What do people think of this indicator?

Hi all,

I rarely use forums, I've traded for a number of years and experienced the usual ups and downs.

I'm a coder, I've spent years on my own crazy complex indicator - basically hundreds of calculations that boil down to a single line indicator. If it goes up, buy, if it goes down, sell.

For the most part, it's served me well. The losses always come from bad money management, over-confidence, etc. This confirms for me that no indicator can make you rich as it's how you trade with that indicator that makes or breaks a trader.

However, I'm confident in the work I've done but I'd like to see how other traders might use the indicator. I think there may be some insight from forum members that will result in:

1. me using my own indicator better
2. me helping other traders do the same

At this stage, I'm only providing CSV file output for 3 months of indicator data, not an MQL or EX4 file. There are two reasons for this:

1. my indicator relies on data that's pre-processed outside of MT4, data from various web sources, etc... it's complex and it's not "packaged" for easy distribution
2. I'd like to see if anyone is interested in providing an unbiased / raw interpretation of the indicator. i.e.; without knowing how it's calculated. What correlations do you see? How do you think the indicator could be used?

The attached CSV files are based on the 1HR timeframe. So the data is simple timestamp + "," + number:

2017.12.28 09:00:00,26.62787319583332
2017.12.28 10:00:00,26.62787319583332
2017.12.28 11:00:00,27.65908473429486
2017.12.28 12:00:00,27.65908473429486

If anyone wants to write a simple indicator to read in the text file for a given currency pair, loop through each line, and plot the numbers (on the 1HR timeframe), and post the MQL file here so others can use it, that would be great.

I've attached a screen shot of how my indicator looks on the GBPUSD chart. A simple orange line. The higher the line goes up the stronger the signal to buy, the lower it goes, the stronger signal to sell. If it's at 0 then there's either no information to display for that period, or it's because of an equal buying and selling signal.

I'm not here to discuss trading styles (stop losses, risk/reward, etc, etc). That's a discussion to be had that's mutually exclusive to any discussion about indicators. I'm here just to discuss how this indicator might help your trading.

UPDATE: Use the file "8 currencies v3.csv"

I've already written two indicators (attached below) that read data from a CSV file and plot it on a MT4 chart. By a happy coincidence your CSV files are formatted perfectly (i.e. date/time in standard MT4 format, followed by a numeric value; file sorted into ascending date/time sequence; .....), so that I can simply use the default settings in my indicators to plot your data. The attached screenshot shows those settings, and the resulting plots on EURUSD.

NOTE: These indicators were written using MetaEditor/compiler build 509. If you want the .mq4 source to compile/run without errors, you'll need to do this first. Otherwise, download only the .ex4 files.

Two potential problems, as I see it:

1. Unless you're willing to share your indicator, you'd somehow need to supply fresh CSV files on an hourly basis, and there would need to be an automated process for downloading them into the correct MT4 folder on my computer, for this to be useful for live trading.

2. As a discretionary trader, unless I have at least a vague idea of the type of data that your indicator is plotting, I'm afraid I wouldn't have the confidence to use the data, when trading a real money account. I would be placing in my trust in a 'black box' that I effectively know nothing about.

How should we best proceed from here? Ideas, anybody?

ah, nice work.

I'll modify Plot External Data.mq4 to work in the latest build. But what's this for:

#include <hanover --- function header (np).mqh>

Well ok, many thanks. I prefer to work with the old build 509 compiler as I've written many indicators and EAs, some of them complex, and I would need to learn OO (I'm an old COBOL/C programmer from the 1970s/80s) and then rewrite them all, at a time when I'd rather focus on improving my trading. When my EX4s no longer work on the latest MT4 builds, or MT5 phases out MT4 altogether, then obviously I'll need to reconsider my position. In the meantime, pips4life has kindly converted the MQH function libraries that I originally wrote, to extend of the old MQL4 language, here. Hopefully the code is fairly self-explanatory.
That's a very challenging task, especially considering that the data was sourced outside of MT4, I mean it could be interrelated markets, economic data, CME-sourced data, astrology, a composite index built from any number of different markets, ...... anything!

