Pair Divergence Method

Thank too much for hard work pippa !!

Hi Bill,

Do you take double-divs (as the enclosed example), where you first get a signal for one pair, and then for the other? Taking positions with correlated (opposite positions for inversely correlated) pairs doubles the risk; at the same time, if the signal is there, why shouldn't it be traded...

MANY THANKS!

When I place the SBT Overlay Ind on the EUR/USD chart it works as it should, (inverting the USD/CHF). But when I load another chart (like GBP/USD) and then place ind on chart and change its pair to EUR/USD or some other pair it never displays the second pair. I've tried FXDD without the small "m" and IBFX with the small "m" but don't seem to be able to get second pair price on chart. Any ideas? I see other charts on thread with change so I know it can be done. Thanks Paul

I did as much as I could on the backtesting the past weekend.
I had some things at home that interfered, and it is extremely slow work, at best.

I didn't get to January.
vbro had previously tested January and had caame up with 20W and 12L.

Here's what I got for February through last Friday, April 17.

This was taking all trades. I applied the system rules mechanically. I let each trade run until SL or TP, even if there was a counter trade. I used 2xATR(14) for both SL and TP. Spreads are included in the results.
I tested a flat 20 pip stop. It worked great the first week or two, then fizzled. It wound up with the same results as the ATR.

February
16 W
15 L

March
30 W
26 L

April
16 W
5 L

Total = 82W / 58L
58.6 % W
41.4 % L

This is a little better than a breakeven system, as is.


I have only seen a handful of mechanical systems that were profitable, when the rules were strictly followed- maybe two or three in 30 years of trading. Most of what you see are systems that claim to be profitable, but exceptions are made while trading it.

Any mechanical system that does better than a coin toss has a solid base. A random coin toss will lose at the rate of (spread x number of trades).
If we had entered on a coin toss for 140 trades, We would have lost about 420 pips. We made a few pips with this.

I'm sure it can be made better- much better, by refining the divergences and not taking every one of them.
The losses seem to mostly occur with the little random bumps that meet the rules for a trade. There are many of these.

The problem is, that backtesting this thing is so difficult.
It is very subject to error and is very labor intensive and tedious. Before I obtained the indicator I used, I overlooked almost half the trades, while trying to test it.

This will make refining it very difficult. Any rule modifications will take a couple of days to test.

Anyway, there it is. Anyone who wants to improve it, is welcomed. ForceDevotee is on to something good.

I won't have much time to give it. It belongs to anyone who wants it.

I wouldn't take them, while trading.
The results probably wouldn't change much if you took them or not.
These pairs do not always maintain correlation. Sometimes they go their own way and counter trades would be profitable.

Fantastic job!!
You do great work.
This will be helpful.
Thank you.

Bill, does this mean you are no longer interested in this method?

No. I'm interested. I just won't be able to do much more backtesting. Not like I've been doing, anyway.

Could I just ask if I understand correctly? Were you using 2xATR for SL, and multiplying that by 2 (against the trend), and 3(with the trend) for the TP?

So for example, if ATR is 20, your TP would be 40 or 60?

I REALLY enjoyed looking at your sons page, the examples of his work were awesome, the Interview really gave me perspective. Really cool.

I am rubbish with creating indicators and that side of things, however I will backtest 2008 as a whole and maybe 2007 if I get a chance because I am manually backtesting other stuff anyway, hopefully before my exams come up .

However I have a problem (See Chart)(Yes sorry I am terrible with indicators)

The time we are testing it over is surely too little a time ?

I think it as a whole looks promising, however, I have a few points to raise that are open to anyone.


Point One) What we are essentially doing is waiting for divergence between correlated pairs.

We are waiting for them to go out of correlation (eg, USDCHF makes a higher high and EURUSD makes a lower high, there is selling pressure on EURUSD so that's the pair we go short on)

However would it not be wiser to . . . Go short on the other one ? Because if selling pressure is coming into one pair, the correlation will eventually correct itself and the USDCHF pair would have more "selling to come" with regards to it ?

We are already shorting the pair that has come under selling pressure, whereas it would be better to sell the one with no selling pressure already ?

So :

1) Divergence occurs
2) (Using the same example) USDCHF makes a higher high, EURUSD makes a lower high.
3) They both turn down on the next close.
4) Instead of going short on the EURUSD which ALREADY has been under selling pressure, we will short the inversion of the USDCHF - The CHFUSD , so we will basically go long on USDCHF instead of going short on EURUSD and we will then wait for TP at whatever.


Like, the whole idea is that of getting in at a better price.

Point Two)

A hedge strategy

Right so the same thing happens again, USDCHF makes a higher high, EURUSD makes a lower high - fine. This shows us that the USDCHF (it's inverted so it's the CHFUSD ) is out of corelation with the EURUSD.

We should short the CHFUSD (It's inverted remember) ( We are going to go long on the USDCHF ) And we will go long onthe EURUSD as well.

We then wait for the Opposite to occur (For them both to fall out of sync again, but this time in the other direction . . if you get me) And then we close the trade.

I have never, applied this but I will try to forward test it. I have no idea by how many say . . . "pips" they would fall out of sync by, but you have to pay 6 to just get the trade going in the first place.



