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

Actually I am not. I will manually exploit the edge anyways.

We are interested in the long run profit. Lot of pips in the 🧺.

Thank you for the indicator swingman ........,.,worksfine

I have read the majority of this thread in relation to the BD indicator at least twice. I am looking for the Proxy EA to trade spike hunter with. I know that the current BD iteration will not use the Proxy EA anymore, and I know that Spike hunter is sitting on the way side for now. I still wanted to test it out though. But I guess that the Proxy EA was not released? Spike Hunter already has the option to work with TradeManagerEA built in so I am under the impression that it will work. Thanks for any answers.

@cardm3n -- an earlier version did work with the proxy EA, but since there was little interest I've since removed it. Even so, that version did NOT automate trading. All it did was add a few buttons to make manual trading easier. You can still use SpikeHunter with a 3rd-party dashboard trading EA -- that would be equivalent to what TradeManagerEA provided. Or you can manually trade using MT4's UI.

Thank you for the reply. I will have to research which dashboard EA will work with it. if you could point me towards one, that would be great. I would trade it manually if it weren't for my work schedule. its almost impossible for me to trade anything under daily time frames.

Also, I am curious to ask why spike hunter was set aside. You mentioned that it repaints, but when I run it through strategy tester, smLazy TMA doesnt seem to. I am also going to go back through the entire thread eventually. So many pages though lol I barely got caught up with the BD stuff. but plenty of info to read is only a good thing.

@cardm3n -- a dashboard EA is also manual trading -- the only difference is the UI is simpler to use than using MT4's UI. A 3rd-party dashboard EA has no connection to an indicator or any trading logic -- it's only BUY and SELL buttons.

@cardm3n -- if you ask this question then it means you have more reading to do -- when you get caught-up it will then become obvious. I may return to SpikeHunter -- but not at the present. I understand what a pain it is to have to read through hundreds of pages of posting -- but that's one of the trade-offs with public forums. It's good for interacting and exchanging ideas / questions, etc -- but if you come late to the party then you'll have to get caught-up to understand why things are where they currently are.

I appreciate your response. I am currently working my way through the thread. Thank you.

MT4 Strategy Tester

I would like to go over some guidelines on how to use MT4 Strategy Tester with the BD EA -- which I will release in the next post. If you don't plan on using Strategy Tester then no need to read this -- but you'd be missing out on a lot. Plus, the more people who test then the better chance we have of flushing-out bugs and finding the best settings for different pairs.

If you are not familiar with using Strategy Tester then read-up on your own -- I'm not holding anyone's hand on how to use it. It shouldn't take more than 10 - 15 minutes to figure out how to use it for testing and optimizing settings. You can start with one of the simple EAs that MetaQuotes includes with your MT4 installation.



Tester Settings: Symbol: GBPNZD <== please stick with GBPNZD for now
Model: Open prices...
Use date: 2018.02.01 To: 2019.07.05 <== this will result in about 100 trades

Period: D1 (only) <== this version of the EA is locked to D1

A note about the data horizon (From: 2018.02.01 To: 2019.07.05) -- whenever back testing there are two things to consider:

1. Use recent data -- i.e. do NOT go back 20 years (unless you have a time machine). Since we trade in the present then we want to use data that reflects the present. If you include old, stale data then it will skew the results. Using old data does NOT make an EA more robust -- there is no evidence for that and, in fact, it's quite the contrary. People say that because we've been trained that more is better. But more old data is NOT better.

2. For results to be statistically significant we need a minimum sample size -- i.e. a minimum number of trades. 100 trades is a good number to shoot for. So I've chosen a data horizon that results in about 100 trades.

Inputs:

There are 3 important inputs -- NumberDaysADR, StopLoss fraction of ADR, and TakeProfit fraction of ADR. To include these in the optimization you'll need to check the CheckBox next to the parameter and provide Start, Step and Stop values. In the screen shot above I've shown examples of the values I've been using.




For the optimizer to work you'll need to check the 'Optimization' check box and then press the 'Start' button. It takes several minutes to complete, but when it is done you can view the results using the 'Optimization Results' tab.

Additional Notes:

1. In my hands there are some pairs I could not test using Strategy Tester -- i.e. the results were blank. I'm not sure why. Perhaps the data horizon is too big and my broker does not provide enough historical data -- I don't know. Perhaps someone here has an answer or can figure out why some pairs can be tested and others can not.

2. In the short-term I am primarily interested to see if people can reproduce my results using GBPNZD-D1. Once we agree that results are reproducible then we can move on to other types of testing.

3. The default settings for the EA where chosen based on my results for GBPNZD-D1. It is for sure these will not be the best results for other pairs:
NumberDaysADR: 20
StopLossFractionADR: 1.8
TakeProfitFractionADR: 0.4

4. There are two additional parameters of interest, but I would recommend sticking with the defaults for now:
MaxSpreadPoints: 40 -- ignores trade signal if spread is > 40 points
MaxOpenTrades: 10 -- ignores trade signal if there are already 10 open trades

5. Hidden Exit rule -- there is one additional Exit rule that I've incorporated that was not part of @bishopdotun's original set of rules. And that is if the strategy generates a SELL signal then it will close any BUY trades that are still open (and vice-versa).

I understand there is a lot of information here -- but the details are important. Please take the time to understand what I've taken the time to write. It will save us all time later. And time is precious...

To be continued...

I tried to run on STest. However my version and you are different. Is it different from the version?

