# There are hardly any guaranteed edges in Forex trading - but here's one that will never fail.

#### WalletInvestor

##### Member
88 27
There is a problem with returning to average - it is eventually a self fulfilling prophesy, but the problem is it can take a lot longer than you think.
What about a casino roulette wheel? Over thousands of spins it will also 'return to average'. Lotto numbers drawn will also do the same but you can't build a strategy around it.

I agree it can take longer than one thinks, you can't predict the future, after all. But I think it is not correct to compare price movements to roulette or lotto as price movements are never completely random, therefore the things that affect each are completely different.

ChartSmart

#### trendie

##### Legendary member
6,795 1,364
An example of a trade closed yesterday on the 5 min EURCAD chart trading back to the 800ema. Price had been away from the 800 ema for a statistically relevant time and at the distance from the mean and at the rate of change of the mean the overwhelming probability was that price would return to the mean. ...

The EMA value used determines whether the price is above the EMA and you're looking for a short, and a different EMA, on the same chart, may be below the price, and you're looking for a long.
So, the EMA chosen sets your expectations of a long or a short trade?

PS: I think I know why I am going to like this thread!

50 2

#### ChartSmart

##### Junior member
11 6
The EMA value used determines whether the price is above the EMA and you're looking for a short, and a different EMA, on the same chart, may be below the price, and you're looking for a long.
So, the EMA chosen sets your expectations of a long or a short trade?

PS: I think I know why I am going to like this thread!
Yes, on any EMA the trade will always be back to the ema itself. So if price is above the ema the trade will be short and if price is below the ema then the trade will be long. The reality is that without the data its impossible to know which ema mean reversion is most likely to happen on. In your example above perhaps price must first move back to the shorter ema before going back to the longer. This is completely normal and not a problem, but the data will clarify which trade is a stronger set up

#### ChartSmart

##### Junior member
11 6
Yes, great question! The answer is a bit of both. It is for this exact reason that we need to know the rate of change of the EMA when calculating probability of mean reversion

Moderator
8,652 980
How are you making it different from just using the EMA on its own as the idea of an EMA is that the front end is weighted more to factor in the rate of change anyway ?

ChartSmart

#### ChartSmart

##### Junior member
11 6
How are you making it different from just using the EMA on its own as the idea of an EMA is that the front end is weighted more to factor in the rate of change anyway ?
The ema on its own gives you zero useful data. How long has price stayed away from the ema on average over the last 250 000 bars of data. What was the longest excursion? How does the current excursion compare to the average and the longest excursion. At the current ROC where is the current excursion projected to end in relation to previous excursions? What is the probability of success considering all these factors. The ema is simply the target - we need to extract the data to understand the DNA of the instrument - how does price behave in relation to its average in the past - this will allow us to make assumptions of how it might behave in the future

#### DallasSteve

##### Member
62 1
Doing some quick back testing there does appear to be an "edge" in this idea of reverting to the average. But the gains in my tests so far are in the range of 1 pip before spread, which means the broker would need to offer very low spreads and commissions. Generally that's not available to US traders. Any suggestions/ideas for trading a system like this from "the land of the free"?

#### Pat494

##### Legendary member
14,621 1,577
" Land of the free" ?
That's a laugh, the crooks of US gambling won't even let US citizens play poker on foreign websites.
As for double zero on the roulette wheel, that is a real con. One zero European style is bad enough !

#### wmram

##### Member
51 18
I think that if you allowed gambling in the US, a large percentage of those gamblers would destroy themselves. Look at the failure rate for short term market traders. "Land of the free" yes but with boundaries for your own good even if you don't see that way. Someone like you might well be able to handle it, but think about all those who would blow up.

#### DallasSteve

##### Member
62 1
But it's OK for the government to sell lottery tickets that are even a worse bet, with no limit to how much the foolish can waste on the tickets. And our US government places controls on Forex trading which have the effect of driving the small traders to brokers that charge a higher spread/commission and drain their accounts even faster (to 'protect' the weak). And don't get me started on medicine. You can't buy the drug you need without first paying \$100 or more to a doctor to get a piece of paper, here in 'the land of the free' and the home of the cowards. That's why medicine cost is out of control: because government is serving the interests of the doctors and drug companies.

#### wmram

##### Member
51 18
In the US if you are an elected leader, you're under constant pressure from media, opportunist, scoundrels, etc. so you work with what you have to work with. I certainly cannot explain the inconsistency in the various laws that are enacted on the federal, state, or city level, but I suspect that most of those laws are the result of a fair amount of bargaining. That said leverage 100x,200x, or more in forex market is going to help most of the people that want it.

#### brfrexloid

##### Junior member
12 2
But how long will it take for those average to be attained can never be estimated. So that then doesn't really help.

269 28

#### tomorton

##### Legendary member
8,378 1,335
Price always returns to it's average.

Always.

If this is true and we can figure out when it is most likely to happen, we have a strong edge on market behavior that has always and will continue to always play out

So what needs to be considered?

Well a lot of traders will tell you distance from the average is key. So if price is much further away from the average than normal, it will probably return to it fairly soon.

Of course this is nonsense. Complete and utter drivel.

On its own, distance from the average tells us nothing apart from the fact that we are in a strong trend or maybe experiencing a significant news event.

A better measure than distance travelled from the average is TIME spent away from the average compared to how long price normally spends away from the average.

Remember, if price always has to return to it's average, which it does, then the longer it has spent away from the average compared to normal, the higher the probability of it returning to the average sooner rather than later becomes.

This combined with distance from the average now starts to give us a very powerful indication of whether price is likely to return to the average sooner rather than later.

The last element which I won't go into to much detail as it starts getting a bit mathsy is the Rate of Change of the average. We can discuss this later on, but it is critical to understand

So, if you are trading mean reversion back to the average of price remember - distance from the average is a start, but TIME spent away from the average is a far more useful concept to master.

Once we know price has spent to much time away from it's average we can start initiating positions to profit from it's return to the average.

Remember, price always returns to its average. Any average. Always.

All the best

Justin View attachment 289307 View attachment 289308 View attachment 289309 View attachment 289310 View attachment 289312
Just need to give you some credit for this principle, which I didn't do adequately last year - time away from the MA.

My principal set-up component is the slope direction of the 50EMA. Obviously, if the slope is consistently upwards, that means that price is consistently above the MA and because its a 50-bar period, it has been up there for some time, probably about 20-25 days.

I also look at other MA / time-related metrics such as the total of consecutive recent weekly closes above/below the 50EMA.

Replies
4
Views
5K
Replies
5
Views
715
Replies
0
Views
430
Replies
15
Views
18K
Replies
3
Views
6K