What is a reasonable system | Forex Factory

Options

Go to
Permalinks
Print Thread

Similar Threads

What is a reasonable drawdown? 17 replies

Tell me if this is reasonable 13 replies

Reasonable fee for programming help? 19 replies

Are You a Reasonable Trader? 24 replies

What are reasonable goals? 8 replies

Attachments: What is a reasonable system
Exit Attachments

What is a reasonable system

  • #1
  • First Post: Aug 30, 2008 9:44am Aug 30, 2008 9:44am
I have been testing many systems I found on this forum. However I would like to use a system which is not so risky but still reasonable. I know different people have different criterias for that and that is why I would like to hear the opinion of other people about a system with the attached results. Do you think this is a risky or reasonable system according to you. I have tested it from 1-1-2000 till today. You can see the results attached. Any opinion is much appreciated.
Attached Image(s) (click to enlarge)
Click to Enlarge Name: Graph.jpg Size: 67 KB
Click to Enlarge Name: Report.jpg Size: 94 KB
  • #2
  • Aug 30, 2008 9:47am Aug 30, 2008 9:47am
looks sexy
  • #3
  • Aug 30, 2008 9:54am Aug 30, 2008 9:54am
Quoting Waterflame
looks sexy
My concern is that I may be too much conservative. the drawdown is 23%. Sure I can push it up a bit, but not sure whether this is wize. It will increase the profit more or less 2 times but on what cost. It still won't fail for the test from 2000 till today but lately the market is getting more volatile and I would like to leave some room for risk. But I am not sure how much is too much.
  • #4
  • Aug 30, 2008 10:25am Aug 30, 2008 10:25am
well, its all about a combination of things and how you trade and proceed your descision taking.

aftger all i have seen now, you are able to trade technically and fundamentals as well, with a system what fits you, and have a good money management...

i can say you are one of thus who have the little something in trading, congrats!
  • #5
  • Aug 30, 2008 10:32am Aug 30, 2008 10:32am
Quoting Waterflame
looks sexy
looks like a backtest...

Id be more impressed with a forward test.
  • #6
  • Aug 30, 2008 10:57am Aug 30, 2008 10:57am
i think he manual traded it, he didn't told that is an Ea backtest, and if, you are right, it would be another case, but i answered him for self performances, not for something other
  • #7
  • Aug 30, 2008 11:07am Aug 30, 2008 11:07am
Quoting bmilanov
I have been testing many systems.... I have tested it from 1-1-2000 till today. You can see the results attached. Any opinion is much appreciated.
If it were a forward test I don't think the picture would have a modeling quality.
  • #8
  • Aug 30, 2008 11:27am Aug 30, 2008 11:27am
It is an EA backtest. I have been trying to fine tune this system. I have a very similar one running on a forward demo for the last month and it is more profitable but I think it will fail one day and that is why I am trying to improve it. I guess my main question is whether 23-25% is a reasonable drawdown.
Also I don't believe the data in MT4 from metaquotes are very realistic - may be the last year or so to certain point. But if I look at the graph in the beginning it is going very very smooth and at the end is extremely choppy which makes me think the data is not very realistic - but I might be wrong.
  • #9
  • Aug 30, 2008 2:56pm Aug 30, 2008 2:56pm
Quoting bmilanov
It is an EA backtest. I have been trying to fine tune this system. I have a very similar one running on a forward demo for the last month and it is more profitable but I think it will fail one day and that is why I am trying to improve it. I guess my main question is whether 23-25% is a reasonable drawdown.
Also I don't believe the data in MT4 from metaquotes are very realistic - may be the last year or so to certain point. But if I look at the graph in the beginning it is going very very smooth and at the end is extremely choppy which makes me think the data is not very realistic - but I might be wrong.
I think that is a martingale system and then you know it is always risk.
  • #10
  • Aug 30, 2008 3:11pm Aug 30, 2008 3:11pm
Quoting bmilanov
Do you think this is a risky or reasonable system according to you.
How can you expect us to comment on whether it's a reasonable system if you haven't even told us the system rules?

And anyway 25% drawdown is way too much for my liking.
  • #11
  • Aug 30, 2008 3:36pm Aug 30, 2008 3:36pm
i don't think i'd take advice from someone that can't even spell "wise."

