(binned per thread starter’s request) MM (Money Maker) Detective | Page 224 | Forex Factory

Options

Go to
Permalinks
Print Thread

Similar Threads

Money management: Capital needed per 0.07 USD per microlot 0 replies

The Volume Detective 411 replies

Attachments: (binned per thread starter’s request) MM (Money Maker) Detective
Exit Attachments

(binned per thread starter’s request) MM (Money Maker) Detective

  • #4,461
  • Jul 6, 2019 9:20pm Jul 6, 2019 9:20pm
Quoting mixedbags
I've been trying to build an EA but even with this simple web-based EA builder I can only manage to create some simple EA. Expert Advisor Builder (https://fxdreema.com/) This is my weekend hobby while sitting at the guard room in this deserted industry building. Good things is that it is free as long as I keep to the weekly quota.
@mixedbags -- this is great -- you have to start somewhere!! Slowly, slowly you'll improve your skills.

If you keep it up then next year at this time people will have forgotten about me and instead be looking to you and @bishopdotun for the latest and greatest indicators and EAs.
  • #4,462
  • Edited 9:40pm Jul 6, 2019 9:22pm | Edited 9:40pm
Quoting robots4me
Everyone should read this post by @dkrock. He has a tendency to delete them, so I wouldn't delay: https://www.forexfactory.com/showthread.php?p=12369150#post12369150
I trade similar with dkrock, ie. with math that calculates the dynamic pricing of the underlying asset. I trust math and nothing else.

His posts always touch on a few same aspects of trading. This one opinion may ruffle some feathers. But it's important traders think deeply about this subject - support and resistance based on historical high/low mostly projected by horizontal and angular lines. If it works for you that gives edge to your trading, fine.

This popular method of horizontal and angular lines mathematically suggests uniformity and assumes a linear time base rate of change. Take note of that.

Probably you are trading support/resistance too, so you are trading fake data. That would also mean you are not able to compare time frames. Support/resistance does not have any probability calculations in it. It is simply trading the last trade all over again. Since the new trade has new traders with new lots sizes and new reasons for trading, it makes no sense to trade the last trade again. S/R trading is highly touted by banks to convince people to risk money at prescribed trading levels so they know where the money is and can steal as much as possible. Otherwise, s/r trading is meaningless. It simply exists as a place to go and get mugged. ....... Dkrock
https://www.forexfactory.com/showthread.php?p=12238533#post12238533
  • #4,463
  • Jul 6, 2019 9:27pm Jul 6, 2019 9:27pm
Quoting bishopdotun
{quote} Great. I will add option 3 3. Close above midline (Doesn’t have to necessarily touch the opposite band)
@bishopdotun -- just to be clear about the rule.

We don't know a Close until we get the first tick of the next bar -- i.e. the Open price. So, the way I'm implementing this is as follows:
1. Check if previous bar crossed the midline
2. Check if current bar's Open is above / below the midline -- if so, then mark as an Exit
  • #4,464
  • Jul 6, 2019 9:33pm Jul 6, 2019 9:33pm
Quoting skyway
{quote} I trade similar with dkrock, ie. with math that calculates the dynamic pricing of the underlying asset. I trust math and nothing else. His posts always touch on a few same aspects of trading. This one opinion may ruffle some feathers. But it's important traders think deeply about this subject - support and resistance based on historical high/low mostly projected by horizontal and angular lines. If it works for you that gives edge to your trading, fine. This popular method of horizontal and angular lines mathematically suggest uniformity...
Quote
mathematically suggest uniformity and a uniform time base rate of change.
@skyway -- @dkrock rarely gets into details, but I bet he and @parisboy would see things eye-to-eye. A while back @dkrock likened PA to a winding river with a boat floating from bank to bank. And this is exactly what we are doing here -- I think. We float from side-to-side, but always tend back to the middle of the river -- before heading to the other bank.
  • #4,465
  • Jul 6, 2019 9:46pm Jul 6, 2019 9:46pm
Quoting robots4me
{quote} {quote} @skyway -- @dkrock rarely gets into details, but I bet he and @parisboy would see things eye-to-eye. A while back @dkrock likened PA to a winding river with a boat floating from bank to bank. And this is exactly what we are doing here -- I think. We float from side-to-side, but always tend back to the middle of the river -- before heading to the other bank.
He calls his method business analytics. He relies on dynamic calculations instead of static and linear lines. He correlates pricing with momentum to make trading decision.

https://www.forexfactory.com/showthread.php?p=12077277#post12077277
  • #4,466
  • Jul 6, 2019 9:50pm Jul 6, 2019 9:50pm
Hello R4M & BD, also to all people chipping in ideas to make this project a success (for me, even at this stage the strategy is very useful and effective). Thank you very much!!!

