Pair Divergence Method

What you did on one trade doesn't matter much.
I'll stick with what I've been saying.

What I did on any trade doesn't matter. However, if you go to the correlation history of the Euro vs GBP it is a fact that the GBP moves lower and has higher peaks. Absolutely keep doing what you have been doing. Just look at your trade and when they line up and you execute your trade look at the GBP and see how much more it made. Thats the facts.

With the 1 period MA on typical price and everything?

Your reasoning doesn't hold any water if someone is trading with fixed fractional money management.

No.
Regardless which pair you trade, your TP will equal your SL.

If you size your trades properly, you will make the same percentage of your account on either pair.

I have found that the higher low or the lower high, hit the TP more often than the other pair does.

billbs, as for your rules, this is correct? Do you consider this bar as a confirmation? This is 1 hour chart,

take this trade basing in price action was un-resistible.

I don't see a divergence.

looking now i see the second top on EUR was 1 bar later than the top on the inverted UC.

the second top on UC was not lower but equal. Even this was a reason for do not trade?

The two peaks being equal is okay.
One being a bar later than the other is not okay.

U/C buy for a loss this morning. October is 14-8 so far by my records

This E/U buy trade was off by a bar on the 1h, but lined up as a larger pattern on the 30m. In progress, but on a Friday afternoon, might have to hold it over the weekend. I don't know if I would do that with only a 42 pip SL...seems a gap against you could do some serious damage you wouldn't expect?

I never hold a trade over the weekend.

U/C buy on the last bar I would consider looking at for the day @ 19:00. In progress.

I probably wouldn't hold over the weekend either if I were trading real money, which makes me wonder when I would close them out, or if I would even take Friday afternoon trades at all?

Hi, traders!
This is my 1st post on FF. I hope, it will be useful.
I created expert for billbss's method. See attachment.
Algorithms used:
- find 2 peaks (or valleys) without any mini-peaks (valleys) between them
- open buy or sell with SL = TP = ATR*2

Backtesting not so beautiful as desired (I used M15 for EURUSD vs USDCHF from 1st June to Today) But this version is draft. So, may be a bugs here.

PS. Expert is autonomical and very fast -- it don't require any custon indicators.
PPS. As you know, Metatrader Tester can't do multicurrency backtesting. So, if you test on EURUSD, then deals on other pair will be ignored.

Hi All

I ran into this thread 2 days ago and just finished surfing thru it.

I trade emini futures but I do monitor FX on a daily basis, so I am very familiar with the pairs discussed about in here.

I too use strategies that are based on divergences - more complex types too - and will give you my 2 cents.

First of all, I interpret divergence as a Temporary Market Inefficiency. Big institutions have algos and systems based on exploiting those inefficiencies, that's why they tend to work.

However, I don't believe that trading divergences alone is the smart way to go or build a profitable system. With this I don't mean to bash the over 40 pages that have been written in this thread. Not at all. You guys have come a long way and what you have done is a wonderful job. Bill, you had a tremendous intuition and my hat goes off to you.

Now the big question is: provided that divergences "tend" to work and provide an edge better than a coin toss, you got to use this edge as a basis to build something upon if you want to maximize its potential.

According to my experience, divergences are a great ENTRY tool. That's about it.

If you want to take this further, you got to develop CONTEXTUALITY. S/R levels are important because if you believe that one of those tiny divergences is the REAL CAUSE of a 100pip move you are utterly wrong! The tiny divergence may be what triggers the move within a larger context.

If you plan on keeping this a mechanical system, then you can't expect much better results than a 55% win rate.

If you want to take this further, then you have the possibility to make serious bucks and develop a profitable strategy - not system -

It's all up to you and where you want to go from here.

I pretty much agree with you.

I stated a number of pages back that this system, is slightly better than a coin toss. That is, however, significant. It means it can be improved on, as it heralds a move in the predicted direction, more often than not.

I don't agree with you about S/R levels. I have found them to be subjective. Five different analysts will come up with five different sets of S/R levels. They will contradict each other and they all think theirs are correct.

Thanks for your advice, ItalianSharp! Help me open up my mind here a little bit...are you suggesting that we should manage trades with more discretion? I.e. be able to know which ones to hold for profit longer than 2xATR, or cut our losses shorter than 2xATR? Or were you suggesting an idea much bigger than that?

