It happened to me too. the two pairs converged and touched and the trade was still negative. That happens because the one you are buying didn’t go up enough or the one you are selling didn’t go down enough or a combination of the two. But on the charts they converge and touch because the way we look at them on the charts is misleading.
The way the charts look when I overlay them with Vitrite, has never shown loss when the pairs touch…if the charts refresh, you loose the effect of the correlation you were trading, because that was relative for that time, and when it refreshes, it creates a different relativity…anyways, if I open a 20 pips trade, I actually wait for the profit to hit $1 then close, so I can be not even looking at charts…if you look back several post, jester has two screenshots demonstrating how the distance between the pairs can change just by changing settings, so it’s all about having the chart setup where most of the time the pairs are touching, correlated together, then have some times of separation, and again the screenshots kelton posted on the first few pages provide perfect examples of what you want the charts to look like.
when I open new charts, I open them at as close to the same time as possible at 1 min, already zoomed out, which shows 999 bars…then I start watching…I never close the charts until I am done for the day…I may have a trade still open, because it hasn’t come back to profit, but I’ll just keep watching the profit/loss until target is reached, and it always comes back!
The answer for this question is that it depends. If pairs are of the same value then when they toches visually you should have a profit. But for me, in practice that doesn’t happens all the time. What I do is to wait until they are in profit and close them.
For manual trading there are some scripts for metatrader that you can add to your chart telling them to close all trades when you reach a determinated profit. You can add that script once you are in and let it to close your trade when ready. That way this system can be entered manually and exited automatically.
I just wait. If your money management is correct, then your trade Will be on profit at some point in the future, just need to wait. I think I’m going to look for the script I said that closes the trades when on profit and post it here for all of you.
For your close trade script, medisoft, just in case in the future I figure out a way to run MT4, can it be programmable to close all open trades (multiple “baskets”) at a set $ profit, say $1 for me, and then if on my next trade I want TP to be $4 profit, change it to that?
I think that using a percent instead of money amount is better, isn’t it?
A percentage would be fine, just wondering how I can manage the closure of multiple trades on the same pairs on, say one opportunity, then only need to close one opened pairs trade on the next opportunity…for example if i have opened three baskets, 6 total positions, starting at 20 pips apart, the next at 40 pips, and the last at 60 pips, then want it to close all at the same time when the pairs touch again, but on the next trade being one that only gets to the 20 pip divergence before coming back to together, i’ll need to know how to customize each trade closure relative to the amount of positions that have opened…BTW, closing 6 positions one at a time on an iPad takes too long! Slippage!
Mmmh, that would need to place the trades using a “magic number”, I mean, you would need to drag and drop a script to place a trade that assigns a unique numeric ID to every trade, and the closing EA will follow the pair by that ID.
It is not very difficult, and maybe it is easier to trade, just drag and drop the script on EU chart, and the script will place both trades, one for EU and other for GU. Two scripts, one for buy EU sell GU and other sell EU and buy GU.
What do you think about that?
So, that script would open both sides/positions simultaneously with the desired lot sizes, while also having embedded the command assigned to the TP goal for that particular trade as a percentage, customized to a magic number? Then another trade could be started at the next greater “divergence point” with settings relevant to it’s own profit goal set to it’s own magic number…this would be nice…at least in the future, since right now I haven’t figured out a way to run mt4, and therefore have access to this.
That’s right.
Also you can use a VPS (Virtual Private Server) to host a Windows XP instance and run MT4 there, and control it with TeamViewer/VNC/LogMeIn
I’m going to look into the VPS thing…so, I would need multiple scripts, looking at 1% per pair for first open at 20 pips, then 0.8% for next open at 40 pip spread, then 0.5% at60 pips, with each order closing at it’s own relevant profit percentage when the spread touches, which I still need to figure out…
Now, if only there was some way to get these orders to trigger using “visual spread” on these overlaid charts, this would be the best…because today, for instance, the spread has been right on tight correlation all day, since I started, but just 2 hours before that, was two great opportunities for about 4% gain that I missed, plus more at 1:00 in the morning while I was asleep…I would like to take full advantage of this great system, but automating entries, to accomplish that, maybe impossible???
Are your 1%, 0.8% and 0.5% profit targets or position risks? If they are profit targets, what is your risk?
