Triple threat exit strategy

Trading Systems Free Forex Trading Systems
442

i think your recommendations for limiting losses early in the trade are excellent. some of the most frustrating days are those that move our way just enough to put us in, then reverse for a big loss. by controlling situations like that, i believe we will be ahead. it makes sense to rely on the quarters theory once we are out there and well into profit, but rely on tighter managment early in the trade.

“3. Once TP is hit, move SL on AIR & RETRACE pos to nearest quarter (already implemented).” i think on this point, we only want to change our thinking to “nearest PROFITABLE quarter” so that we don’t turn a gain into a loss. this is what i was discussing earlier about trades on 7-9 and 7-16.

thanks.

Read more: 301 Moved Permanently

443

we may find that basing monday trades on friday data works better. fortunately, your ea will let us look at it either way and adjust our delta entry points. i will do some more work on this. btw, oanda also provides saturday data. i’m not using oanda for these trades at this time, but what would the ea do in that case? would it still go back to friday for a range to trade on sunday or would it pick up that saturday data and base on that?

444

If broker provides Saturday candle (on daily chart), then I need to modify code and have to put a similar logic like that of Sunday candle. This is how EA works - on what ever TF chart you, EA gets current day’s shift on daily chart. If current day is Monday, and true for sunday candle then take 2nd previous candle else take just previous.

But for the time being, lets dont use Oanda. We will come back to it once we have our EA profitable and in good condition…

445

i ran some quickie tests for july using the ea as it is now. of course, these should be considered preliminary and prior to any improvements we might make, but can give us a baseline for comparison later. july was an exceptionally good month and august is punishing, but we hope for more july’s than august’s. since some of these currencies are not us dollar based, these results are in dollars, not pips: g/j -535, a/u +365, e/u +1215, g/u +399, e/j +597.

446

no problem.

447

Let me modify and test for 7-9 & 7-16. After TP hit, moving SL on AIR & RETRACE to to “nearest PROFITABLE quarter” is always good.

448

wenlong, you have shown a previous interest and seemed sincere. i am sure you are just startled by that large negative number and that your comment meant no ill will toward the thread.

actually, this drawdown is not ridiculous at all and might even have been anticipated from our prior research. for your benefit and others who might also be concerned, let me explain a bit. here is an equity chart for the eur/usd for 19 months. that covers our original test period and adds july.

e-u, equity chart.jpg700×414 23.4 KB

you can see at least four months where this pair lost pips. other pairs have also had three or four losing months. now, there exists a mathematical probability that while e/u was losing, some other pairs might have been losing at the same time, epecially if we broke the data down into weeks instead of months. depending on how many pairs start losing at the same time, we may see an overall drawdown in the portfolio. that is what we are seeing now as the markets are producing very little movement in any pair and several are losing simultaneously.

is this to be expected or is this the first losing month for a method that will never perform again? who can say for sure? that’s why there is a constant disclaimer that “past performance is no indication of future results.” all we can do is keep our eye on the performance and give it up if we are sure it has quit working.

i cannot think of any method where it would be impossible to collect either historical and/or contemporary results. for crying out loud, just look at a chart and see what would have happened if you had done such and such. but, a contributor offering a system for consideration by other traders has many choices on whether to report that information and how to report it. these are some variables, some legitimate, some malicious, that factor into that reporting or lack thereof:

  1. the contributor may not know how to go about collecting performance data;
  2. the contributor may be too lazy to do the work required;
  3. the contributor may have an ego in the way and find it necessary to save face in losing periods by simply misrepresenting the results;
  4. the contributor may not offer results data because he/she knows the system is “iffy,” doesn’t know for sure whether it performed or not, probably doesn’t want to know for sure, and doesn’t want you to know;
  5. the contributor may be offering something that is just an idea, still in the development process, and is legitimately seeking input to determine whether the approach has any validity.

these, and many other determinants keep us from having valuable information that might help us decide whether we want to pursue a method or not. without implying sinister motives to any contributor, there are any number of trading approaches on this site and others that use attractive marketing verbiage, “amazing, highly accurate, profitable, no losses, 40 pips a day, sure thing,” etc. many of them offer no results information of any kind, some offer results that cannot be replicated and verified, and a few are so obscure when it comes to the actual trading rules that stephen hawking couldn’t figure out what to do.

so, which way would you prefer to have your results? how about this. how about we do our best to build at least something of a historical picture of how the method probably performed in the past. then, we start collecting and publishing contemporary results so you can see how the system is doing since it was introduced. how about we get our egos out of the way and do our best to report these results without bias or preference. the results are transparent and the collection methodology is explained. that means you can do your own research and see what you get. this way, you get to see the data, warts and all. you get to decide whether you have any interest in learning more about the method, contributing to its improvement, or using it.

