Random entries - Let's take it to the extreme | Forex Factory

Options

Go to
Permalinks
Print Thread

Similar Threads

Is random really random? 231 replies

Coin tossing and random entries 44 replies

Pure Risk to reward trade System - 1 to 3- Totally random entries 178 replies

Expert Advisor with random entries 3 replies

Random entries discussion... 7 replies

Attachments: Random entries - Let's take it to the extreme
Exit Attachments

Random entries - Let's take it to the extreme

  • #1
  • First Post: Nov 27, 2019 2:22am Nov 27, 2019 2:22am
Hi all,

Before you go on, let's keep an open mind.

Here's a thought experiment. We like to talk about how edges decay over time (some swear by an average of 3 years before the efficacy of a strategy falls).

What if all trades were entered at random and all you could control is the exit strategy? A system based in random entries has to no edge to speak of, so by that logic, there is no edge to decay.

But how would you trade it?

I've been dabbling with this wild thought for a while now and wanted to see if there is a like-minded individual who wants to discuss this audacious experiment.

If there isn't enough interest in this area, I shall take my work elsewhere.

All ideas are welcome.
  • #2
  • Nov 27, 2019 2:39am Nov 27, 2019 2:39am
Hi carddard i also was thinking about this,here is EA based on Random Entry Points with SL and TP

Note:about Exit Strategy for Random i can propose use High and Low from H1 TimeFrame

when price will break 4 high we must exit for buy orders

when price will break 4 low we must exit for sell orders

Thanks
Attached Image (click to enlarge)
Click to Enlarge Name: 1.png Size: 378 KB
Attached File(s)
File Type: ex4 random.ex4   9 KB | 588 downloads
File Type: mq4 random.mq4   3 KB | 642 downloads
  • #3
  • Nov 27, 2019 3:05am Nov 27, 2019 3:05am
Quoting inner21
Hi carddard i also was thinking about this,here is EA based on Random Entry Points with SL and TP Note:about Exit Strategy for Random i can propose use High and Low from H1 TimeFrame when price will break 4 high we must exit for buy orders when price will break 4 low we must exit for sell orders Thanks {file} {file} {image}
Hi inner21,

Your random entries don't seem to be random. They are entered at specific points.
  • #4
  • Nov 27, 2019 3:06am Nov 27, 2019 3:06am
you'd probably won't go far wrong by flipping a coin every start of new bar
heads buy tails sell, 0.5% risk per trade
set stop to the high or low of previous bar accordingly
trail the stop of any trade that survives to SMA 20 or any Trailing Stop you like
In the end its just a stupid numbers game
  • #5
  • Nov 27, 2019 1:03pm Nov 27, 2019 1:03pm
Random entry = Random results. You might still be profitable after cost but by a fine margin. Live trading will be incredibly hard. It's hard to keep believing in your strategy when you've been in a 3 year drawdown.

I Dont agree that strategies stop working after 3 years. That simply means your strategy was not robust. You either curve fitted or market fitted.

The million dollar question is how to find a robust strategy? The starting point to answer this is to limit your involvement in the market. The less active you are in the market the better...
When you see it, BET big.
  • #6
  • Nov 28, 2019 4:46am Nov 28, 2019 4:46am
Doesn't seem like there is any creative approach to this. Maybe I can kickstart the discussion with a suggestion. This is what I have been testing - Upon x period of internal, open a random buy or sell trade. Attempt to close this and continue to open and close a series of randomly entered trades. The idea is to focus solely on the exit method.
  • #7
  • Nov 28, 2019 5:53am Nov 28, 2019 5:53am
A quick backtest on some of the possibilities (2019 Jan to 2019 May):
Attached Image (click to enlarge)
Click to Enlarge Name: 2.JPG Size: 283 KB
  • #8
  • Nov 28, 2019 6:43am Nov 28, 2019 6:43am
Quoting carddard
Doesn't seem like there is any creative approach to this. Maybe I can kickstart the discussion with a suggestion. This is what I have been testing - Upon x period of internal, open a random buy or sell trade. Attempt to close this and continue to open and close a series of randomly entered trades. The idea is to focus solely on the exit method.
Banks already effectively do this when brokering options, via delta hedging. They don't get to choose the entry, but the exit is predetermined. Maybe worth some of your time.
  • #9
  • Nov 28, 2019 4:19pm Nov 28, 2019 4:19pm
If the most "optimized" A.I.s and algos suck, do you really think this whole random thing will work?

