Are mechanical systems robust on forex markets

[gekko_101]

Hi everybody,

I have been watching the forex markets for about 12 months now and for the last six I've been writing my ideas into Tradestation. About 5 months ago I wrote what I thought would was a pretty solid system - low reliance on optimisation, worked on Swiss, Euro and Pound without modification and had pretty good stats.

I'm pretty conservative so I watched it for the next three months and for each month each of the currencies produced pretty solid gains (av around 150 points a month, which I think is good). So then I start trading it and the returns have fallen right away (average of 10 - 20 points a month) and the drawdowns have been pretty big.

So my question is this - have other people found this with mechancial systems, particularly forex systems. I mean are they just going through a rough patch or are all systems I build only going to work for a limited time until the market participatants start figuring it out.

How have other people gone with robustness of their systems - do good systems turn bad or was it just luck that they were good in the first place.

I appreciate any comments or advice,

gekko

 

[wasp]

I know exactly what you mean...

Happened to me at first... perseverance is the key. IF you know that you will win more overall than you will lose you just keep going and believe.

You will lose more in costs, slippage and quieter markets but it should work itself out in the long term.

In theory...!

 

[philhackett]

I use a three moving average system using Fibonacci numbers as follows.
EMA @ 5 coloured black;
EMA @ 13 coloured red;
EMA @ 34 coloured pink;
Bollinger Bands coloured black

When the Bollinger expands and the 34 starts to slope, indicating a trend, enter a trade when the 5 crosses the 34. When the 13 also crosses the 34 consider adding to your trade.

When the 5 dips through the 13 exit and stand aside until the 5 either re-crosses the 13 to continue the trade in the same direction or crosses the 34 to start a trade in the opposite direction. If Bollinger has squeezed to narrow bands and the 34 is horizontal, wait for new market enthusiasm.

If you have Heikin Ashi candles you will find this is a very easy system to use and you will be in the market most of the time, only standing aside when the 5 is between the 13 and 34 which gives you time to make a cuppa.

Remember that moving average systems are the most successful systems. They may not be trendy but they work far better than all the expensive software being touted in the market these days.

Back test my 5 13 34 system and you will be able to estimate your daily, or in your case, nightly earnings.

If you feel that you need indicators as well, set a Macd (without histogram) at 8 13 9 to mimic Joe DiNapolis Macd and add a slow stochastic as well. Go to eSignal university learn how Joe uses the Macd and stochastic as trading tools, but I suggest that once in the trade you remain in until a crossover takes you out. It means gritting your teeth but long term will be very profitable.

I hope that this helps. Please let me know how you get on with it.

Sincerely

Phil

Hi everybody,

I have been watching the forex markets for about 12 months now and for the last six I've been writing my ideas into Tradestation. About 5 months ago I wrote what I thought would was a pretty solid system - low reliance on optimisation, worked on Swiss, Euro and Pound without modification and had pretty good stats.

I'm pretty conservative so I watched it for the next three months and for each month each of the currencies produced pretty solid gains (av around 150 points a month, which I think is good). So then I start trading it and the returns have fallen right away (average of 10 - 20 points a month) and the drawdowns have been pretty big.

So my question is this - have other people found this with mechancial systems, particularly forex systems. I mean are they just going through a rough patch or are all systems I build only going to work for a limited time until the market participatants start figuring it out.

How have other people gone with robustness of their systems - do good systems turn bad or was it just luck that they were good in the first place.

I appreciate any comments or advice,

gekko

 

[SvenFoster]

very interesting mr phil. could you clarify which timeframes you feel work best?

I use a three moving average system using Fibonacci numbers as follows.
EMA @ 5 coloured black;
EMA @ 13 coloured red;
EMA @ 34 coloured pink;
Bollinger Bands coloured black

When the Bollinger expands and the 34 starts to slope, indicating a trend, enter a trade when the 5 crosses the 34. When the 13 also crosses the 34 consider adding to your trade.

When the 5 dips through the 13 exit and stand aside until the 5 either re-crosses the 13 to continue the trade in the same direction or crosses the 34 to start a trade in the opposite direction. If Bollinger has squeezed to narrow bands and the 34 is horizontal, wait for new market enthusiasm.

If you have Heikin Ashi candles you will find this is a very easy system to use and you will be in the market most of the time, only standing aside when the 5 is between the 13 and 34 which gives you time to make a cuppa.

Remember that moving average systems are the most successful systems. They may not be trendy but they work far better than all the expensive software being touted in the market these days.

Back test my 5 13 34 system and you will be able to estimate your daily, or in your case, nightly earnings.

If you feel that you need indicators as well, set a Macd (without histogram) at 8 13 9 to mimic Joe DiNapolis Macd and add a slow stochastic as well. Go to eSignal university learn how Joe uses the Macd and stochastic as trading tools, but I suggest that once in the trade you remain in until a crossover takes you out. It means gritting your teeth but long term will be very profitable.

I hope that this helps. Please let me know how you get on with it.

Sincerely

Phil