I could well be talking complete and utter rubbish with my second point (And with the first the truth be told) But they are just some ideas I had in my head that I thought I would share.


Like if you look at whats happening from a candlestick point of view, when the pairs fall out of sync it is usually because an engulfing pattern has occured on one, or a recognizable candlestick pattern of some sort, and then what we are doing is entering on the break and close of it.


Anyway, people in the thread let me know what you think of my ideas, and feel free to comment on them.

I'm new to currency trading, so grab a grain of salt, but Bill's basic premise has been working like magic for me. I've been using it on Guppy against several other pairs. I haven't restrained myself to highs and lows that line up precisely.

I don't pretend to have any special insight, but I may have stumbled on something worth trying. I have the candlesticks showing, but very faintly. (I wish I could see them on the second pair, but so far haven't found an indicator to do so). I look for a trade on the M15, and once I see one forming I switch to the M5. Since we are waiting for the candle to close in the correct direction, this gets you into the trade at a better price. I wait for the M5 candle to close in the right direction, and pull the trigger. Maybe it's beginner's luck... not sure, but it seems to work well.

This is a great job bro, keep it up! Any dig up comes back to FF.

i would like to thank you and others for posting and working on this method. I have opened a live account today to test it. I think for a manual system we might want to watch the p/l when deciding exit point after opening a position, rather than setting a fixed tp and sl point value. Even the mechanical back testing result seems falling short of expectation, in real manual trading the result might be far satisfactory. I am the "grab and run" type of trader, i do not set a big tp and my loss is small.
I feel at ease with this methodology, so i want to trade it live.

Here are two indicators I found that you might find useful.

Second Chart, at the bottom, displays a second currency pair as small candles in an indicator window below your current main pair. This is showing USDCHF on 5 min below EURUSD on 5 min.

Custom Candle comes from a strategy called 3-in-a-box. It draws large candles over the normal candles you see on screen. For example, here we have larger 15 min candles on top of 5 min candles. You can fiddle with the colors to get things to be less confusing.

WARNING: If your strategy is based on moving averages on the longer timeframe, like 15 min, remember that if you are watching activity on 5 minutes your MA won't look right. And just multiplying out the 5 min MA x 3 to match the 15 min won't give the same result.

Good luck!

Hi pippa, does your indicator work for mini account? I dun see the yellow SMA lines showing

I made a bad discovery and a good discovery last night. I thought I was finding phases where the 1 min or 5 min for one pair was moving a few bars sooner than the other pair, perfectly predicting movement a few minutes early. But, sadly, I figured out that during slow periods, if there is no volume for a 1 min or 5 min bar, then MT4 just leaves out a bar (I think that's why.) So the further back I looked the further the lines were really just out of sync because of missing candles.

To fix it I changed the indi to match date/time stamps for each bar to verify the bars are synced.

The good news is that even on 5 min and 1 min there are now many, many more matches showing up than ever, even on short time frames. I also made a few other important improvements:

1) Bars are matched by date/time stamp for perfect syncing
2) I added a secondary pair variable so you can type in any pair name you want. For micro we're seeing that you'd need to maybe type USDCHFm or whatever the exact name used by the broker is.
3) I changed the tweak variable to be in pips. If you need to move the secondary line by 100 pips, you type 100 instead of .01. If you're using 5 decimals, and need to move 35 micro pips you'd type 3.5.

Wish list: make it so you can use a secondary pair that isn't inverted

I love this system! Thanks for the hard work and sharing of it!

I have two trade ideas for GBP/USD based on USD/CHF and EUR/GBP. Both say long GBP/USD.

What do you think?

Jigsaw,

This is off topic, but nice avatar & I want ask where you got the idea from. You are correct....some people just don't appreciate life, but I do & try to live each day to its fullest. I think that was the theme that SAW had in mind along with some get even tactics. He did raise a valid point about people not appreciating life, but using another method to get the point across would have been better.
________________________

AceTrader

Thanks for the indicator, it's beautiful. Do you have any idea what I could be doing wrong? When I try to put EUR/JPY on an EUR/USD chart, I have to flatten the EUR/USD almost completely to see the highs and lows of the EUR/JPY.

I literally just got the idea from SAW initially, I saw that movie first when I was pretty young and the idea was like a spectre in my mind which has refused to be exorcised since then.

I understand that the whole "torturing people with violence to get them to appreciate life thin is NOT A GOOD IDEA" (Like obviously I do not agree with it) but in ways you have to be that extreme to shake people out of what they take for granted.

It's like in fight club when Tyler Durden puts the shop assistant on his knees and points the gun to the back of his head telling but he never had an intention of killiing him, it was just to show him how sweet life really is. I think his quote was "When he wakes up tomorrow his breakfast cereal will be the best he has ever tasted".

I know that this is not . . "Achievable" lol , in real life, but I like having stuff around like this - http://www.flatrock.org.nz/topics/od...ing_a_meal.jpg just to remind me how lucky I am. That child is crawling to the nearest food centre over a kilometre away, nobody knows what happened to her in the end.

I guess some people may find looking at that kind of macabre, but for me it instantly places me back in life, where all my petty worries dissapear,. . And I instantly see once again see life anew.

Wow, I ranted there heh

Take care Ace,
Jig

EDIT - Again, sorry for the off topic