This is silly.
  • #10
  • Nov 28, 2019 4:37pm Nov 28, 2019 4:37pm
Quoting The-Flipper
If the most "optimized" A.I.s and algos suck, do you really think this whole random thing will work? This is silly.
Curve fitting is real. The most optimized algos are curve fitted and would not survive live trading. Read your statement and ask yourself if you have exhausted all "good strategies" and still can't find success. You either go with the crowd or think outside of the box.

I hope you find success my friend.
  • #11
  • Nov 28, 2019 4:44pm Nov 28, 2019 4:44pm
I have begun a series of experiments with random entries, and am tracking them via trade explorers / fxblue to test the efficacy of this concept.
For those who are interested and have something meaningful to share, please feel free to do so via this thread, or contact me directly.

Like many others on FF, I don't seem to be able to get meaningful inputs from others via a thread. Maybe the remaining good guys on this forum are not sharing their work. Which is probably the best way to approach this to protect their IP. Perhaps I should too, but I'll give this one last chance.

Trade well guys!
  • #12
  • Nov 28, 2019 8:01pm Nov 28, 2019 8:01pm
Quoting carddard
I have begun a series of experiments with random entries, and am tracking them via trade explorers / fxblue to test the efficacy of this concept. For those who are interested and have something meaningful to share, please feel free to do so via this thread, or contact me directly. Like many others on FF, I don't seem to be able to get meaningful inputs from others via a thread. Maybe the remaining good guys on this forum are not sharing their work. Which is probably the best way to approach this to protect their IP. Perhaps I should too, but I'll...
You 've received a couple of good answers, you might just not be willing to listen. It works but after commissions, swap and slippage it barely stands above the water level. The pain to gain ratio is very High, not only deep drawdowns but also long drawdowns. Still better than what most do with their accounts if you can stick with it.
When you see it, BET big.
  • #13
  • Nov 28, 2019 9:00pm Nov 28, 2019 9:00pm
Quoting Rennaissance
{quote} You 've received a couple of good answers, you might just not be willing to listen. It works but after commissions, swap and slippage it barely stands above the water level. The pain to gain ratio is very High, not only deep drawdowns but also long drawdowns. Still better than what most do with their accounts if you can stick with it.
I hear you, and you have further confirmed the fact that the quality of discussion may not be what I am expecting.
No worries, I'll leave this thread to run its natural course, and continue the work with my team.

All the best guys, and happy holidays!
  • #14
  • Edited 8:13am Nov 29, 2019 4:49am | Edited 8:13am
Quoting carddard
I have begun a series of experiments with random entries, and am tracking them via trade explorers / fxblue to test the efficacy of this concept. For those who are interested and have something meaningful to share, please feel free to do so via this thread, or contact me directly. Like many others on FF, I don't seem to be able to get meaningful inputs from others via a thread. Maybe the remaining good guys on this forum are not sharing their work. Which is probably the best way to approach this to protect their IP. Perhaps I should too, but I'll...
Cardard,

You might want to read through 60minuteman's fascinating thread "Systematic Trend Trading". Both 60minuteman (Darren Littlewood) and FF member Copernicus discuss the aspect of random (i.e. trend) trading through the lens that the elusive trading edge lies in trades that have an asymmetrical risk:reward achieved by the use of small, hard SL to limit losses and trailing stops to capture the rare right-tailed large gains IF/WHEN price breaks out in the direction of their trade. Hence, both point out that long-term trading success (i.e. edge) comes from focusing on exit strategies more than entry strategies in order to capture these rare large asymmetrical gains. Darren BRIEFLY discusses a few favorite PA-based entry patterns, but quickly acknowledges that his entry patterns really don't matter - its just a reason for him to enter trades and it's the disciplined adherence to his exit strategy that results in his long-term success. Copernicus discusses the use of "filters" to latch onto the asymmetrical trades that lead to the rare long-term gains, and that the filters limit the depletion of the account due to the constant whipsaws that is the bane of random (i.e. trend) trading - so his entry approach is less "random", but the overall edge is still due to limiting losses and letting profits run. Anyways, a very very good read....