There is a very simple way to do this, with no indicators, or overlaid charts. Apologise for harping on about it, but, the way to do this is to not trade the synthetic EG (by trading a long EU and short GU). Trade the EG directly. The movements in the EG directly relate to the divergence between the EU and GU. It can’t get any simpler. If they diverge one way, the EG moves down, if they diverge the other way, the EG moves up. That’s it. You judge the divergence by measuring how many standard deviations the EG moves, and this can easily be done by using bollinger bands. Ok, another apology, you do need an indicator, you need bollinger bands…but it really can be that simple, and it would certainly make your scripts much simpler as well, if you want to automate this to some extent.
If you are not convinced about it, then take a look at my previous posts where I explained it, and then ask any questions. I’d be happy to answer any questions or doubts that you might have…
Note that from your description, I think I traded that exact event, very easily. Despite everyone being on holiday in the UK (me included 
), I had another 30 pips from this today. EG gapped right down on the open last night and went below -4SD and half way to -5SD. Amazing I thought… I am a bit wary of buying in a down trend, but at -4 or -5 SD, for me, those are good entry conditions. So, I bought when it started to come back up, right on the -4SD mark. Closed it out when it had reverted to -2SD for a very easy 30 pips.
Those are actual positions, so, 1% is 0.1 lot on $1000 account, may not be accurate percentage, but that’s for reference…now to your single pair strategy, it would only work good for me if after diverting to say, 2 SD’s, the price came all the way to the mean, which could come down after a drop, like what happened when the EG gapped down, yesterday…not so sure about that…
Also, why not trade every single other major with BB?
That is exactly what I am targetting - about 2Sd of movement. Sometimes 3 or 4SD depending on price action. However, if it moves against me, it means that the pair have diverged further. So, I can then take another position (I do the same thing as every one else. I take a position, and then if it moves against me, I take another position…).
My SL is calculated about 3SD away from my entry. If I am entering when the price is already at almost 5SD (as it was in this case), then the chances of it moving to 8SD is very very small.
Any random major won’t work because the vol is too great, so you can only do this with the currencies that are highly correlated. But, when they are highly correlated, I do use this method. Two other good examples are the AU/NU (trading the AN) or the EU/EJ (trading the UJ).
I also hold the opinion that, of all the currencies traded, the EU/GU/EG hold a unique situation because of the volume of these currencies. EU is the most traded and GU is 2nd or 3rd (can’t remember). So the EURGBP holds the unenviable position of trying to tie together the Euro and the Dollar…
Today was a VERY slow day, only 1 open trade, no closing trades 
Jedster, do you think that in addition of E/G, AUD/NZD, USD/CAD and EUR/USD are good pairs to look correlation errors?
checking [Forex - forexticket](Forex Correlation - Mataf<br /><br />I) found that GBP and USD have better correlation than EUR and USD, but EUR and CHF have good correlation.
Thanks to Jedster for the idea, how many periods do you look back in the bollinger bands, do you use M1 chart too?
Personally, I haven’t done anything with the USDCAD. I just took a quick look at it appears to have considerably higher vol, the swings appear to be much larger, so I’m not sure I would be too keen to start trading that. What is the correlated pair that you are thinking of in order to be trading the USDCAD?
The EG (from EU/GU) and AN (AU/NU) by comparison have a lower vol, and tend to plod along a bit, so I feel they are more suited to this strategy. However I suppose that also means that they suit my level of risk apetitite, whereas the USDCAD doesn’t.
I think of the EG as taking up the slack between the Euro and the Dollar. Of course, EG has its own economic influences, but I think it is generally quite dull. The British economy is so closely linked with Europe, it takes serious economic changes to make big changes in the EG. As a result, and very generally, I think of it as simply acting as a buffer between the Euro and the Dollar. That’s how I think of it anyway, but it is why I am confident to trade it for this strategy.
AN is in a similar position because of the relationship between Australia and New Zealand…
The GU has better correlation with what, compared to the EU?
For the CHF, you need to condsider what I posted last week about the cap. That will/is affecting all the CHF pairs at the moment.
I use 200 period on the 1H chart. I know that for equities, the more common method is to use a yearly period on the daily chart, to determine the standard deviations. However, what I want is a long period so that small moves don’t affect it unduly, and yet it will move slightly up or down over time if there are relatively long term divergences.
Note that I don’t always wait for price to revert back to the mean. In a trend, quite often that won’t happen (or not for a long time anyway). So, I am looking for 2SD or 3SD worth of movement, depending on what the entry conditions are…