Bill, I understand that your goal is to keep this system as mechanical as possible in order to reduce subjectivity to almost zero. And I agree with you on this because I am a very rule-oriented trader too and I prefer to have everything lined out in front of me before I make any trading decision. In fact, whenever I take trades that slightly step out of my trading plan or miss a couple of rules, I tend to struggle psychologically because my confidence level is not the same. So even winning trades that I take using my discretion are not as good as they could be.

With that said, I strongly believe you got to develop a set of rules to determine in what context you are taking the divergence setup.

S/R levels can be discretionary but some are very common levels that everyone can see. Prior low/high of the day is a level everyone pays attention to. Prior open/close too. You can extent the same to weekly/monthly levels as well. These are KEY AREAS all traders pay attention to, don't you agree?

If you trade intraday, then you got to develop a set of rules to identify intraday S/R levels. Who cares if others may agree/disagree with your S/R levels. Didn't you develop rules to identify clear hills/valleys? And do you think the way you define hill/valley is the same as every other trader? No way. And you shouldn't even care about it. What matters the most is that you got rules to define everything. It takes lots of work and patience, but it's possible.

Now you may ask me: how do I know what are going to be the key S/R levels for the day? And how do I update the S/R levels during the day?

Since I am a big fan of logic and simplicity, I'll say that a key S/R level is an area from which the market made a clear visible move. I'd call Support any level from which the market bounced convincingly and Resistance any level from which the market retraced convincingly. You got to develop rules to define "convincingly" of course. Once you have a set of rules to define S/R levels, then you can draw them on your charts and closely monitor price action as soon as those zones are tested.

If there is anything that I've learned about the markets - any market indeed - is that they love to test those S/R levels over and over again. The biggest moves and nicest moves always occur at key S/R levels. Why? Simply because those are the levels where the biggest battles between bulls/bears take place, because the biggest volumes occur at those levels, and once either side is defeated the market will quickly run to the next level and start another battle. Rarely do we see markets slicing thru S/R levels unless the move is news driven and there is a MAJOR FUNDAMENTAL reason driving the market a certain direction.

Once you develop a way to define/identify S/R levels, I'd recommend you also develop a set of rules to determine whether the market is trending, ranging, or reversing.

It is my personal preference to avoid all countertrend trades simply because I am not interested in catching tops/bottoms. I prefer to leave that tough task to someone else with bigger pockets than mine. I patiently wait for the market to show me a bottom and reverse, and only then join the new trend by buying/selling the obvious retracement.

That is when markets trend and reverse. When markets range, then there is no trend or reversal so you either play the range or sit out until the market breaks out either way.

I don't mean to give a TA lesson in here because this is your thread and you are the boss.

I am only trying to help because I am a big believer in divergences and you've shown good intuition and skill to get this going. Now it's time to move on and make it a VERY PROFITABLE strategy. You can do it but you must add a few things into the mix.

Leaving it the way it is now is a shame. I bet you that most of the losing trades will either be countertrend or outside any meaningful S/R level.

Kind regards

IS

Well, trade management only scratches the surface of what I am referring to.

It's about understanding the context in which you take any trade that can make a big difference.

There only are three types of trades that I know of: countertrend, with-the-trend, trend reversal.

The best trades are with-the-trend and reversal trades - beginning of the new trend -

I am not good with countertrend trades and, honestly, I am not even interested in them. Too much work and sweat. I prefer to leave those to other traders.

If you take one of the divergence trades described in this thread without being aware of what context the setup appears in, then you are trading blindfolded. It's like driving a car in the dark with the lights turned off.

So if you guys want to take this strategy further and make it VERY PROFITABLE, you got to work on different things.

As I said earlier, the divergence is what triggers a move not the cause of the move.

Whew! Just finished reading the thread and setting up the template. Looks very interesting. Simple but powerful. I like the use of divergence coupled with correlation. Will forward test this week.

It's clear, a lot of hard work by dedicated people went into originating (Billbss) and enhancing this system.

Before I get started, could someone please explain how to draw the shortened trendlines that appear on a lot of the charts to highlight the divergences on both pairs? I am with IBFX.

Many thanks in advance!

Go into the properties of the trendline after you create it and un-check the "Ray" box on the parameters tab. You only have to do this once, and all subsequent trend lines should be "point to point."