Anyway, I think I'll move on now, unless/until the two issues that I noted in my original reply are addressed.
_____________________

Just before I disappear, however, I've made a couple of quick enhancements to the attached indicator:

YaxisValues: valid entries are now as follows:
 Field is left blank: no (additional) Y-axis will be plotted
 A series of up to 50 ascending (integer or decimal) values, separated by commas: these levels will be plotted on the chart. Example: -30,-20,-10,0,10,20,30
 Four entries: an asterisk (*), followed by the starting/lowest level, step/increment, and ending/highest level, all separated by commas. Example: *,-30,10,30 would generate the same levels as in the previous example (start at -30, increase in steps of 10, end at +30)
 A question mark (?), optionally followed by a positive integer, separated by a comma. Example: ?,5 will divide up the highest and lowest data values plotted into 5 equal intervals, calculate the step value across each interval, and plot a level at each step. For example if the highest data value is 50 and the lowest is -50, then ?,5 would plot levels at -50,-30,-10,+10,+30,+50
 A question mark (?): same as the previous, although it will divide up the chart into the default value of 10 intervals, i.e. ? delivers the same result as ?,10

Note that you must press F8 and check 'Show object descriptions' on, if you want the Y-axis number values to print above the plotted horizontal lines (as shown in the screenshot).

YaxisSettings: new options are:
 Leave the first entry blank: the line will start plotting at the left of the chart
 Leave the second entry blank: the line will be plotted as a ray, extending to the far right of the chart. Any negative value will also cause a ray to be plotted

NOTE: This indicator was written using MetaEditor/compiler build 509. If you want the .mq4 source to compile/run without errors, you'll need to do this first. Otherwise, download only the .ex4 file.

I've modified Plot External Data.mq4:

1. I couldn't work out why you used: int h = FileOpen(FileName, FILE_CSV|FILE_READ,'~');... why "~" ? I made it use ","
2. Checks for errors using GetLastError() after attempting to open the file
3. No timestamp reformatting. Assumes it's in the correct MT4 format
4. in the refactoring step, allows values of < 0 otherwise the indicator can never plot a negative number
5. Increases the array sizes of dt and val, rather than hard coding to a size to 9999
6. Added a grey line at the 0 point so you can easily see when the indicator is positive or negative

See attached image (EURUSD) as example, and new MQL file.

I've also attached two examples (zoomed into EURUSD) of how I use the indicator. I've attached "short term" and "long term" images with notes on the images.

I've also attached new CSV files (and the latest ones are also uploaded to that FTP folder) as I outputted the wrong data initially - sorry about that.

Yes, CSV files would need to be outputted and uploaded somewhere each hour. I've attached a .BAT file so you can use Task Scheduler on Windows to run it each hour, or run it manually, or you could implement some other similar solution. It uses FTP. I've created an FTP account for this thread and that account is used in the .BAT file. If there's any traction on this thread I'm happy to schedule FTP uploads of CSV files each 30 minutes.

NOTE: in the .BAT, change the save location of the CSV files to your MT4 installation. So replace "294B6FCE6F709DE82DA4C87FDBF1DE36" with your ID.

As for trust in the data... of course, your comments make sense. It's hard for me to write a basic set of rules as to how the indicator is calculated and so having faith in it is understandably difficult. It's taken me over 2 years of refinement and there's about 100 algorithms running to generate that simple little line you see in the indicator.

But what can I say... um... I'm largely technical in my trading. I focus heavily on volumes (Wyckoff, etc), and therefore "smart money" is always top of mind for me. Perhaps 30% of the number crunching is related to VSA. I spend a lot of time in probability analysis in order to reduce trading risk. So I do a lot of number crunching on news cycles, and the analysing market response from previous news announcements in order to gain an insight into the probability of market shifts on future news cycles. About 30% would be on that. The remaining emphasis would be spread across all sorts of little algorithms; upper timeframe trend, price histograms, volume histograms, RSI, Fibs, Pivot Points, and of course good ol' support and resistance. All of it focuses on probability analysis rather than following the generally accepted rules / settings on such indicators. So I take the concept of support and resistance and calculate the probability of each support and each resistance zone being hit. All algorithm results are given weightings and merged into a single buy / sell line in the indicator.

I feel as though the indicator regularly predicts short and long term shifts as hopefully you'll see from the attached examples. That's all the indicator does. As for whether I open a trade, at what lot size, what risk I take, etc... that's all another complex method and not worth discussing as everyone has their own psychology / methods surrounding that.

