Trapping System: Tunnel Martingale | Page 23 | Forex Factory

Options

Go to
Permalinks
Print Thread

Similar Threads

Martingale, Anti-martingale, and Compounding 40 replies

Martingale, Reverse Martingale, Modified Martingale, Maths 1 reply

How to change this Martingale to Reverse Martingale ? 3 replies

Trapping System: Way to Fully-Automated Trading System 47 replies

Martingale vs. Non Martingale (Simplified RoR vs Profit) 1 reply

Attachments: Trapping System: Tunnel Martingale
Exit Attachments

Trapping System: Tunnel Martingale

  • #441
  • May 28, 2020 6:44am May 28, 2020 6:44am
Now, let's talk about Smart Bucket. This is getting interesting
The idea is to compound in the piggy bank, the losses you suffer for the past trades, and then add it up to the next trade's target profit. This means if the previous/current trade is in the loss for $5 and closed due to a condition in the configuration, the losses will be added into the bucket. More loss for another $5, add again into the bucket which in total would be $10. If your target profit is $20, then for the next session/cycle, your target profit would be $30 ($20 from target profit, and $10 from the bucket).
Practically saying, your previous losing trades will be "zero"-ed once the target is reached.

There are 2 types of Smart Bucket: Normal Smart Bucket, and Martingale Smart Bucket (this one is experimental).
So, these Smart Bucket essentially is only for mitigating the loss suffered. In a quick short word, take the loss now, convert into no loss on the next trade. If they still got the loss, accumulate again the loss and convert into no loss on the next trade, so on so forth.
Attached Image (click to enlarge)
Click to Enlarge Name: TM-SmartBucket.png Size: 51 KB

Imagine the picture above (no 5). If the price moves and executes both 1st and 2nd order into positions, when the price move past the center (A), it will close the 0.02 position with loss and put the loss amount (in non-negative) into the bucket. And this bucket will be considered into the profit taking too (as described above).
If price goes up till (B), it will not close the cycle and continue. Because essentially, it's not reached the needed take profit as there's addition from the bucket. But say if price goes to (C) with considering the bucket, it will close the cycle. At this point, the previous loss trade "converted" into no-loss or $0 trade.
If you ask me to code/fix your EA... it's probably not for free...
  • #442
  • May 28, 2020 7:15am May 28, 2020 7:15am
Quoting radityo.ardi
OK, just to settle in on what's going on with Tunnel Martingale from the last update. For those who don't follow, let's get a refresher. Tunnel Martingale basically consist of 2 main things: Ceiling, Floor. Both are the price line which guards if the price goes towards past Ceiling, it will put pending order on the Floor 2 times lot size of the previous (Ceiling). If this pending order gets executed, it will then again add pending order on the Ceiling 2 times lot size if the Floor. Until then it breaks out, (in example 0.16 of the lot size), it...
I looked into this concept some years ago and never figured out how to reduce the risk.
Maybe you will solve this issue
  • #443
  • Jun 1, 2020 2:58am Jun 1, 2020 2:58am
Quoting Merka
{quote} I looked into this concept some years ago and never figured out how to reduce the risk. Maybe you will solve this issue
This is a treacherous path to go down indeed. It's something that I have spent a lot of time on (out of curiosity) on demo but never brought live, because of the fact that investors do not see this as viable investing. But or the hobby trader / gambler who has a higher risk profile, it is definitely fun to try out.

There are ways to reduce your risk by cutting down on the number of open trades - but at the same time that exposes you to the current open leg (either buy or sell) and you need to constantly hedge the open order in hopes that there is enough volatility to bring you back on track.

It also requires a certain amount of equity to trade more than a pair at any time in time. Another reason why I refused to bring this to live.