@anhthu049 -- that's the dashboard EA. In MT4 a developer has two choices -- to build an indicator or build an EA. An indicator does not have access to APIs for buying and selling. An Expert Advisor does have access to APIs for buying and selling.

Please note that EA does NOT necessarily mean it automatically trades. The dashboard EA you refer to has buttons -- pressing a button generates a BUY or SELL trade -- hence it was built as an EA. But it can only be used for manual trading.

I haven't yet released the automated trading EA -- I will do that shortly.

R4M BD Trading EA (R4M-BD-ADR-EA.ex4)

This is the trading EA that relies on signals generated by R4M-BD-ADR indicator. To use the EA simply attach it to a D1 chart of the pair you wish to trade -- it MUST be a D1 chart. As with the other indicators and EAs, it can be downloaded from the other thread -- "Custom Indicators for TMA-ADR Reversal Trading and BBMA" (https://www.forexfactory.com/showthread.php?t=920925). All indicators and EAs are bundled in a *.zip file.




There are a few settings to be aware of:

* The most important is 'StopLoss fraction of ADR'. The current value of 1.8 works well for GBPNZD, but probably is suboptimal for other pairs.
* 'TakeProfit fraction of ADR'
* 'NumberDaysADR' -- number of days used when computing the ADR (20 days is a good default value).
* 'MaxOpenTrades' -- if the number of open trades exceeds this value, then new signals are ignored.
* 'MaxSpreadPoints' -- if the current spread exceeds this value, then new signals are ignored.

It is best not to overload a chart running a trading EA with a bunch of indicators. Each chart is allotted a fixed time-slice of the CPU to perform all its computations. Once the time-slice has been exhausted then MT4 switches the time-slice to the next chart -- regardless whether the previous chart had finished all its computations. Each indicator you attach to a chart needs to perform its own computations and these count against the chart's allotted time-slice. Having said that, I think it is probably okay to attach the R4M-BD-ADR indicator to the chart so you can visually confirm that trades occur as expected.



The screen shot above shows an example of the BD EA and indicator running on the same chart.

Transparency

One thing people may not be aware of or appreciate is how transparent @bishopdotun and I have been in presenting his strategy along with indicators and EAs. No magic sauce -- no riddles, no puzzles, no treasure maps to decipher. No 10000 hours of screen time required. The strategy's rules are simple and that is what I've implemented.

So, now, when you use the indicators and EAs you can understand and follow exactly what is happening. Furthermore, if you spot something unexpected I hope you will post a screen shot and explanation so I can reproduce it and see if there is a solution. In this way we can continue to refine and improve the BD indicators and EAs.

And, finally, though I am not comfortable releasing all my source code *.mq4 and *.mqh files -- I've decided to share a snippet of code for those who are interested in developing their own indicators or EAs based on the signals generated by the R4M-BD-ADR.ex4 indicator. Below is how my EAs use the iCustom() API to access the signals. If you were to develop something interesting, then I hope you'd be gracious and generous to share with the rest of us.


#define INDEX_EXPORTED_BUY (14)
#define INDEX_EXPORTED_SELL (15)
#define INDEX_EXPORTED_EXIT (16)

//+------------------------------------------------------------------+
//| |
//+------------------------------------------------------------------+

// NOTES: iPeriod is PERIOD_D1
// iPercentADR is '100'
string g_strCustomPath_BishopDotun = "R4M-BD-ADR";
bool CustomValue_BishopDotun(string strSymbol, int iPeriod, int shift, int iPercentADR, int& iExportedBUY, int& iExportedSELL, int& iExportedEXIT,
int iMaxBars = 500, int iNumberOfDaysToAverage=20, ENUM_EXIT_CONDITION eWhichExitCondition=101)
{
ResetLastError();

iExportedBUY = (int)iCustom(strSymbol, iPeriod /*timeframe*/, g_strCustomPath_BishopDotun,
"", "10080", iPercentADR, iMaxBars, iNumberOfDaysToAverage, eWhichExitCondition,
INDEX_EXPORTED_BUY, shift);

iExportedSELL = (int)iCustom(strSymbol, iPeriod /*timeframe*/, g_strCustomPath_BishopDotun,
"", "10080", iPercentADR, iMaxBars, iNumberOfDaysToAverage, eWhichExitCondition,
INDEX_EXPORTED_SELL, shift);
iExportedEXIT = (int)iCustom(strSymbol, iPeriod /*timeframe*/, g_strCustomPath_BishopDotun,
"", "10080", iPercentADR, iMaxBars, iNumberOfDaysToAverage, eWhichExitCondition,
INDEX_EXPORTED_EXIT, shift);
int iLastError = GetLastError();
if (0 != iLastError)
{
// NOTE: 4072 ==> "Cannot load custom indicator"
string strMessage = StringFormat("*** [FAILED]: %s: %s: ErrorCode: %d: -- %s",
"CustomValue_BishopDotun", strSymbol, iLastError, ErrorDescription(iLastError));
Print(strMessage);
Alert(strMessage);
return (false);
}

return (true);
}

This is the champion.
I think it is time, I invite my family to fine dinning again.

I just tendered my resignation.

I didn't play with the settings yet. The strike rate is good.
1 question - over a 1 year period it just have 3 to 5 transactions, will you consider it as low?

May I interest you in the latest and hotest Riviere, Lattice One, Sky Everton, Juniper Hill or anything else that might interest you let me know.

Today duty at Sentosa guard post. Maybe Forest for lunch.

Trading wisdom.