Quoting bmilanov
. Sure I can push it up a bit, but not sure whether this is wize.
  • #12
  • Aug 30, 2008 3:39pm Aug 30, 2008 3:39pm
i've got a few questions....

1) are you sure the EA is opening on the bid and closing on the ask, and vice versa?

2) is it a scalping EA (TPs/SLs less than 20)?

the graph looks great, but that's not the concern. you need to make sure the trades are realistic (that is, not based on the current bar's TA)

if so, your bet is that the currency pair's characteristic's will not deviate from it's previous 8 years.... that is what you base your risk on.

perhaps you should only risk 10% of your entire account per six months, and reevaluate. i do the same thing myself.


PS) Just noticed 9 mismatched chart errors on the report..... i would definitely look at the charts in the backtest and make sure the entries are where they are supposed to be.... hopefully you know the formulas
  • #13
  • Aug 30, 2008 8:33pm Aug 30, 2008 8:33pm
Quoting tdion
i've got a few questions....

1) are you sure the EA is opening on the bid and closing on the ask, and vice versa?

2) is it a scalping EA (TPs/SLs less than 20)?

the graph looks great, but that's not the concern. you need to make sure the trades are realistic (that is, not based on the current bar's TA)

if so, your bet is that the currency pair's characteristic's will not deviate from it's previous 8 years.... that is what you base your risk on.

perhaps you should only risk 10% of your entire account per six months, and reevaluate. i do the same thing myself.


PS) Just noticed 9 mismatched chart errors on the report..... i would definitely look at the charts in the backtest and make sure the entries are where they are supposed to be.... hopefully you know the formulas
To answer one of the previous questions - it is not a martingale system. Part of the money management is using an idea from martingale i.e. it uses certain number of lots and after a loss it uses a bit more for sometime, but after another loss it doesn't increase again.

The EA open and close are correct.

It could be considered in a way as a scalping as the TP varies I would say average between 5-50 pips but it is most of the time low. The TP is 1/3 of the average of the range for certain period of time. The entry also depends on the average range. A position may be open for just few minutes or sometimes months.

It is a grid system which is exiting depending on how many positions we have opened. The more positions we open the TP decreases and eventually it becomes negative.

It is only effective on EUR/CHF as it as ranging system. I briefly tried few other currencies but most are losing and none could come any closer to the EUR/CHF.

I have played with the settings and found out that the best results I get is when the gap between the trades is 50 pips - working for all 8 years. But if I change it to 40 or 60 the system doesn't work in some years. But then I noticed if I use a gap of about 80 -120 pips there are no losing years. Ofcourse the return is lower in this case but the risk is also smaller.

The return seems to vary a lot for some years it is 25% for some over 200% but in average a bit below 100% annually. I am not sure whether this is a reasonable return having in mind the drawdown of 23%.
  • #14
  • Aug 30, 2008 8:56pm Aug 30, 2008 8:56pm
hi again

if a lot of your TPs are below 15, the backtester will often give false results. in a nutshell, it can't tell if the price hits the tp or sl first, because they are both happening in the same 1 minute bar.... so it gives the better of the two outcomes almost always.

well i admire what you have done. it is very similar to my own ea development... and i wish you success.

just remember, past performance never guarantees ANYTHING. those beautiful ranging EURCHF markets can disappear without warning, crashing your system because it exploits your system's weakness.

Quoting bmilanov
To answer one of the previous questions - it is not a martingale system. Part of the money management is using an idea from martingale i.e. it uses certain number of lots and after a loss it uses a bit more for sometime, but after another loss it doesn't increase again.

The EA open and close are correct.

It could be considered in a way as a scalping as the TP varies I would say average between 5-50 pips but it is most of the time low. The TP is 1/3 of the average of the range for certain period of time. The entry also depends on the average range. A position may be open for just few minutes or sometimes months.

It is a grid system which is exiting depending on how many positions we have opened. The more positions we open the TP decreases and eventually it becomes negative.

It is only effective on EUR/CHF as it as ranging system. I briefly tried few other currencies but most are losing and none could come any closer to the EUR/CHF.

