Trading using only MA's?

[DaveB]

What are the best two Moving Averages to use together on, for example, a 2hour time frame or any other time frame and what would they be set at, eg exponential, simple etc.

Thanks
Dave

[craig66]

There is no best combination of MA cross settings, every combination has drawbacks in different market conditions. If you are determined to use MA crosses I would do it on a H4 or daily timeframe and just try different combinations until you see something you like. I don't think pure Moving Average cross systems really work anymore, too low of a MA setting and you get whipsawed to death, larger MA settings don't tend to cross until the market is going in the complete opposite direction. As for type of MA, it really does not matter in the long run.

Search for the 'bunny cross system' if you want an example of a MA cross system, but read the thread carefully because there was alot of discretion used to make the system a 'success'.

If you are just starting out, I would give yourself a head start and just forget about MA cross systems (and multiple indicator systems in general, adding another statistically insignifigent indictor does not make a better system), I spent so much time testing all sorts of TA cross systems & none of them work, I quick internet search will quickly reveal the futility of these approaches. RSI, Stochastic etc, it's all s**t. All you can do is trade in the general direction (taking into account news & support/resistance) of the market and give yourself an edge though risk & money mangement.

Sorry for the rant, I'm just trying to save you the sea of crap I had to wade though.

[ramdas]

There is no best combination of MA cross settings, every combination has drawbacks in different market conditions. If you are determined to use MA crosses I would do it on a H4 or daily timeframe and just try different combinations until you see something you like. I don't think pure Moving Average cross systems really work anymore, too low of a MA setting and you get whipsawed to death, larger MA settings don't tend to cross until the market is going in the complete opposite direction. As for type of MA, it really does not matter in the long run.

Search for the 'bunny cross system' if you want an example of a MA cross system, but read the thread carefully because there was alot of discretion used to make the system a 'success'.

If you are just starting out, I would give yourself a head start and just forget about MA cross systems (and multiple indicator systems in general, adding another statistically insignifigent indictor does not make a better system), I spent so much time testing all sorts of TA cross systems & none of them work, I quick internet search will quickly reveal the futility of these approaches. RSI, Stochastic etc, it's all s**t. All you can do is trade in the general direction (taking into account news & support/resistance) of the market and give yourself an edge though risk & money mangement.

Sorry for the rant, I'm just trying to save you the sea of crap I had to wade though.

What you have said about MA cross is correct, I have also tried many MA Cross stratergy and at the end realise crosses are like Death cross.. enter late and reach at Graveyard.

But Moving average works...But it is the way how we uses it.
I have posted one setup on Forex-TSD forum where i use Simple moving average as Support and Resistance. you can have a look at that thread.

here is link
http://www.forex-tsd.com/suggestions...ple-setup.html

Best Regards
Ramdas

[DaveB]

Thanks for your comments guys, I will bear in mind what you have said. Maybe trend lines and support and resistance is the best approach. I will see how I get on using those.

[initialsBB]

What was written above really amazes me since my results are completely opposite.
I've been writing trading systems for 1.5 years now and while there are many concepts that can be profitable, MA crosses for certain currency pairs (not all of them though) is one of the most profitable approach. After having written so many systems, i did not achieve any dramatically better results than those from MA crosses. So even if i had used only that, it would've been sufficient.
MAs have been around for a very long time, everyone uses them and they still work! They are trend-following and as long as there are trends they will work.

[DaveB]

Thanks initialsBB, I do like the idea of using MAs because they are simplistic and easy to read. Having said that, could you recomend any pair to use for cross over signals for a particualr time frame that would be overall successful. I do appologise in advance as I have little experience of MA crossover trading.

Regards,
Dave

[initialsBB]

The best way to determine the length of the MAs or whether that pair is suitable for MAs is through backtesting, preferably with TradeStation. You can test 2 types of exits - fixed stop loss/take profit values or trailing stops with highest high, lowest low function. If you use constant values you should go for at least 1:2 risk/reward ratio. And not to mention that you need at least 100 trades.
I must stress that "visual" or manual backtesting can be extremely deceiving - very often something that looked good on charts returned poor results on backtests. For me the most important parameter is Return on Account (ROA). I test wide range of values and plot ROAs in an Excel-style chart. If there are sharp differences among close values then probably MAs are not good way to predict the future for that pair. Generally i look 15% above and bellow certain length - if performance does not chage dramatically then that's a good sign that it can predict profitable opportunities.
Very good pair is GBPUSD. Each year it generates two trends of approximately 800-1200 pips. Just to catch those two would suffice to end the year with profit

[craig66]

...which you will need for all the whipsaw trades that will be generated. Perhaps I should clarify slightly. MA's are better all that other c**p, I use MA's, just not MA crosses. If it works for you, cool! But I just can't get backtest's which show (very) profitable systems.

[permanentjaun]

After months of looking at the market and trying to create a system I've come to the conclusion that it's pointless. If you're looking for a MA system to make your decisions easy then don't bother using them.

I use a system of S/R and trendlines. It's real easy to trade. It takes me about 2 seconds to draw a line, literally. Then I can see where my trades are in the future. I know when to call it quits on trades. I know when to reverse. It takes very little time to draw and is very accurate. I've recently come to this conclusion and it's almost boring since there isn't much need to come to forums and talk about trendlines and S/R. Just realize that the most successful trading isn't going to come from an EA unless you can't handle to follow a system on your own.

[DaveB]

Hi permanentjaun, that sounds interesting as it was mentioned earlier in the thread. May I ask a question about trend lines, you obviously need two points to draw the liine so if using for example, a smaller time frame, say the 5min, I presume you would take the second candle as your second point and then draw it. Would this be correct. The reason I ask is because the 5min chart does provide plenty of short term trends and support and resistance.

Regards,
Dave

[walt8880]

After months of looking at the market and trying to create a system I've come to the conclusion that it's pointless. If you're looking for a MA system to make your decisions easy then don't bother using them.

I use a system of S/R and trendlines. It's real easy to trade. It takes me about 2 seconds to draw a line, literally. Then I can see where my trades are in the future. I know when to call it quits on trades. I know when to reverse. It takes very little time to draw and is very accurate. I've recently come to this conclusion and it's almost boring since there isn't much need to come to forums and talk about trendlines and S/R. Just realize that the most successful trading isn't going to come from an EA unless you can't handle to follow a system on your own.

I'm just starting also, but after studying a lot of strategy ideas and indicators and such, am coming to the same conclusion about trend lines, and S/R with perhaps a couple MAs thrown in for confirmation.

If I can ask, what time frame are your trading in and hojw do you estimate Stop Limits?

[permanentjaun]

daveb,

Right now im playing with an MA trendline system. I have a chart with a Hull MA 3 plotted against the chart. This way the volatility of the market is smoothed, but the MA still keeps up with movement and doesn't lag too much. I may end up dropping the idea and just using the actual real time candles themselves. It's just something to play with.

walt8880,

I don't really have any stop limits. I plot my trendlines, S/R, or channels and figure where on the chart price would have to go for the trend, s/r, or channel to be broken. I give price enough room for its typical movement so I don't get stopped out on noise.

Right now I am trading on a 1 hour chart. I like to back the chart out though so I see as much info as I can. There's no point in only looking at 10 hours of data on a chart when the trends form over 30+ hours. If I could back a tick chart out to show a week I probably would. That's something I'm going to request for the next version of GFT dealbook 360; the ability to zoom out of charts infinitely. They can't now because they only allow a maximum number of bars on a chart, but I think they're changing that. Matt

[permanentjaun]

BTW, I wouldn't use MA's for a confirmation signal. If you must use indicators use MACD or RSI to confirm a signal. They're well known and somewhat reliable.

[initialsBB]

After months of looking at the market and trying to create a system I've come to the conclusion that it's pointless. If you're looking for a MA system to make your decisions easy then don't bother using them.

I use a system of S/R and trendlines. It's real easy to trade. It takes me about 2 seconds to draw a line, literally. Then I can see where my trades are in the future. I know when to call it quits on trades. I know when to reverse. It takes very little time to draw and is very accurate. I've recently come to this conclusion and it's almost boring since there isn't much need to come to forums and talk about trendlines and S/R. Just realize that the most successful trading isn't going to come from an EA unless you can't handle to follow a system on your own.

That's exactly what my starting idea was. I don't know if you know that trendlines can also be curvilinear . Moving averages automatically plot curvilinear trendlines and anticipate support/resistance levels. I use another lagging average to calculate the percent of penetration needed in order to initiate position. In this way it's much easier to code and check the credibility of the idea

[permanentjaun]

Yea I'm sure trendlines can be curvelinear. That get's a little confusing though. I'd rather up/down trend lines to create sloping channels or S/R. I think if you use a curvilinear trendline you're curve fitting too much and might get false breakouts. Not sure how moving averages really anticipate levels since they don't extend past the current candle. At that point youre essentially drawing a trendline, so why not just draw a trendline ya know?