What makes a ‘good’ forex trading system?

Love Trading Buy/Sell Arrow Signals? Try This!

During the Summer months I usually back away from my computer a little and as I can not really afford to take time off ‘trading’ I seem to reduce the amount of forum and website posting I do. This allows me more time outside to enjoy the Summer but it also gives me time to think of new trading ideas.

I am always trying to think of better ways to trade or better forex trading systems for myself more than anything. I check binary bets and other fixed odds bets regularly to see if there are any trading opportunties. I collect end of day data and sit at a spreadsheet for hours manipulating data until something stands out. Quite often what stands out is that the markets are pretty balanced and most ideas end up in the bin as they usually end up with an evens probability or winning with and evens risk to reward ratio. Effective a ‘waste of time’.

In my search for mechanical forex trading systems I have learnt that there is a limit to how profitable a system can be. This, based on my theory, is because of the nature of the markets that you can not increase the strike rate without effecting the risk or reward. And there is a natural balance between the three. This means that you will imo never find a mechanical day trading system that has say 85% winners whilst risking less than you can win. Maybe someone will prove me wrong but they haven’t so far and I doubt they will in the near future.

So this brings me to my question. At what point do you consider a trading system to be worth trading? What happens if I say I have a mechanical 62% strike rate trading system with a even risk to reward ratio? Sounds good but the reality is tha (taking into account days where neither stop nor target are hit) could only yeild you 7 to 8 pts a day!!

Further such a system you would need to trade a one or more contracts to make it pay financially (if you are trading for a living) but then the nature of the game would mean you need to be able to take 3 or 4 losses in row and thus could easily be up to £1200 loss in week. I can feel you shaking your head but the system is profitable and consitant.

So what is wrong? What strike rate do you want? And at what risk to reward?

Yes, I know you see people on forums never taking losses etc but in reality you are dreaming if you think you are going to trade for a career and not take losses. So I would like to know…what is the minimum you would expect from a trading system before you would trade it?

Answers in the comment box please.

4 Responses

  1. FX_Trader says:

    For me a good trading system has a minimum 1:2 risk/reward ratio. I’ve programed over a dozen FX robots based on systems that I’ve learned or created and having extensive back & forward testing have discarded most of them because they may not hold over extended periods of time. The 2 systems I do use that I’ve created (1 technical & 1 fundamental based) both have a risk/reward ration of 1:3 and have an input that allows me to increase or decrease my risk/reward ratio. These systems allow me to have more lossing trades than winners and still remail profitable over a given period of time due to the risk/reward ratio.

  2. Jason says:

    good man. With a risk to reward ratio like that it would be very hard to lose.

  3. Tom Leeson says:

    I am a trader and developer, I love to trade the GBPUSD, I find that modelling a robot on my discretionary systems is the ideal solution and an ongoing quest – sometimes it will take trades obediently whereas i would have filtered and vice versa however it is hard to put my brains biological parrellel super computer which we all have into a robot.
    I am better at trading than my robots currently ha! But only for 1 system in all the others they are superior as they hav eno emoption and never miss a trade :-)

    Keeping track of and consciously learning what we are doing when we trade is important, vebalising and having written rules is a key to system trading. I find that a system for me must have a positive expectancy, the winning ratio i like to see around 50% and the risk reward must be more than 1:2 or I am not happy psychologically and therefore cannot trade it.
    A 35% trend following system with a large but random risk reward is not my cup of tea either. I like to see what i am getting, the only way I can do this is to trade the daily session ATR using a range trading technique, so i am going from one end of the daily candle to the other, my ideal target is a reverse signal or target which is >= 10 day average ATR of the daily time series foe the GBPUSD this is around 150pips this is the average maximum favourable excursion. I aim for 100% of the MFE, with a mean reversion hybrid range trading HILO system this is possible with a trend trading it is never ever possbile as you take a slice out of each move. Each system is the opposite of the other.
    Mean reversion has a static technical based target, winners are cut and losses too,
    trend trading systems have dynamic target ie a stop and reverse – winners have to run.

    If you combine these ideas into perhaps a HI LO Range trend trader then you take trades with the trend, your target is not the mean it is the other side of the mean. It is the high watermark starting from the low watermark.

    So i have learnt to use a system that allows me to see exactly what is on offer. I can see i ahve 150pips on offer and i dont like a stop bigger than 15pips. I aim for 10pip stop and try to get my risk reward 1 to 15, that is the ideal. So then lets look at the winning ratio – 35% is a typical trend trading system, well this was no better originally so I had to optmise the system to increase the winnin ratio without reducing the MFE or how much i can take from it. I found this was possible using combinations of solid tehcnicals, trading with the trend or counter at extremes, with confuence of higher time frames. I have not found this easy to code into a robot yet! I have not consistently obtained more than 1 to 4 risk reward in a robot as yet, 1.2 to 1.3 yes… The difficulty lie within the aggregation of information from higher time frames and trading time frames… into a single signal. The robot has become a signal aggreation order execution engine, it can listen to many signals, setups, trend direcitonal filters from many charts, backtesting it is not easy due to technical limitations…. my quest continues… The great thing is that my journey for this oober robot means that i take breaks from the market and do not over trade.

    constant position size is imporant is it lets you see the real profile of a robot, however you can increase your risk reward when you have found a stable reliable parameter set
    with dynamic position sizing controlled by the stop size and max risk factor.
    For example if you have a money management rule of a max of 2.5% risk which is 60pips and you have a trade which which can be 15pips to get the stop above that 9month high and session high – then 2 units trading 3 lots per unit peeling 1 unit off at 20pips, moving to break even and letting the 2nd unit run to the 100% ATR and beyond will dramatically improve your risk reward….. however it is vital you are ok with 1 lot before trying such a stunt :-) So tip get the RR profile working with 1 lot first…then backtest and try 2 lots….

  4. Jason says:

    Thanks for you insight.

Free Trading Systems. Metatrader Indicators.Expert Advisors.Forex Forum. Privacy Policy    
© 2009-2010 Great Trading Systems. All rights reserved.