. I meant you probably didn't set it up properly. For example my system is based on M15 charts but you need to have access to M1 charts in order to simulate the tick price action. Withou use of an M1 chart it results in a much more optimistic outcome which is not true.
I'm not upset . The way my post read sounded rude but wasn't meant to be.
0 traders viewing now