PipMaker v1 - Price action based EA - page 33

 

I hope any of the people developing this EA could check something. I think I have discovered a major bug in the MT backtester.

When placing a market order with 'Ordersend', it will sometimes open 2 orders instead of 1, passing the most recently opened order ticket to the calling subroutine. The 'extra' order is present in the 'Results' menu but is not displayed on the chart so you end up with a pile of 'orphan' orders that weren't supposed to be placed.

I only discovered this as I was storing the opened orders in an array so they could be more easily analyzed. I also noticed that 'clumps' of losing orders were not being closed as they should have been.

Hopefully this is confined to the backtester but it does render its results pretty meaningless if true (which I'm almost certain is...)

 

You don't...there are not any guarantees in forex Trading...

swisscheese:
and how do i know befor going live?
 

Many of the problems discussed here can be avoided if you work from longer time frames.

You can also try an IBFX account.

I would encourage demo testing at least on the 1H and with the LotIncrement set to zero. Also try a much smaller profit setting.

I think this may be a very robust EA without the added risk of a negative progression.

Dare to experiment with this stuff. Run a number of platforms with different settings and start them all with new accounts at the same time.

Maybe this has been mentioned, but some additional code to only trade within a specific spread width could be added as well.

 

I don't quite understand most of the posts on this ea. The ea works great. If most of the people here would just set up a few platforms and start forward testing they would see. Also, I suck at programming but I can read code a little. All the functions are right there in the code. Another thing, many people are complaining about backtesting????? My backtest data sucks big time and has gaps in it yet the ea still manages to come out positive. For the people with questions about bactesting, have you analyzed your data to check for accuracy?? You may be surprised. Also, for the people shooting for 100% returns every two weeks, hell yeah you're gonna have some drawdown....risk=reward. I'm out.

Nick,

ps Awesome ea

 

Here are the 19 pair that IBFX offers and their normal spreads:

EUR/USD (2)

USD/JPY (3)

EUR/GBP (4)

USD/CHF (4)

NZD/USD (5)

USD/CAD (5)

AUD/USD (5)

GBP/USD (6)

EUR/JPY (6)

EUR/CHF (6)

CHF/JPY (7)

AUD/JPY (8)

EUR/AUD (8)

AUD/CAD (8)

EUR/CAD (9)

GBP/JPY (9)

NZD/JPY (10)

GBP/CHF (10)

AUD/NZD (11)

I am trading 6 pairs with demo-dollars of 2K with 0.01 microlot inputs in Pipmaker at IBFX with v9.1 on a 15 minute setting: GBP/USD, USD/CHF, USD/JPY, AUD/USD, NZD/USD & USD/CAD. I am again faced with this incredible drawdown that this system seems to live with. To some of you 40% DD is ok, if you are getting the yield...but I am a 15% Intraday DD guy...I am already at 10% DD in under a day.

These are my four input changes from the defaults:

Lots: 0.01

SmallestLotSize: 2

TrendPeriod: 80

TrendtimeFrame: 15

DATE STARTED 08/22/07, BANKED DEMO PROFIT $204.16, DD:-$193.94

ES

 

Your points about the validity of backtesting and the quality control are very good and I agree.

But...

I would rather make half the yield and hold under 15% DD (there is a ratio of yield to DD that I benchmark with). I am not being negative...just searching for the correct number of pairs ...heck!... I am using the minumum Lots that I can to rank and benchmark this system with. Else I must change the pairs...increase the capital...or live with the DD...This is a personal choice that this system constantly faces me with.

When trying new systems...Pro-Traders look at the ratio of yield to drawdown FIRST! to see how robust the system is. I have yet to find 3 to 1...but still looking

ES

faifarni:
I don't quite understand most of the posts on this ea. The ea works great. If most of the people here would just set up a few platforms and start forward testing they would see. Also, I suck at programming but I can read code a little. All the functions are right there in the code. Another thing, many people are complaining about backtesting????? My backtest data sucks big time and has gaps in it yet the ea still manages to come out positive. For the people with questions about bactesting, have you analyzed your data to check for accuracy?? You may be surprised. Also, for the people shooting for 100% returns every two weeks, hell yeah you're gonna have some drawdown....risk=reward. I'm out.

Nick,

ps Awesome ea
 
ElectricSavant:
Here are the 19 pair that IBFX offers and their normal spreads:

EUR/USD (2)

USD/JPY (3)

EUR/GBP (4)

USD/CHF (4)

NZD/USD (5)

USD/CAD (5)

AUD/USD (5)

GBP/USD (6)

EUR/JPY (6)

EUR/CHF (6)

CHF/JPY (7)

AUD/JPY (8)

EUR/AUD (8)

AUD/CAD (8)

EUR/CAD (9)

GBP/JPY (9)

NZD/JPY (10)

GBP/CHF (10)

AUD/NZD (11)

I am trading 6 pairs with demo-dollars of 2K with 0.01 microlot inputs in Pipmaker at IBFX with v9.1 on a 15 minute setting: GBP/USD, USD/CHF, USD/JPY, AUD/USD, NZD/USD & USD/CAD. I am again faced with this incredible drawdown that this system seems to live with. To some of you 40% DD is ok, if you are getting the yield...but I am a 15% Intraday DD guy...I am already at 10% DD in under a day.

These are my four input changes from the defaults:

Lots: 0.01

SmallestLotSize: 2

TrendPeriod: 80

TrendtimeFrame: 15

DATE STARTED 08/22/07, BANKED DEMO PROFIT $204.16, DD:-$193.94

ES

I'm actually testing in demo account $3.000 with only 3 pairs. Obviously, trading 6 pairs would give you twice the DD I'm having, and a bit more (talking about %), as you're trading a $2.000 account instead of a $3.000 as I am.

CeroOnda

 

Sorry I had a look only for 5 min. so I might repeat something .

The graph shows wins, because it reports only closing trades, and not open trades. The fact is that winning trades are closed from this EA and loosing are not....

so the graph makes illusiion in your eyes...

 

ok...got it...I hit my 15% DD so I again will start a new Demo...with EUR/USD & USD/CHF...with 0.01 microlots and 2k.

ES

CeroOnda:
I'm actually testing in demo account $3.000 with only 3 pairs. Obviously, trading 6 pairs would give you twice the DD I'm having, and a bit more (talking about %), as you're trading a $2.000 account instead of a $3.000 as I am. CeroOnda
 
ElectricSavant:
ok...got it...I hit my 15% DD so I again will start a new Demo...with EUR/USD & USD/CHF...with 0.01 microlots and 2k. ES

My testing shows USD/JPY better than USD/CHF.

Just my opinion.

I'm attaching my statement too.

CeroOnda

Reason: