Why exactly do EMA crossover systems not work? | Page 4 | Forex Factory

Options

Go to
Permalinks
Print Thread

Similar Threads

Help. EMA Crossover with different EMA cross for exit 0 replies

EMA crossover (ema 14 & 34) 4 replies

Please help code an EMA Crossover 12ema x 24 ema cross 1 reply

Systems, Systems and Systems... 1 reply

EMA Crossover Systems 1 reply

Attachments: Why exactly do EMA crossover systems not work?
Exit Attachments

Why exactly do EMA crossover systems not work?

  • #61
  • Oct 21, 2010 1:26am Oct 21, 2010 1:26am
EMA crossovers don't work because there is not efficient indicator that gives definitive signals. Indicators must be interpreted and used correctly. If the market were as easy as "when this line crosses that line" everyone would be billionaires by now.

I always have a MACD up on my chart, unfortunately, I find that very few can explain what the MACD is. When it's high above the 0 line the bull trend is coming near an end, when it's low below the 0 line the bear trend is coming near an end. The crossover for a (12,26,9) has too much lag and a (6,12,5) or scrapping it for a TRIX(5,3) has too many false signals. Never use indicators only for buy and sell signals.
  • #62
  • Oct 21, 2010 8:35am Oct 21, 2010 8:35am
Quoting DKallDAY

I always have a MACD up on my chart, unfortunately, I find that very few can explain what the MACD is. When it's high above the 0 line the bull trend is coming near an end, when it's low below the 0 line the bear trend is coming near an end. The crossover for a (12,26,9) has too much lag and a (6,12,5) or scrapping it for a TRIX(5,3) has too many false signals. Never use indicators only for buy and sell signals.
Including yourself, MACD is simple, get 2 EMA 1 fast and 1 slow, then you subtract the 1 from the other, it's merely the difference between the 2, not magic, it means currently the market is possibly going in a direction, but it doesn't mean the market can't change direction.

It also has a average of the value for well, no real logic other than why not.
Nothing to it, but to do it!!! Stick to the plan FOOL!!!!
  • #63
  • Oct 21, 2010 8:52am Oct 21, 2010 8:52am
Wow I see someone resurrected a zombie thread.

The OP said MA cross systems "don't work" without specifying the way in which they "don't work".

Fact is they do work, but may not be for everyone. MA crossing systems fall in the realm of trend following systems. These systems have fairly low win rates (50% or less) but can have positive expectancy if you're looking for high reward:risk trades.

People are generally drawn to systems with high win rates because they're seen as "better". But a high win rate system can still have negative expectancy.

I suspect the OP was looking for an MA cross system with a 90% win rate. These don't exist.
You are in a maze of twisty little passages, all alike.
  • #64
  • Oct 21, 2010 11:39am Oct 21, 2010 11:39am
Quoting Turveyd
Including yourself, MACD is simple, get 2 EMA 1 fast and 1 slow, then you subtract the 1 from the other, it's merely the difference between the 2, not magic, it means currently the market is possibly going in a direction, but it doesn't mean the market can't change direction.

It also has a average of the value for well, no real logic other than why not.
I'm not sure I followed much of that reply... hahaha can you please explain again?

Do you mean "including yourself" as having human interpretation rather than just waiting for signal crosses?

Quote
it means currently the market is possibly going in a direction, but it doesn't mean the market can't change direction.

Exactlllyyyy, which is why it should only be interpreted as a suggestion rather than a definitive trading signal.
  • #65
  • Oct 21, 2010 12:09pm Oct 21, 2010 12:09pm
Quoting stevegee58
Wow I see someone resurrected a zombie thread.
Hahahaha I didn't even notice that, 3 years old, well my posts are worthless haha.
  • #66
  • Oct 21, 2010 1:07pm Oct 21, 2010 1:07pm
I'm glad this thread was brought up from the grave as I'm now forward testing an EMA system. Simple crosses won't hold but with some discretion, and a little bit of thought EMA's can be as good as any other instrument in the Forex world.
  • #67
  • Oct 21, 2010 1:10pm Oct 21, 2010 1:10pm
If you think about it, every long-term trend has to start with an MA cross by definition.

It's the crosses without a long-term trend that you have to watch out for.
You are in a maze of twisty little passages, all alike.
  • #68
  • Dec 18, 2010 8:30am Dec 18, 2010 8:30am
Guys,

On the subject of EMA crossovers, there is no magic combination. 5 and 10, 10 and 20, 55 and 200 etc. The list goes on and on. I do know however that there is uniformity on the chart timeframes that are used to analyze currencies - 1hr, 4hr, daily, weekly and monthly. Plot EMA's on your chart to reflect these timeframes. On a 15mn chart, place a 4 period ( hourly ), 16 period ( 4hr ), 96 period ( daily ) and a 672 period( weekly ) EMA. Add in fractals and pivot points. The 4 period EMA and the 16 period EMA crossovers are pretty reliable and when the 16 period ( or 4hr ) crosses the 96 ( daily ), you will regularly see a continuation of the trend. You can apply the same principle to higher timeframes e.g. on an hourly chart, use a 4, 24 and 168 combination and on a daily, use 7, 30 and 365. Turning points are often at the longer period EMA - this is where fractals are useful for either swing high or swing low. On the 14th of December 2010 - €/$ hit a high of 1.3497 - this was exactly where the 365 EMA was on the same day - the price then retraced over the next 3 days. Have a look at these combinations on your charts, it will help in your trading decisions.
  • #69
  • Dec 20, 2010 8:29am Dec 20, 2010 8:29am
Quoting mestx41
MA crossovers work when you understand them as a form of trendline, if you don't understand why they work, you'll never be able to use them. Yet I can say that they are the most important tool in my trading.
i use them as you do
  • #70
  • Dec 20, 2010 8:39am Dec 20, 2010 8:39am
If you purely buy and sell when they cross, be prepared for a killing in a chop market.
  • #71
  • Dec 20, 2010 8:56am Dec 20, 2010 8:56am
EMA cross = gr8 not to forget one's own trading plan
  • #72
  • Dec 20, 2010 9:05am Dec 20, 2010 9:05am
Why not just use a single MA and price? Here are the results of price crossing a 50 day SMA. these results have NO money management, it is an always in the market method. 50 SMA vs Price

Here is another MA cross over method that is still being used and remains profitable to date..... 77 MA Cross

Here are the results of that method for 2008. johnedoe has been banned since then but I have been tracking the method. EUR/USD
  • #73
  • Dec 20, 2010 9:10am Dec 20, 2010 9:10am
EMA's work quite well if you use them correctly. You need to know the disadvantages and advantages.

They can confirm a trend change, or tell you where the current trend is going. Some people use this as a buy signal.

The problem with that is, you get that signal usually somewhat late.

Another problem is that in a ranging market an EMA crossover could easily backfire.

Some EMA's are used as support and resistance.

It's really up to you how to make do with them.
  • #74
  • Dec 20, 2010 9:35am Dec 20, 2010 9:35am
Quoting tradpat
EMA cross = gr8 not to forget one's own trading plan
Well of course you have to have a plan...
  • #75
  • Dec 20, 2010 6:36pm Dec 20, 2010 6:36pm
1. Fake Price (smaller timeframe is the killer)

there are super duper a lot of fake price due to manipulation by banks and other sharks in the market

2. Some EMA and SMA lines have bouncing effect, therefore when u use the lines together they will generate a lot of false signal

3. Large institutions only use EMA lines as money lines, lets say price above EMA 60 is BULL and price close below EMA 60 is BEAR. Sometimes the price will be bouncing along this line and generate a lot of false signals.

4. When EMA crosses each other, it's usually 4 bars after the move started, therefore it is a lagging indicator, usually after 4bars there will be pullbacks which you'll see losses temporarily or if it's a fakey, then it will go the other way round
  • #76
  • Dec 20, 2010 6:46pm Dec 20, 2010 6:46pm
I rely heavily on MA crossovers, but if you don't know how to use them properly....they will crush you....especially in a highly leveraged market like FX. If the application of MA crossovers were easy and straightforward, there is no way I could profit from them. The trick is to figure out what causes MA crossovers to fail and then work backwards from there.
  • #77
  • Dec 20, 2010 6:59pm Dec 20, 2010 6:59pm
MA Crossover strategy would kill your account equity in a tight ranging market.

MA Crossovers tend to work best for trending markets but I still find it not an idea way of trading.

Whats interesting the see is traders using a MA Crossover strategy on low timeframes. (i.e. scalping, etc)

MA Crossover is ideally used on higher timeframes as lower timeframes moves too fast for such lagging indicators. Scalping using a MA Crossover will always put a trader 'behind' the market.
  • #78
  • Dec 21, 2010 6:45am Dec 21, 2010 6:45am
Quoting Jhig
MA Crossover is ideally used on higher timeframes as lower timeframes moves too fast for such lagging indicators. Scalping using a MA Crossover will always put a trader 'behind' the market.
Lol, no. Higher timeframes have bigger signal/noise ratio and periods that the market is not traded heavily is suppressed into number of smaller samples. Lower timeframes have more noise and different market "moods" so anyone trading lower timeframes would have no problem trading higher periods.

MA lags the same on a 1min or 1hr chart.

MAs don't work by themselves because you are using fixed period and that means you are generally telling the market is always behaving the same way because MA will have a static freq. response. Fortunately market behaves different a lot of the time but mostly the same as the immediate past.
  • #79
  • Dec 21, 2010 7:29am Dec 21, 2010 7:29am
Statistically speaking, the crosses of moving averages are more reliable the higher the time frame you test, considering that you adjust the targets based on the ATR of that TF.
  • #80
  • Dec 22, 2010 8:34pm Dec 22, 2010 8:34pm
They can work quite well; just add a little filtering & a healthy dose of shrewd.
My favorite crochet hook is size J. It's a J-Hook!
0 traders viewing now