if you had a crystal ball and in january of this year you had known that some trading method would have three losing months and three winning months, but at the end of that time it would be up +5,000. your crystal ball has a little glitch and can’t tell you specifically which months will win and which will lose, so you would have to trade 'em all. would you? well, that’s why we collect data and do win/loss ratios, average loss versus average win, etc. because we presume and we hope that the future may be at least something like the past and we would very much like to have three losing months, three winning months, and +5,000 pips.

a final consideration. what is a loss of -1,200? if you had started trading with this method in july, it would just be a drawdown against previously accumulated equity and you would still be positive. if you just started at the beginning of this month, yes, you are down -1,200. that is, if you had traded all five pairs. forex is the most risk flexible opportunity anyone could ever imagine. first, you don’t have to trade the entire portfolio. second, a pip or a dollar can actually be anything you want it to be, from nothing on a demo account, to one cent, one dollar, a thousand dollars. if you are trading demo, you have lost nothing but time. if you are trading at .01, you started first of the month, and you traded all five pairs, you are down about $12.00. i went live at the beginning of the month with three pairs: g/j, a/u, and e/j, so am down about -750. but, how much is that in dollars? it is whatever i felt comfortable risking at the time.

the process for any new application is overview it, study any available performance data, try to replicate any available performance data, look at specific trades to get a feel for price action within the method’s structure, determine outcomes, demo trade, trade small and increase your risk only as your account builds. this is at least a one to three month process. you can see how different an approach like this is from, “don’t ask questions, buy when the arrow’s green and sell when it’s red, make quick and easy riches.”

450

Hi pipwoof.

Thanks for this excellent strategy… I’m liking what I’ve been reading so far - will finish everything later when I get back from the office.

But I have a concern though which I hope you can answer…

If I can only afford six positions (wrt to my risk management) - which one would be the best option to take?

First:
GBP/JPY - 3 positions (take profit, air, retrace)
AUD/USD - 3 positions (take profit, air, retrace)

Second:
GBP/JPY - 2 positions (take profit, air)
AUD/USD - 2 positions (take profit, air)
EUR/USD - 2 positions (take profit, air)

Third:
Your recommendation

Thank you.

451

fitz, thanks for your interest. right now dinesh is contributing a considerable amount of time to try to get us a working ea. i just happened to be up this morning at 3:00 am with a sick dog and found him online. we were able to go back and forth with ideas and i think he is getting very close to completing a tool that will help us decide the worth of this idea and how to go about using it. i would be very cautious about going live account at this time. we hope to have more information soon, so this is a good time to look back at charts, demo trade, and see what happens with this method.

i understand what you mean by risk managment that limits the number of pairs and positions you can take. good for you for having the sense to use careful money management. i went live at the first of the month, but with very small risk. i am continuing with that, but trading only the gbp/jpy, aud/usd, and eur/jpy. preliminarily only, these have shown the most potential. personally, i would not want to eliminate position 3. i would rather trade two pairs with three positions than three pairs with two. retrace is the one that withstands the 30 pip reversals that take air out. retrace holds on to rack up the occasional big number that makes a real difference in our bottom line.

best wishes for your trading.

452

Thanks pipwoof.

I’ve finished reading everything, including the Daybreak system. And I’m going to test this on demo as soon as I can.

Actually, I’ve been trading live for around 4 months (after 3 months of trading demo) - and has been enjoying modest profitability using the “traditional methods” (if you can call it that way) of studying fundamentals and looking at price action, chart patterns, etc.

I do enjoy doing those, but indeed - that can take up a little too much of your personal time.

So now, I’m looking for a mechanical system which could perhaps free up some of the time I use for analysis. And I’m considering using half of the positions I can afford to such a system, and trade half of the positions using the “traditional methods”.

Am I correct on that assumption that doing so would eventually free up some of my time because I’ll be studying less pairs/charts?

In any case, I do have one question… which I don’t know if I just missed… but with this system, are you placing OCO entry orders on both the high and the low? Or are you just placing one entry order on the directional bias of the chart/price action?

Also, I just read Robopip’s system review of TT… I’m curious about your reaction to that… specially on the tweak of making the stops maxed at 25 pips instead of placing them at the quarter levels between 25-50 pip stops.

Thanks again.