If anyone is happy to put the indicator on their charts for a while and see if it's predicting things well enough for them, then I'd love to hear how they are interpreting the indicator and using it to trade.

UPDATE: Latest MQL file attached
UPDATE: Use the file "8 currencies v3.csv"

It sounds like your indicator is the result of some extremely dedicated and impressive research. Given the amount of time, effort and expertise that you've obviously put into it, I totally understand your reluctance to share your IP publicly. I would feel exactly the same way myself.

I started my research into the financial markets in 2001 and specifically Forex in 2006. Although trading -- and occasionally tinkering with automated and semi-automated systems -- is really not much more than a hobby for me, I'm happy enough with my existing methodology (S/R, volume profiling, orderflow, and statistical analysis are all concepts that I find valuable also). However, when I saw that nobody had replied to your post, and I had a couple of 'CSV data plotting' indicators that I thought might be useful here, that was my main reason for replying.

Nonetheless, your indicator looks very interesting. Ability to anticipate bias can be useful for both entries and exits. From experience I know that it usually takes me 6-12 months to assimilate and test a new trading concept, to the point that I feel comfortable running with it on a live account. Hence I would need to decide whether it is worth the time and effort, given my current situation.

In the meantime, thank you very much for sharing, and I wish you all the best.

amazing ,
just subscribed until be hot this one

subscribed...
will try to use this indicator...

contrarian indicator?

Well, the goal isn't to be a contrarian indicator but it looks like that I agree. I tend to avoid any labels at all with life in general, and therefore my trading. I just do whatever the statistical and other analysis indicates, and if that results in what people call "contrarian", then faire enough.

Keep in mind that the algorithms running will naturally adapt to different currencies and market conditions. So it may seem "contrarian" on that particular chart, and that particular time but that doesn't mean it'll always be like that.

If you're looking at my buy and sell trades on the chart, keep in mind that my sells were based on the indicator but I entered way too early. As I mentioned in my first post, it's my psychology / entry / exit / etc that needs improving. I actually think the indicator is great. I'm the problem So my buy trades on that chart are putting me into a hedge.

As of today (Monday 8th), the EURUSD indicator is definitely telling me to sell. Except I'm already in 3 sell positions so I'm not going to open another, although I'm sure I'll regret not opening another! So, this is why I'm finally opening up my indicator to the public. I need help using my own indicator!... and I think people will definitely benefit from it (but then again, everyone who posts here thinks they have something fantastic to offer lol)

ForeverNewb, I'm actually not too sure what your markups on my charts mean. I thought maybe they were pointing to the trades I had open but looking again I don't think that's the case.

Just to confirm... A "sell signal" is either when the line quickly moves down, or is clearly in the negative. The opposite for a "buy signal"

ForeverNewb... ah, now I see what you're doing. Well, it does appear that just as the line crosses from positive to negative, it would have been a good time to BUY, even though going into negative actually is a SELL signal.

That's an interesting observation though. I would never enter a SELL just as the line crosses into negative; I'd wait until it's clearly sitting in the negative and continuing to go down, but I do tend to enter too early. So perhaps my indicator is picking up the strength or weakness too early, and therefore perhaps it's worth me looking at using the "crossover points" as potential trades... so, for example, when it goes from a "clearly" positive position for a "while" and drops down and just touches into the negative, I should BUY, knowing that it's about "mid-trend", and then close that position and SELL if / when the line continues further into the negative.

"clearly" and "while" would need to be defined.

i wish i know how to code

I've modified my indicator to show the buy line, sell line, and then the final line (= buy minus sell).

This helps to show the reason why, for example, the final line is in negative... is it because of lack of buy signal? lack of sell signal? maybe it's a combination of buy and sell signal? etc.

New MQ4 file and CSV files attached. I'll update the previous posts with the latest files to assist new readers to the thread.