Can i add another option for exit? we can use the good old "Trailing stop" after price touched a few pips on the midline. This will squeeze some pips after the midline but most certainly we'll be stopped not being able to reach the other band. That's a trade-off. but it's ok sometimes price does not touch the opposite band and it returns to midline. The sure-ball gain is midline, plus some pips after it.

best regards
  • #4,467
  • Edited 10:39pm Jul 6, 2019 9:59pm | Edited 10:39pm
Quoting benjun9
Hello R4M & BD, also to all people chipping in ideas to make this project a success (for me, even at this stage the strategy is very useful and effective). Thank you very much!!! Can i add another option for exit? we can use the good old "Trailing stop" after price touched a few pips on the midline. This will squeeze some pips after the midline but most certainly we'll be stopped not being able to reach the other band. That's a trade-off. but it's ok sometimes price does not touch the opposite band and it returns to midline. The sure-ball gain is...
@benjun9 -- hold that thought -- perhaps after version 1.

I've coded and used trailing stops in the past and it was never clear to me whether or not there was a benefit. On the surface it makes sense, but in practice there are trade-offs that make it less effective.

At this point -- before considering modifications or additions I'm looking for evidence it will work. For example -- charts showing where the current strategy exits and where it could have been improved if a trailing stop had been employed.

Also, there are a variety of different types of trailing stops -- e.g. fixed, ATR, envelopes. Which would you choose, and why?


EDIT: @benjun9 -- I just reread your post and on second reading it now makes more sense to me -- my apologies. My brain saw "Trailing Stop" and immediately went into lecture mode. What you suggest falls more under money management -- rather than strategy rules. Let's see what @bishopdotun's thoughts are.
  • #4,468
  • Jul 6, 2019 10:43pm Jul 6, 2019 10:43pm
Quoting robots4me
Below is an example of applying a 10-period, simple MA to the MMDetective indicator: {image} Look at the portion of the price line labeled "Rising 10 Simple MA". Do you see how short the purple lines are and how long the blue lines are? This is not rocket science, and simply confirms -- visually -- why you should not be selling when a 10 SMA is rising. Similarly, looking at the portion of the price line labeled "Falling 10 Simple MA", you can see that the "lifetimes" of the purple lines are much longer than the blue lines -- visually confirming...
this is pivot point, very good alternative to approach a trade, but probably a confirmation would be also necessary.
  • #4,469
  • Jul 6, 2019 10:52pm Jul 6, 2019 10:52pm
Quoting raf10
{quote} this is pivot point, very good alternative to approach a trade, but probably a confirmation would be also necessary.
@raf10 -- thank you. And that was about 4464 posts ago. No problem commenting about it -- but currently I'm on to something else.
  • #4,470
  • Edited 11:06pm Jul 6, 2019 10:53pm | Edited 11:06pm
Quoting robots4me
{quote}At this point -- before considering modifications or additions I'm looking for evidence it will work. For example -- charts showing where the current strategy exits and where it could have been improved if a trailing stop had been employed. Also, there are a variety of different...
For close beyond mid-line exits, try this mtf accelerator change in the 4th decimal place and 2nd decimal place for jpy pairs, or exit at ADR,BD whichever is earlier.
Attached Image(s) (click to enlarge)
Click to Enlarge Name: 1.PNG Size: 45 KB
Click to Enlarge Name: 2.PNG Size: 60 KB
Click to Enlarge Name: 3.PNG Size: 52 KB
Click to Enlarge Name: 4.PNG Size: 65 KB
Attached File(s)
File Type: mq4 MTF_AC.mq4   7 KB | 409 downloads
  • #4,471
  • Jul 6, 2019 11:39pm Jul 6, 2019 11:39pm
Quoting skyway
{quote} For close beyond mid-line exits, try this mtf accelerator change in the 4th decimal place and 2nd decimal place for jpy pairs, or exit at ADR,BD whichever is earlier. {image} {image} {image} {image} {file}
@skyway -- it's another momentum-type indicator. Currently we are not using any filters with the BD indicator because its D1 signals are so accurate.

The SpikeHunter indicator does use a momentum-filter based on @Nih98's FX FSI indicator plus Stochastic Oscillator. Are you saying you've lost confidence in uncle @Nih98 ?

Below is a screen shot of the R4M-Momentum-Filter indi overlayed on the MT4_AC indi. I think R4M-Momentum-Filter gives accurate results -- plus, sometimes it reacts more quickly.

If you can show an example where adding a momentum filter actually reduces false positives in the BD strategy then I would gladly consider it. I just don't want to get into adding filters for the sake of adding filters. There needs to be evidence that it actually reduces false positives.

Attached Image (click to enlarge)
Click to Enlarge Name: MTF-AC-Indi.png Size: 17 KB
  • #4,472
  • Jul 7, 2019 2:30am Jul 7, 2019 2:30am
I'm ready to release an initial version of a trading EA that follows @bishopdotun's strategy and rules -- things look really promising. However, I've decided to do this over several posts. This is not a "teaser". Rather, there a number of disclaimers that people need to understand in advance. Also, I want to minimize the amount of chaos that is likely to follow. This EA can be run in MT4's Strategy Tester and there are settings that can be calibrated. So, if you are not yet familiar how to do this then you have a few hours to figure it out. The last thing I want to deal with are questions of how to use MT4 Strategy Tester, or why your results may be different than mine, or any of that stuff. So -- this is a heads-up. You can use any EA (even the default ones that come with MT4) to quickly learn how to use that tool and, especially, how to optimize / calibrate EA settings.

A few disclaimers:

1. This is not a Holy Grail EA. Yet, considering its very simple logic and conditions it works well across multiple pairs. Definitely some pairs seem to work better than others -- I'm not sure why. I'm hoping with all hands onboard we can collectively figure out the optimal conditions to use this EA. I am NOT presenting this as a completed project. Please do not ask for the best settings -- I don't know them. I'm hoping people can help fill-in the missing gaps by using MT4 Strategy Tester.

2. At this point I am only interested in D1. In fact, I've locked the EA so that it only works on D1. This does NOT mean it will forever be locked on D1 -- just that we know D1 works well and I want to take full advantage of having everyone on the same page when it comes to testing. That makes sense, right?

3. To begin with, I only want to test GBPNZD-D1. This pair seems to give the best results -- and I'll explain why in the next post. Before testing other pairs I want to make sure that people can reproduce my results on GBPNZD-D1. If everyone were to test different pairs then your feedback is less useful. So, to begin with, I want to work out any issues surrounding GBPNZD-D1 -- and then we can move on to other pairs.

It is midnight for me, but I still want to write a couple more posts. Again -- if you are not yet comfortable using MT4 Strategy Tester and if you are interested in helping to test this EA, then please get familiar now. And, just to be clear -- MT4 Strategy Tester is completely and totally NOT TO BE TRUSTED. However, in the absence of better tools it is still better than nothing. And it can be used to confirm that a strategy is executing as expected. However, the results you get using MT4 Strategy Tester usually have no relation to reality when trading in a live, Real account. And that is because (a) it uses different data, (b) there are known and unknown variables that differ from live trading, (c) and more.

To be continued...
  • #4,473
  • Jul 7, 2019 3:09am Jul 7, 2019 3:09am
A couple of days ago I posted screen shots of a few examples of the types of statistics I was seeing. Just to give you a flavor, the screen shot below is for GBPNZD-D1 using the data horizon 2018-02-01 through 2019-07-05. If you plan on helping with the testing, then I would request we all use the same data horizon.

Attached Image (click to enlarge)
Click to Enlarge Name: stats-GBPNZD-D1.png Size: 56 KB


The statistics that interest me most are:
1. Profit Factor
2. Win Ratio
3. Average Profit per trade, Average Loss per trade

In this example the Profit Factor is 3.71 -- amazingly, amazingly good. This will not hold up in a live, Real account -- but it is a good sign. Also, the WinRatio of 64% is very good -- especially considering the ratio of Average Profit Per Trade to Average Loss Per Trade is about 2:1. So, not only are there more wins than losses but, in addition, each win is about twice the size of each loss. Also, the worst draw down is < 1%.

As I mentioned in the previous post -- results from MT4 Strategy Tester are not to be trusted. However, in the absence of 1-year's worth of Demo testing then it is all we can go on for now.

Another thing to note -- in the parameters section there are 3 settings that have a big effect on the strategy's performance:
1. StopLossFractionADR -- our StopLoss is a fraction of the ADR (e.g. 1.4)
2. TakeProfitFractionADR -- our TakeProfit is also a fraction of the ADR. In the example here this fraction is '2', which basically means it will never get hit. Instead trades are closed by the Exit condition.
3. NumberDaysADR -- number of days used when computing the ADR

I need to hit the sack shortly. However, I'll finish posting and release the EA before the London market opens tomorrow.

To Be Continued...
  • #4,474
  • Edited 3:31am Jul 7, 2019 3:16am | Edited 3:31am
Quoting robots4me
I'm ready to release an initial version of a trading EA that follows @bishopdotun's strategy and rules -- things look really promising. However, I've decided to do this over several posts. This is not a "teaser". Rather, there a number of disclaimers that people need to understand in advance. Also, I want to minimize the amount of chaos that is likely to follow. This EA can be run in MT4's Strategy Tester and there are settings that can be calibrated. So, if you are not yet familiar how to do this then you have a few hours to figure it out. The...
That is a very good message to read on Sunday morning here after the breakfast and before going to the golf course, my friend (guessing we are in a similar age)...
By the way, may I pay you a compliment...you are the best OP I have seen here on FF in many years, almost always present around the clock with full nick, very objective and last but not least very conciliatory...and not forgetting an awesome coder!

Please stay tuned, jdva
I never lose - either I win or I learn...
  • #4,475
  • Jul 7, 2019 3:29am Jul 7, 2019 3:29am
Quoting robots4me
I'm ready to release an initial version of a trading EA that follows @bishopdotun's strategy and rules -- things look really promising. However, I've decided to do this over several posts. This is not a "teaser". Rather, there a number of disclaimers that people need to understand in advance. Also, I want to minimize the amount of chaos that is likely to follow. This EA can be run in MT4's Strategy Tester and there are settings that can be calibrated. So, if you are not yet familiar how to do this then you have a few hours to figure it out. The...
Great news on a Sunday morning.
  • #4,476
  • Jul 7, 2019 3:33am Jul 7, 2019 3:33am
Quoting robots4me
A couple of days ago I posted screen shots of a few examples of the types of statistics I was seeing. Just to give you a flavor, the screen shot below is for GBPNZD-D1 using the data horizon 2018-02-01 through 2019-07-05. If you plan on helping with the testing, then I would request we all use the same data horizon. {image} The statistics that interest me most are: 1. Profit Factor 2. Win Ratio 3. Average Profit per trade, Average Loss per trade In this example the Profit Factor is 3.71 -- amazingly, amazingly good. This will not hold up in a live,...
No curve fitting
No training
No complicated maths

Coffee pls ️
  • #4,477
  • Jul 7, 2019 3:35am Jul 7, 2019 3:35am
Quoting jdva
{quote} That is a very good message to read on Sunday morning here after the breakfast and before going to the golf course, my friend (guessing we are in a similar age)... By the way, may I pay you a compliment...you are the best OP I have seen here on FF in many years, almost always present around the clock with full nick, very objective and last but not least very conciliatory...! Please stay tuned, jdva
Quote
you are the best OP I have seen here on FF in many years

@jdva -- I appreciate the compliment -- thank you. And the tangible and intangible lessons I have learned here at FF have been worth more than gold to me -- so, it's been a pleasure.

Also, I feel that @bishopdotun's strategy was a gift handed to us on a silver platter. And that changes the complexion of everything -- to be able to work and share remotely on something this cool doesn't come around that often. So, again, thank you @BD...
  • #4,478
  • Jul 7, 2019 3:40am Jul 7, 2019 3:40am
Quoting bishopdotun
{quote} Great news on a Sunday morning.
BD, my name sake. Thanks for bringing this trading idea. Placed two trades last week purely on your strategy and both were good.
To R4M, you are one guy who won't give up. I started with u when this thread started, right from George's. Thank you once again.
My level of knowledge is not close to yours that's why I'm not commenting as I should. A student should learn to listen, Lol.
God bless everyone
Only Jesus Saves
  • #4,479
  • Jul 7, 2019 3:41am Jul 7, 2019 3:41am
BD set to 'Exit when price crosses OPPOSITE band'. The 'red X' to exit is appearing on same side band. This is a false visual cue of an exit on the same side? Or will it hold the trade and exit the opposite side accordingly?
Attached Image (click to enlarge)
Click to Enlarge Name: Capture.PNG Size: 45 KB
  • #4,480
  • Jul 7, 2019 3:53am Jul 7, 2019 3:53am
R4M ADR All Pairs (R4M-ADR-AllPairs.ex4)

R4M ADR All Pairs is a new indicator that computes the ADR for all symbols in your Market Watch window. The results can be sorted alphabetically, ascending or descending. Also, you can change the number of days used when computing the ADR.

All indicators are bundled in a *.zip and can be downloaded from the other thread -- "Custom Indicators for TMA-ADR Reversal Trading and BBMA (https://www.forexfactory.com/showthread.php?t=920925).

Attached Image (click to enlarge)
Click to Enlarge Name: R4M-ADR-AllPairs.png Size: 11 KB


The reason I developed this indicator is because I noticed when testing the new EA that symbols with larger ADRs seemed to perform better. Don't quote me on this -- it's a cursory observation. You can see that there is quite a range of ADR values -- from a high of 115 pips (GBPNZD) to a low of 33 pips (USDSGD). This may (or may not) come in handy as we test the limits of the new EA.

Because of this variation in ADRs then I think the parameters for StopLossFractionADR and TakeProfitFractionADR may be particularly sensitive to which symbol is being traded. The TakeProfitFractionADR may not be so important because the Exit condition usually gets hit first. But the StopLossFractionADR may need to be tailored to each pair -- we'll see.
1 trader viewing now