But as a mental exercise, it is very fun to play around with. If anyone has enough guts to try it out on live, I'm willing to share!
  • #444
  • Jun 2, 2020 12:37am Jun 2, 2020 12:37am
Quoting carddard
{quote} This is a treacherous path to go down indeed. It's something that I have spent a lot of time on (out of curiosity) on demo but never brought live, because of the fact that investors do not see this as viable investing. But or the hobby trader / gambler who has a higher risk profile, it is definitely fun to try out. There are ways to reduce your risk by cutting down on the number of open trades - but at the same time that exposes you to the current open leg (either buy or sell) and you need to constantly hedge the open order in hopes that...
I do agree martingale has its risk. But isn't it the same concept with other trading systems as well?

Just think of this, someone posted what claimed to be a winning strategy, then another too with another strategy, but all of these I found 1 thing in common. They have entry requirements. If not followed, then the strategy won't work. The funny part is that most of us ignore the fact that martingale does require entry requirements. Am I right?

When I walk past MRT or public transport and sees the advertisement about investing in this, that, whatever, those always with one distinctive fine print: "Investments are subject to investment risks. The risk of loss in leveraged trading can be substantial. You may sustain losses in excess of your initial funds." or anything along that line.

Another example, when you trade EURUSD with technicals, say you use 2 EMA and 1 BB. Your signal tells you to buy, but then you would have another requirement that this strategy only works perfectly when you trade during EURUSD trading window. Other than that, it won't work nicely. Isn't it the same like when you apply martingale, but only at the time when a news has been released? Isn't it the same?
Of course we could say, it's not the same. But what are the other similarities between those? What about something like can I set up the entry price for martingale the top and bottom? Of course if you set the top and bottom too tight, you would end up in bleeding your own account. But what if we can find the best distance between the top and bottom to work this out? Isn't this the same as we apply technicals where the EMA period parameters one is 14 and another is 50?
I'm just saying it's not fair for people to say martingale is gambling and others are not.

Then some other time, someone experienced (just an example to show you the idea, which I've also encountered in real life in the past) "okay, I'm using 2 EMA and 1 BB. Year 2015 until last year it was working perfectly. This year, it's in bad shape. this strategy doesn't work throughout the time". Well, then we all are in fact gambling our money out. We all do if you see the bold line which the advertisement forex company put on.


We need to realize the fact that every single strategy (fundamentals, technicals, including martingale since it's considered algorithmic) in this world, pose a risk and each of them has their own requirements in order to make profits. These are facts we should not ignore.

Now the question is, how do we reduce risks in martingale?
If you ask me to code/fix your EA... it's probably not for free...
  • #445
  • Jun 2, 2020 12:48am Jun 2, 2020 12:48am
I like to learn. But too busy trading. Will comment later. Thanks and
Regards
  • #446
  • Jun 2, 2020 1:07am Jun 2, 2020 1:07am
Quoting radityo.ardi
{quote} I do agree martingale has its risk. But isn't it the same concept with other trading systems as well? Just think of this, someone posted what claimed to be a winning strategy, then another too with another strategy, but all of these I found 1 thing in common. They have entry requirements. If not followed, then the strategy won't work. The funny part is that most of us ignore the fact that martingale does require entry requirements. Am I right? When I walk past MRT or public transport and sees the advertisement about investing in this, that,...
Don't get me wrong radityo.ardi, I completely agree that all strategies have inherent risk. No doubt the entry / exit requirements matter like in any strategy. I'm not challenging that. It's a personal preference of how much risk you want to undertake and in what form. My problem with this is that the absolute risk profile is not the only constraint that I am working within. The amount of capital and margin requirements matter as well.

What I mean by this is, if I want to put my money between these 2 options:

1) Open 1 trade of 0.01 lot each for 28 pairs
2) (1) but with multiple entries with varying lots on 28 pairs

I would almost always go for (1) because firstly I spread my exposure around different pairs and I know that in the worst case scenario, I always have only 1 trade (or 2, depending on whether you buy / sell at the same time). From an outsider point of view, it is also easier to attract investors because there is no overexposure in any particular instrument. That is also why I said that however, this is an interesting experiment to try out regardless of how attractive it is to investors, because it's an intellectually stimulating exercise and nothing is more exciting than an unorthodox method that proves the critics wrong.