Updates (all previous files in previous comments have been updated):

1. .BAT file downloads any CSV file (*.csv) so any new currencies exported to CSV will be downloaded to your MT4 install
2. changed location of files to /CSV in order to support (1) above
3. MQL file only calculates on first tick (it was previously calculating on every tick)
4. MQL file works out the file name based on the Symbol(). Previously you had to select the file name in the indicator options
5. My system will regularly upload CSV files to the FTP server

truelifeajf,

apologies for the unclear message, it seems I f*k up somehow with the image attachment and lost part of the message... although I am an avid reader of this forum, I am a little poster (so much to read ..duhh) and that may well be only the second image I post. As the preview didn't show the image as full but only as "downloadable", tried a few things, and lost some of my writing in the process... I ll try to remember for next time

In short, I was pointing to your 1st message in which you said:
but looking at the images only (my childish side) it didn't look like that, so simply was pointing to when it crossed down to the negative, reasonable buy signal, and when it crossed to the positive, reasonable sell signals, with (eyeballing) small loss, and much bigger potential wining (though no exit strategy yet!!! just eyeballing to the max of move of price, hence the "potential"). Was therefore wondering if I misunderstood anything as this is opposite to what I understood of your 1st message. I was also wondering if in your programme you may have forgotten a negative sign or other "bug".

that is something I always wondered: is a move up because lack of selling or increased buying? and all the variations of the same questions. So far I haven't been successful with devising a concept for indicator that convinces me. All I ended up with in my research is that nothing I tried seems to beat the simplicity of a few MAs and some indis based on these to highlight things sometimes difficult to pinpoint with the naked eye. Complex indicators with a bunch of calculations simply slowed down my MT4... That lightbulb came after a period of playing with volume/ticks/etc and finding out that signals were actually coming at exactly same time as what some slightly tweaked MAs would give.
That is also something that stroked me when I saw your single line, it didn't look better than more common oscillators (overlooking the contrarian stuff).
Now with your three lines, that is getting very interesting (conceptually) because that is something I never made on lower than daily TF, so I am very curious about which data you are using (though I may have an idea, likely wrong, but not something I could code to test anyway). I think I finally need to download your data to try out more stuff

In conclusion, I am wondering if your indi is working well for you (as you say "For the most part, it's served me well") simply because somehow you trust/understand it more than more common oscillators (you put so much effort into it, you know what it is based on, etc.. you say yourself, you are confident in the work you've done), not because it is intrinsically better than a MACD for ex.
But that is based only on a few images so far and my little experience, so don't take my word for it ( I took your message as seeking constructive criticism, so that is what I am trying to do) .

Thanks for your comments ForeverNewb,

Unlike an indicator which is calculated purely on price action and volume (as that's the only data available to MT4), my indicator uses data not available within MT4. Therefore, while it may look like an oscillator, and you may well find an oscillator that is close to the line that we've been using in our example screen shots (EURUSD, going back 2 months), I can assure you this indicator will deviate from any correlation you find between this indicator and an oscillator indicator.

As mentioned in previous posts, about 30% of the calculations is based on volume (tick volume, huge data stores), 30% on news and other fundamental data that can be analysed, and then a whole range of basic trading principals, again with a focus on statistical analysis of those principals rather than the principals themselves... SR, some price action, pivot points, price density histograms, etc.

It's interesting that the "contrarian" concept came up simply because my goal is to follow smart money (lots of VSA stuff going on in that indicator), and smart money is obviously contrarian (because they're the ones making money while everyone else isn't). So, VSA combined with my other calculations (about 100 of them)... if that's showing a contrarian indicator then great

But that's not to say it'll always present itself that way. I have no fixed trading strategy. But this indicator is the "boiling down" of a huge array of considerations I would like to make before considering any trade... I seem to have boiled it down to a single line - something I didn't expect I'd do, but here I am.

So yes, the indicator is working well - because it's an indicator, not a trading strategy. My trading strategy is up and down, usually because I get too confident, enter too early and then have to deal with the consequences For that reason, I'm wondering who might be interested in testing out my indicator without me (at this stage) revealing all the details of my many years of calculations. Insights from people here may help me with my trading strategy.

I did a few tests.
Unfortunately I am not on PC, and being seriously retarded computer-wise, I couldn't open the csv in MT4, so tried through excel, then some math prog.
What I did is test EU and AU, multiply your data (the resultant single curve, not the buy and sell; in red in graphs below) by -1 (for convenience wrt contrarian stuff) and tried to scale it with corresponding close price (by eyeballing, so not optimum fit) (in black in graphs below; haven't tried more exotic prices which may fit better) (the green line is the zero line of the indicator):





without knowing what is in your indicator and the logic behind it, it is difficult to tell more than: you have found a complex way to reproduce a price oscillator.