In my own experience, my back-testing of random entry methods invariably had negative expectancy ...I focused on varying the risk:reward (Hard SL/hard TP) or having NO SL/hard TP of various amounts, or small SL/trailing TP), and grids of various types, in order to obtain an edge...suffice to say...

Anyways, I do hope you keep the discussion going...

Best,

Dave
  • #15
  • Nov 29, 2019 5:48am Nov 29, 2019 5:48am
Quoting davelansing
{quote} Cardard, You might want to read through 60minuteman's fascinating thread "Systematic Trend Trading". Both 60minuteman (Darren Littlewood) and FF member Copernicus discuss the aspect of random (i.e. trend) trading through the lens that the elusive trading edge lies in trades that have an asymmetrical risk:reward achieved by the use of small, hard SL to limit losses and trailing stops to capture the rare right-tailed large gains IF/WHEN price breaks out in the direction of their trade. Hence, both point out that long-term trading success (i.e....
I have followed C's journey before he became a commercial member, and dabbled with some of the tools he used, but couldn't find success. Since then, I have been experimenting with new parameters by isolating the exit condition.
  • #16
  • Nov 29, 2019 5:53am Nov 29, 2019 5:53am
Quoting carddard
{quote} I have followed C's journey before he became a commercial member, and dabbled with some of the tools he used, but couldn't find success. Since then, I have been experimenting with new parameters by isolating the exit condition.
Interesting... what exit conditions? Related to risk/reward or consistent PA/TA pattern? Please feel free to PM me if you don't wish to discuss on the open forum...
  • #17
  • Dec 1, 2019 11:37am Dec 1, 2019 11:37am
Quoting davelansing
{quote} Cardard, You might want to read through 60minuteman's fascinating thread "Systematic Trend Trading". Both 60minuteman (Darren Littlewood) and FF member Copernicus discuss the aspect of random (i.e. trend) trading through the lens that the elusive trading edge lies in trades that have an asymmetrical risk:reward achieved by the use of small, hard SL to limit losses and trailing stops to capture the rare right-tailed large gains IF/WHEN price breaks out in the direction of their trade. Hence, both point out that long-term trading success (i.e....
Long term Success comes from both entries and exits. You need to be careful about how and where you play. If not you arrive at a random walk with a little profit to show for the magnitude of your efforts.
When you see it, BET big.
  • #18
  • Dec 1, 2019 4:50pm Dec 1, 2019 4:50pm
Quoting Rennaissance
{quote} Long term Success comes from both entries and exits. You need to be careful about how and where you play. If not you arrive at a random walk with a little profit to show for the magnitude of your efforts.
I don't disagree. There are many ways to skin a cat.
What I'm experimenting on is completely disregarding the entry criteria to see if any positive expectancy is derived at all.
  • #19
  • Dec 1, 2019 4:51pm Dec 1, 2019 4:51pm
Quoting davelansing
{quote} Interesting... what exit conditions? Related to risk/reward or consistent PA/TA pattern? Please feel free to PM me if you don't wish to discuss on the open forum...
None of those. Happy to discuss in private but it has to be a 2-way conversation.
  • #20
  • Edited Dec 2, 2019 2:38am Dec 1, 2019 5:41pm | Edited Dec 2, 2019 2:38am
Quoting Rennaissance
{quote} Long term Success comes from both entries and exits. You need to be careful about how and where you play. If not you arrive at a random walk with a little profit to show for the magnitude of your efforts.
Rennaissance,

That's the conclusion that I've come to as well by way of extensive backtesting (see my previous post on backtesting random entries with specific exit strategies).... but the arguments - and copious amounts of DATA - presented by 60minuteman (Thread: "Systematic Trend Trading", and Copernicus (Threads: "Diversified Trend Trading Approach", and "Systematic Trend Trading"), and many others, also supports the idea that asymmetrical risk:reward trading (i.e. trend trading) has long-term profitability- especially when compounded. With that said though [for profitable trend trading], you also have to accept the long periods of draw down and little to no gain (or loss) on a month-to-month basis with the HOPE that you never miss the handful of trades that lead to those extraordinary large asymmetrical wins. Thankfully there are many ways to profitable trading, but with the caveat that a method that works extremely well for one trader can blow up another trader's account...

Best,

Dave
0 traders viewing now