I have spent countless hours on trying to crack the martingale puzzle and quite literally have hundreds of algorithms just based on hedging / martingale but have not used any of them live because of the reasons stated above.

With that said, I would still love to see how this progresses because I like to keep an open mind when it comes to retail trading. Never say never!
  • #447
  • Jun 2, 2020 1:08am Jun 2, 2020 1:08am
Quoting Merka
{quote} I looked into this concept some years ago and never figured out how to reduce the risk. Maybe you will solve this issue
because all of us think that martingale will solve all problems without us thinking through what's the entry requirement. It's a wrong mindset (me too in the past). So, we tend to think martingale has to be run all the time ignoring all the hurdles and quiet hours. We tend to think to take profit small, then do another round again next time.

1 thing for sure, every news is released, the price always go to somewhere new and sit there for some time till another news is released. Sometimes mixed up, goes up there, then go down, but the range is not that something like sideways. Sometimes this can go like 100 pips - 400 pips. One day this year EURUSD reaches 1400 pips a day!
Imagine that if you put martingale at the right time let it move a bit (to execute another few pending orders) then shoot that 1400 pips without you closing the order. How much profit you will get? You can stop your martingale after that day and go to Bali.

Martingale doubles your risk when it's not on your side, but once it shoots up and you let it go, your profit potential also doubles.
If you ask me to code/fix your EA... it's probably not for free...
  • #448
  • Jun 2, 2020 1:15am Jun 2, 2020 1:15am
Quoting goodways100
I like to learn. But too busy trading. Will comment later. Thanks and Regards

Quoting carddard
{quote} Don't get me wrong radityo.ardi, I completely agree that all strategies have inherent risk. No doubt the entry / exit requirements matter like in any strategy. I'm not challenging that. It's a personal preference of how much risk you want to undertake and in what form. My problem with this is that the absolute risk profile is not the only constraint that I am working within. The amount of capital and margin requirements matter as well. What I mean by this is, if I want to put my money between these 2 options: 1) Open 1 trade of 0.01 lot...
Yes, man. absolutely...

Anyway for both, I am doing this since no one is exploring much. So far the result is somewhat good, but I need more than what I'm seeing it currently. Gotta backtest it for at least the span of a year.
If you ask me to code/fix your EA... it's probably not for free...
  • #449
  • Jun 2, 2020 1:18am Jun 2, 2020 1:18am
Quoting radityo.ardi
{quote} {quote} Yes, man. absolutely... Anyway for both, I am doing this since no one is exploring much. So far the result is somewhat good, but I need more than what I'm seeing it currently. Gotta backtest it for at least the span of a year.
I have some old ideas that I have tested and abandoned, let me see if I can dig them out - Who knows, it might come in handy, but happy to contribute in the discussion!
  • #450
  • Jun 2, 2020 1:53am Jun 2, 2020 1:53am
Quoting radityo.ardi
OK, just to settle in on what's going on with Tunnel Martingale from the last update. For those who don't follow, let's get a refresher. Tunnel Martingale basically consist of 2 main things: Ceiling, Floor. Both are the price line which guards if the price goes towards past Ceiling, it will put pending order on the Floor 2 times lot size of the previous (Ceiling). If this pending order gets executed, it will then again add pending order on the Ceiling 2 times lot size if the Floor. Until then it breaks out, (in example 0.16 of the lot size), it...
One of the ideas that I played around with in the past is to adopt a dynamic "tunnel" - The ceiling and the floor is not fixed to any single point, but adjusts according to price momentum. The shortfall of this is that the positions have to be managed at an equilibrium at all times - Meaning to say the orders are placed according to where the current price momentum is going towards, and we don't try to predict that direction because we can't. But that means that I had to find ways to close out these positions actively to avoid the accumulation of a series of too many unfavourable trades.