I have played with the settings and found out that the best results I get is when the gap between the trades is 50 pips - working for all 8 years. But if I change it to 40 or 60 the system doesn't work in some years. But then I noticed if I use a gap of about 80 -120 pips there are no losing years. Ofcourse the return is lower in this case but the risk is also smaller.

The return seems to vary a lot for some years it is 25% for some over 200% but in average a bit below 100% annually. I am not sure whether this is a reasonable return having in mind the drawdown of 23%.
  • #15
  • Aug 31, 2008 5:04am Aug 31, 2008 5:04am
Quoting bmilanov
To answer one of the previous questions - it is not a martingale system. Part of the money management is using an idea from martingale i.e. it uses certain number of lots and after a loss it uses a bit more for sometime, but after another loss it doesn't increase again.

The EA open and close are correct.

It could be considered in a way as a scalping as the TP varies I would say average between 5-50 pips but it is most of the time low. The TP is 1/3 of the average of the range for certain period of time. The entry also depends on the average range. A position may be open for just few minutes or sometimes months.

It is a grid system which is exiting depending on how many positions we have opened. The more positions we open the TP decreases and eventually it becomes negative.

It is only effective on EUR/CHF as it as ranging system. I briefly tried few other currencies but most are losing and none could come any closer to the EUR/CHF.

I have played with the settings and found out that the best results I get is when the gap between the trades is 50 pips - working for all 8 years. But if I change it to 40 or 60 the system doesn't work in some years. But then I noticed if I use a gap of about 80 -120 pips there are no losing years. Ofcourse the return is lower in this case but the risk is also smaller.

The return seems to vary a lot for some years it is 25% for some over 200% but in average a bit below 100% annually. I am not sure whether this is a reasonable return having in mind the drawdown of 23%.

If a EA just work in one pair I want not say it is good.
  • #16
  • Aug 31, 2008 5:13am Aug 31, 2008 5:13am
i will come right back and say it is EXCELLENT

every pair has a different ATR, trend characteristic, spread, and swap. it is the SMARTEST thing to identify the characteristics of a pair, and match your system to it. one size DOES NOT fit all in forex usually.

Quoting Heino
If a EA just work in one pair I want not say it is good.
  • #17
  • Aug 31, 2008 7:59am Aug 31, 2008 7:59am
Quoting tdion
hi again

if a lot of your TPs are below 15, the backtester will often give false results. in a nutshell, it can't tell if the price hits the tp or sl first, because they are both happening in the same 1 minute bar.... so it gives the better of the two outcomes almost always.

well i admire what you have done. it is very similar to my own ea development... and i wish you success.

just remember, past performance never guarantees ANYTHING. those beautiful ranging EURCHF markets can disappear without warning, crashing your system because it exploits your system's weakness.
Thanks Tdion,

After seeing your post I checked the trades and even thought many of them are with TP 10-20 I haven't seen 1 which is in the same minute. The open positions are open for an average of 30-50 min. This system actually is placing trades mostly after big moves and it is off the 30 min chart. I guess at that time the move has calmed down and that is why they last so long.
You mentioned you have a similar EA. Would you mind posting the results as I did from a backtest? I would like to compare and decide whether I should spend more time on it trying to improve it - or may be it can't do better than that. From what I've seen by fine tunning it the most in the last 3 years could go not more than about 75% annually. In the period 2000-2006 the profit is much higher.
Thanks for the warning - I know nothing is for certain in Forex.
  • #18
  • Aug 31, 2008 1:26pm Aug 31, 2008 1:26pm
well normally i wouldn't do this (don't want PMs asking me for it, etc...) this is the forward test for 2008.... a very good year....

i have tested it back to 2001.... some years are better than this.
Attached Image (click to enlarge)
Click to Enlarge Name: StrategyTester.gif Size: 6 KB
  • #19
  • Aug 31, 2008 11:02pm Aug 31, 2008 11:02pm
Quoting tdion
well normally i wouldn't do this (don't want PMs asking me for it, etc...) this is the forward test for 2008.... a very good year....

i have tested it back to 2001.... some years are better than this.
impressive. do you trade it live also ?
Twitter: @forexpokerpro
  • #20
  • Aug 31, 2008 11:14pm Aug 31, 2008 11:14pm
in the near future....

Quoting BBPI_fxtrader
impressive. do you trade it live also ?
0 traders viewing now