453

fitz, i greatly respect your willingness to read through what is already a considerable amount of material for both daybreak and triple threat. though threads can get lengthy, i do believe that if someone finds a method they believe has potential, they owe it to themselves to at least leaf through the posts and read the pertinent information. thank you for doing that.

i am in the process of considering more conservative loss control. there is always a delicate balance between giving a position stop-loss room to develop and giving up too many pips initially. between my “hungary helper,” dinesh, and i, we have pretty much determined that the quarters theory should govern when we are a little further into profit, but exercise different stop-loss methods in the beginning.

i have complimented robopip on what i believe is an excellent and fair review of the system. i was very pleased to get the favorable evaluation which considers both strengths and weaknesses. since there is no grail, this must be done with any method. his reduction of the stop-loss is completely consistent with the direction we are headed. i am hoping that dinesh will have more time this week to work with me on completing an accurate ea.

at this time, both the ea built by rpotor and the one built by dinesh will trigger orders in both directions. in backtesting, we will see both long and short orders on days when both were hit and we will see results as if both trades were taken. it works a little differently in real life. in the united states, hedge positions are not permitted and neither of my brokers permit oco orders. if i have a long order open and prices reverse to the short entry point, the trade will simply not be taken. a window will pop up that says, “hedge not permitted.” i will remain in the original position, but not add the hedge position. however, if my stop-loss is hit on the long position, then prices fall further to hit my short entry, that will be taken because the long has previously been closed out.

i prefer to take only one trade per pair per day. unlike daybreak, this method has been published as one requiring attention to positions. considering our current state of development with the ea’s, the most effective way to be sure that i am taking only one trade is to manually cancel the opposite orders after entries are triggered. at this time, the ea’s simply will not do that automatically.

as far as time, this is a demanding method. manually, it would require six entry stops for each pair you are considering and steady monitoring for when moving stops is necessary. using any kind of directional bias, including entering only the side of the trade that is closest to trigger, reduces effectiveness.

the ea’s can be very helpful here, but i urge you to watch them, be sure you understand what they’re doing, and adjust them if necessary. at this time, i believe both rpotor’s and dinesh’s have the entry points and initial stops managed. once we begin a new day, attach the ea to a chart and it will automatically enter all those orders. that is, rpotor’s will actually place resting entry stops and show those in your pending orders window. dinesh’s works a little differently and will not show pending orders, but will initiate the order when conditions are met. also at this time, rpotor’s adjustment of stops for the two positions still open after tp is hit is more accurate than dinesh’s. i understand that dinesh is working on a segment of the method that describes how stops on positions 2 and 3 are treated once position 1 takes profit. that is, rather than go immediately to quarters management, which might place position 3 in a potential loss, the method describes moving those stops to “near breakeven,” i.e., a small loss or a small gain. at this time, i am recommending that we place those trailing stops at quarters that achieve a profit rather than letting a gain turn into a loss.

personally, i have comfortably placed rpotor’s ea on a chart and gone to bed or left the house. so far, that has worked out and i come back to find stops in the right places. i am using rpotor’s v3.21 for that.

it sounds like you’re the kind of trader who does the necessary homework. make that a consistent habit. contrary to much popular opinion, the first goal of trading is not to make money. the first goal of trading is to not lose money.

454

not having mt4 programming skills, i have, in the words of tennessee williams, “…always depended on the kindness of strangers.”

i have had two participants in the thread who came very close to making a working ea. rpotor did a lot of initial work and built an ea that seems to accurately enter and trail positions. denish has also contributed significantly to his ea and i think it is close to completion.

as we see the need to adjust our initial stop-loss methods, our ea would need to have some testing flexibility. i worked with denish as recently as a few days ago and hope he is still on board to move this direction. i have not heard from rpotor for a long time and fear that he may have burned out from giving too much time and attention in the beginning.

despite the fact that august’s results are currently sucking the big one, i am optimistic that robopip’s favorable review may attract some additional interest. is anyone with programming skills willing to come on board and help complete an ea? i always thought the variable moving stops would be the challenge, but those have pretty much been accomplished. it would seem that all we need now is the ability to experiment with initial stops.

thanks for your consideration.

455

pipwoof,
maybe we should calculate PF or expectancy for each TP and then decide which ones we keep? Or we can also try a fixed TP?