You picked two charts that happened to have a high correlation over the last couple of months.

I added a plot of correlation (white line) to EURUSD and yes, for all of November the indicator line, when inverted is closely correlated to the price. But... from about 12th Dec correlation shifts drastically.

AUDUSD is a similar story.

Had you picked another chart, e.g.: GBPUSD, you would see far less correlation.

And I also uploaded a screen shot for USDJPY correlation which when eyeballed, appears correlated but in fact isn't when we calculate the correlation (white line).

And it makes sense there's correlation (inversely) because there are many VSA calculations in my algorithm and, for example, identifying weakness in a downtrend, and buying into it, is very much a VSA concept... a contrarian concept.

But even with common inverse correlation, I think it's easy to overlay the inverse of the indicator on the price chart and figure that it's just a price oscillator and conclude that the indicator is simply saying, "if the price goes up, then sell, and if it goes down, then buy, and the more it goes up or down is a stronger signal to buy or sell."

But the interest is whether the indicator line is in positive (buy signal) or negative (sell signal). See attached "trade points EURUSD.png". I've highlighted 2 places where I would have entered a buy, and 2 for a sell.

1. The first one is a BUY, price drops, indicator rises with inverse correlation
2. The second one is a SELL, price rises, indicator drops with inverse correlation
3. The third one is where my point sits... A similar thing happened with the price... it rises, BUT, the indicator is in positive (BUY). So in this case, the price is rising but the indicator is saying BUY. It's easy to miss that if you just overlay the indicator on the price chart and look at the OVERALL correlation.
4. The fourth... This is where the indicator and price lose correlation, and is a SELL signal

So my point is that even when there's inverse correlation (the price goes up, and the indicator line goes down), the indicator going down doesn't mean sell. The indicator with a negative value means sell. The more negative it goes, the more confident I can be that selling is a good move. If the indicator line is in the negative, and it jumps up a bit (but still is in the negative), then it often represents a scalping opportunity - just some short term strength, even through the indicator is showing an overall sell signal.

I should point out that the main indicator line (orange one) mostly sits between 40 and -40. So generally, if the line moves up to about 20, I would buy, and if it moves down to around -20, I would sell.

I've attached a section of the EURUSD chart, but on the 5 minute timeframe. It's to show another example of when the indicator appears to just be a complex price oscillator but isn't really.... upon closer inspection we can see the price started going up for 5 days while the indicator flatlined, just in the positive. After 5 days of price rises, the indicator then dropped into the negative, and the indicator didn't really get low enough to be a sell signal until a few days later as the price started hitting what appears to be a double top. So, yes, the inverse of the indicator line is correlated to the price, but whether it's moving before or after price, and whether it's moving above or below 0 is the important bit.

Finally, I've attached 3 examples showing how the indicator can sometimes give a signal a little late (example 2), sometimes early (example 3), and then example 1 where it gave a signal to sell but then after a bit of price movement down, the indicator line started moving back into the positive. So sometimes new information comes in that "undoes" a signal it just gave.

Good morning truelifeajf,

thanks for the additional explanations, it is clearer now how you read the indi.

I guess my point is: does your indi perform better than less complex indi? If it does, what do you care what other people think of the indi? if it doesn't, then it is just another line. But not knowing the logic of the indi makes it hard/impossible to really provide an evaluation.
And of course, as you already mentioned, the indi is one thing, the trading is another.
For ex, here is a simple contrarian PO indi I made quite some time ago. no hard/tight SL (just one far incase of the unthinkable happens), 50 pips targets, 8 out of 8 winners for the same EU period you show. Yet I never got the balls to trade it, still working on my iron cast... the wide security stop is something I can't sleep with.
What do people think about this indicator?...



In other words, how do you benefit from the complex calculation of your indi? As I said in my earlier post, I believe (obviously may be wrong) it is simply the confidence it gives you because you know the indi and it makes sense to you. You can give all the examples you want, if you don't explain the logic of the indi, difficult to go beyond beliefs. Don't get me wrong, I am not asking you to disclose your work in public (though I am very curious about it). But maybe you should rethink the question you are asking.