This is an example of a build up of too many trades while holding on stubbornly until a profit target is reached.
It's all fun and games in demo but in live it's something to shit your pants over.
Attached Image (click to enlarge)
Click to Enlarge Name: 1.JPG Size: 241 KB
  • #451
  • Jun 2, 2020 2:05am Jun 2, 2020 2:05am
Quoting carddard
{quote} One of the ideas that I played around with in the past is to adopt a dynamic "tunnel" - The ceiling and the floor is not fixed to any single point, but adjusts according to price momentum. The shortfall of this is that the positions have to be managed at an equilibrium at all times - Meaning to say the orders are placed according to where the current price momentum is going towards, and we don't try to predict that direction because we can't. But that means that I had to find ways to close out these positions actively to avoid the accumulation...
Does that execute all the time, back-to-back one to another?
I gotta know, to confirm that my findings are correct.

Quoting carddard
until a profit target is reached.
That is the sentence I've always had in mind, but actually you can "release" that and let it shoot all the way. Just like trailing stop does.
If you ask me to code/fix your EA... it's probably not for free...
  • #452
  • Jun 2, 2020 2:10am Jun 2, 2020 2:10am
Quoting radityo.ardi
{quote} Does that execute all the time, back-to-back one to another? I gotta know, to confirm that my findings are correct. {quote} That is the sentence I've always had in mind, but actually you can "release" that and let it shoot all the way. Just like trailing stop does.
It may or may not, depending on the settings. If you want to tap on the higher volatility sessions, you can do that. In the screenshot, I restricted hours to the London / NY session. But to be fair, trades are left open until the next trading day.

Reg. profit target - Yes agreed. We should use a trailing stop that would come in handy if prices continue to move favorably away from your reference point.
  • #453
  • Jun 2, 2020 2:17am Jun 2, 2020 2:17am
Quoting radityo.ardi
{quote} That is the sentence I've always had in mind, but actually you can "release" that and let it shoot all the way. Just like trailing stop does. {quote}
By the way - With both buy and sell trades open at the same time, how is your approach in determining the "average price" for the sum of these orders, so that you can establish a reference point for the trailing stop. With just buy orders or just sell orders it's straightforward. But with both legs going on at the same time, do you have any ideas?
  • #454
  • Jun 2, 2020 2:25am Jun 2, 2020 2:25am
Quoting carddard
{quote} It may or may not, depending on the settings. If you want to tap on the higher volatility sessions, you can do that. In the screenshot, I restricted hours to the London / NY session. But to be fair, trades are left open until the next trading day. Reg. profit target - Yes agreed. We should use a trailing stop that would come in handy if prices continue to move favorably away from your reference point.
ok noted.
In my mind, martingale can't be run back-to-back, even if you limit the trading hours. IMO must be one trade at a time, when something we know potentially could yield large pips. Other than that, sideways of different sizes may occur and potentially lead to sucking up the risk we have.
If you ask me to code/fix your EA... it's probably not for free...
  • #455
  • Jun 2, 2020 2:35am Jun 2, 2020 2:35am
Quoting carddard
{quote} By the way - With both buy and sell trades open at the same time, how is your approach in determining the "average price" for the sum of these orders, so that you can establish a reference point for the trailing stop. With just buy orders or just sell orders it's straightforward. But with both legs going on at the same time, do you have any ideas?
Trailing can't be managed by normal order as usual. It has to be managed inside the EA itself but still uses the same concept.
Usually, trailing stop is attached to an order with some number of pips to be the "distance".

Now, the trailing stop must be attached to the whole chain of orders by trailing NOT the pips, but the profit $. This must be managed inside the EA.


Apart from that, below is just a backtest about CPI m/m and Core CPI m/m from Jan 2019 till May 2020. This is what I'm talking about...
The summary of this specific martingale strategy is only 1 chain of trades at a time, at one specific time, then force it close either on target or after some duration.
Attached Image (click to enlarge)
Click to Enlarge Name: Screenshot1.png Size: 199 KB

I'm about to test the other type of news.