456

:slight_smile: I’m still here, just had a busy week and I couldn’t really check in on the forum. Just today I noticed a little problem on the EA (using the Londonbreak version). The EA was working fine until today I noticed that it just failed to update the SLs of those trades that it opened yesterday but was still running today in profit. The fixed version for the Londonbreak EA is 1.08 and it can be downloaded from here:

http://www.rpotor.com/forex/TTE/Londonbreak-TT-OrderEntry-v1.08.ex4 (compiled, “plug and play” :slight_smile: version)

http://www.rpotor.com/forex/TTE/Londonbreak-TT-OrderEntry-v1.08.mq4 (original mq4, you need LibGMT.mqh for compilation by yourself)

This bug also affects the EA which is running the original TT rules but it is very likely we haven’t run into it yet. I will not bore you with the details, but the logic of the EA is such that this bug is not likely to come up, but it is there nonetheless. Fixed hopefully :slight_smile: in the version below:

http://www.rpotor.com/forex/TTE/Daybreak-TT-OrderEntry-v3.22.mq4

Unfortunately I don’t think I can devote the necessary time in the near future to tweak the EA as thoroughly as it is requested in prior posts above, but I’ll try to keep an eye on the topic and see what dineshydv comes up with. I think his EA might be more professional and better in the end, but I like to double and triple check a code before I fire it up, especially if it involves the trading of real or even demo money, but unfortunately I have not yet seen a verifiable mq4 version of dineshydv’s EA here just a compiled ex4 version. Anyway, if I find the time I think I’ll try to tweak my EA as well, since I can learn good things about EA programming from this exercise, but yeah, my pace might slow down a bit. :slight_smile:

457

Hopefully this was “fixed” in the meantime as you’ve read through the topic, but still: woris, I’m sorry for your bad experience, 1.01 was the first prototype so to speak and as the topic developed, so did the EA. :slight_smile: See my post above this one for the current versions, which are working fine (or they should until we find other bugs :slight_smile: I’ll try to steal some time somewhere to set up a very basic “homepage” for the EA, so I will not pollute the forum with links to each and every version, just point interested people to a link where the current versions could be downloaded from.

458

I think this is basic Forex market behavior, as you’ve correctly identified. Both of those orders were sell orders and sell orders are taken out when the Ask price reaches the price level set as the SL. So, while it’s true, that for example the Bid was 1.5697, however with a spread of 3 pips, that would mean the Ask price at that moment was 1.5700 and your broker correctly closed the trade. I dunno how it would affect our calculations in the EA if we would add the spread to the stop losses, I have to think about this and “demo it” in my head. :slight_smile:

459

[QUOTE=rpotor; I’ll try to steal some time somewhere to set up a very basic “homepage” for the EA, so I will not pollute the forum with links to each and every version, just point interested people to a link where the current versions could be downloaded from.[/QUOTE]

It is great see you back! your suggestion will be nice and appreciated by all, and hope Pipwoof accepts this?

460

Rpotor, which of the ea version should we use, 1 or 2 as below?

1.http://www.rpotor.com/forex/TTE/Lond...ntry-v1.08.ex4 (compiled, “plug and play” :slight_smile: version)

  1. http://www.rpotor.com/forex/TTE/Lond...ntry-v1.08.mq4 (original mq4, you need LibGMT.mqh for compilation by yourself)

  2. http://www.rpotor.com/forex/TTE/Dayb...ntry-v3.22.mq4

461

rpotor, very good to see you back. at this time, i must say that your version 3.21 is operating more according to the original tt than denish. i will upgrade to the new version. you may be right, in the end his version may be more “professional,” but, as a beginner with mt4 i must commend your work and i am, frankly, more interested in functionality than professionalism.

denish has done a lot of good work and i enjoyed being on line with him the other morning, going back and forth with ideas. however, he will have to resolve the issue of moving to near b/e on pos 2 and 3 after tp, which your program already seems to do effectively. i wish i could appear less selfish here, but you did hint that you might find time to tweak your ea and i hope you do.

i believe the only things missing from your version would be limiting to one trade per day and allowing a fixed initial stop loss rather than using the original quarters idea. it now seems that the big losses occurring when the quarter is too far away increase the risk more than necessary and we can control better with straight fixed stop losses.

i’ve thought about adding the spread to stop points, but believe it will be more work than it is worth. after all, a stop at 75 or 73 probably won’t make that much difference in the long run. i will certainly understand if other priorities keep you from adding the final tweaks. i do realize how much time and mental focus you have put in already and it is appreciated.

i wasn’t entirely sure whether your objective was to learn more about mt4 programming, consider trading the method, or both. whatever, i hope you are closer.

462

though not good, i have not backed off from posting august results. just lost a computer to the “fbi moneypak” virus and geeksquad is several days out. i can communicate with this one, but the other was my main trade computer and had all my charts, indis, etc. i have since bought a 1 tb backup hard drive and will be using it in the future. always living and learning, even at my age. when i get it back, hopefully complete with all those red and yellow horizontal lines that took hours to place, i will update results.

if you retain any interest in the method, take time to read robopip’s evaluation. it was more than fair and, this month’s performance notwithstanding, encourages me.