Just don't look at the capital, $, profit, etc. Focus on the strategy requirements.
If you can see on the equity chart above, most of the trades closes with 1 open position only. Except the one I put mark, apparently touches the ceiling, then the floor. But that's it.
If you ask me to code/fix your EA... it's probably not for free...
  • #456
  • Jun 2, 2020 2:50am Jun 2, 2020 2:50am
Quoting radityo.ardi
{quote} Trailing can't be managed by normal order as usual. It has to be managed inside the EA itself but still uses the same concept. Usually, trailing stop is attached to an order with some number of pips to be the "distance". Now, the trailing stop must be attached to the whole chain of orders by trailing NOT the pips, but the profit $. This must be managed inside the EA. Apart from that, below is just a backtest about CPI m/m and Core CPI m/m from Jan 2019 till May 2020. This is what I'm talking about... The summary of this specific martingale...
Yes, I understand your point about the trailing stop. I guess using $ value increments is 1 way to trail the entire group. I can visualize how using a pip step (with a directional bias) could work as well, as long as the net profit continues to ascend according to the pip step. I'll find time to test it out in code.

Noted on your approach - It's very similar to one of the ways I've tested in the past - With the ideal scenario being just closing with 1 open position with a regular trailing stop. Only when price moves unfavorably do I kick in a "tunnel martingale".
  • #457
  • Jun 2, 2020 6:17am Jun 2, 2020 6:17am
Quoting carddard
{quote} One of the ideas that I played around with in the past is to adopt a dynamic "tunnel" - The ceiling and the floor is not fixed to any single point, but adjusts according to price momentum. The shortfall of this is that the positions have to be managed at an equilibrium at all times - Meaning to say the orders are placed according to where the current price momentum is going towards, and we don't try to predict that direction because we can't. But that means that I had to find ways to close out these positions actively to avoid the accumulation...
As opposed to a fixed $ closure, I've added a group trailing stop and TP is relying solely on that.
The equity profile looks something like this. You will notice the steeper incline during each group close.
I think another viable way to do this is using the most recent swing high / low, instead of a fixed pip step.
Attached Image (click to enlarge)
Click to Enlarge Name: 1.JPG Size: 257 KB
  • #458
  • Jun 2, 2020 11:28pm Jun 2, 2020 11:28pm
Quoting carddard
{quote} As opposed to a fixed $ closure, I've added a group trailing stop and TP is relying solely on that. The equity profile looks something like this. You will notice the steeper incline during each group close. I think another viable way to do this is using the most recent swing high / low, instead of a fixed pip step. {image}
Did you change any of the test parameters? I hope not.
last time I see equity reaches down towards 2K which is not good. Now the risk is somewhat acceptable from my point of view. Congrats!
If you ask me to code/fix your EA... it's probably not for free...
  • #459
  • Jun 3, 2020 12:51am Jun 3, 2020 12:51am
Quoting radityo.ardi
{quote} Did you change any of the test parameters? I hope not. last time I see equity reaches down towards 2K which is not good. Now the risk is somewhat acceptable from my point of view. Congrats!
Nope, didn't change the parameters. Yeah I guess it helps to trail profit but the pre-requisite of having enough volatility and movement is important. Otherwise it would continue to bounce off the same region for a while and accumulate open lots.
  • #460
  • Feb 28, 2021 12:14pm Feb 28, 2021 12:14pm
Quote
As opposed to a fixed $ closure, I've added a group trailing stop and TP is relying solely on that.
The equity profile looks something like this. You will notice the steeper incline during each group close.
I think another viable way to do this is using the most recent swing high / low, instead of a fixed pip step.

carddard, could you send the parameters you are using to get to the BT shown above?
Anyone with good and consistent results?

Tks
0 traders viewing now