![MQL5 - Language of trade strategies built-in the MetaTrader 5 client terminal](https://c.mql5.com/i/registerlandings/logo-2.png)
You are missing trading opportunities:
- Free trading apps
- Over 8,000 signals for copying
- Economic news for exploring financial markets
Registration
Log in
You agree to website policy and terms of use
If you do not have an account, please register
The topic is called Market Patterns.
What do I know about it? A lot (specialising in Level II analysis).
I'll show here from today several of these regularities in pictures. And there is one pattern with 100% efficiency, but I'll show screenshots of it not today.
Below is a pattern of Potentials for ONE symbol. The drawing of the graph without mathematics is attached.
Potentials are calculated reliably - tested on SaxoBank instruments (net client positions).
The essence of the strategy (medium term) is arbitrage of price differences and potentials.
The theory is that when the Ask potential increases - price should increase (tested fact), when the Bid potential increases - price should decrease. In addition to the difference, we should calculate the speed of change of the potential difference, the efficiency of the potential difference increment.
Also, the model of potentials, if we consider cycles, shows trends - both local intraday and global.
The picture for 29.07.2013-31.07.2013 on EURUSD.
I can post on any symbol for the period from October 2011 to the current date.
I should add that more interesting market information is available when analyzing major currencies (usd, eur, gbp, aud, jpy, nzd, chf), i.e. when calculating their potentials (multi-threaded analysis of 28 pairs) and tracking the dynamics of liquidity flow between them. I may post pictures, but not right now.
The work is carried out by directions (in Level II Spot):
1. Spectra (and related market insiders)
2. Potentials .
3. Flow patterns (liquidity between currencies)
4. Identification of qualitative parameters from Level II (currently 21 parameters)
5. Neural networks
I have encountered your blue screenshots and flowchart several times in the forum, I decided to use them until they were removed.
But experience showed, that compared to pure price analysis - connecting volumes from Level2 only noisy and worsens results.
I am giving you the report of real-time trading, you may challenge it if you have something, but not abstract images and stories.
I've seen your blue screenshots and flowchart in the forum several times already, I decided to use them until they were deleted, sorry if I infringed on the copyrights.
But experience has shown that compared to pure price analysis, including volumes from Level2 will only make noise and worsen the results.
Here's a real-time trading report, you can argue with it if you have something else, but not abstract images and stories.
It's not like anyone is arguing with that. See above: Williams' is not a fractal, but a candlestick pattern. Compare the Mandelbrot Set and the five candlesticks of Williams' "fractal".
Well, yes. Williams' fractal itself isn't worth much, so I guess it was a marketing ploy on his part to attract attention.
The fact that the market can be described by fractal geometry is, I think, obvious?
And if it can be described, then it is possible to predict to some extent. Or am I mistaken?
About 7 years ago I modeled "pictures" of Mandelbrot sets and the like, they obtained very similar to market quotes, if I may say so.
Where do you get your volumes from?
I've tried to use those that appeared in mt5, and of the third-party platforms, I've investigated dukascopy, clasterdelta, mbtrading to varying degrees.
If i've tried them, i've had no problems with them, but now i've decided to question ProstoTak, maybe he will prove that his fxopenapi is the best one, since i haven't checked it thoroughly and he found 100% of them.
I've tried those that appeared in mt5, while I've investigated dukascopy, clasterdelta, mbtrading from third-party platforms to varying degrees.
I've already started using them, but now I've decided to question ProstoTak, maybe he can prove that fxopenapi is the best one, since I haven't checked it thoroughly and it seems to be 100%.
You need to study the nature of volumes before jumping to conclusions... It's just as if he's PRing the system).
Study, that's the domain of science figures and students, and I only test their conclusions with experiments).
Study, that's for scientists and students, and I just test their conclusions with experiments).
What's the point of multiplying random variables? Experiments can randomly turn out to be resultant and this will be a subsequent factor to reduce the deposit).
You mean the algorithms for calculating Bid, Ask potentials from ProstoTak? So you can ask him what his simplex methods are, but I have everything in the dark - the algorithms are packed in the black boxes of neural networks).
Gentlemen, could you please tell me where I can get level2 data for analysis, searching for patterns, etc.? I am intrigued by this data)))
I was advised to download SoftFXQuotesDownloader from server tpdemo.fxopen.com data is divided into archives and divided into subdirectories with the average duration of half an hour. It is time consuming to glue them into one file manually, may be someone knows how to do it automatically?
I would also like to get data from another source, please advise perhaps other servers for QuotesDownloader, or other ways to obtain such data.
Thank you.
Gentlemen, could you please tell me where I can get level2 data for analysis, searching for patterns, etc.? I am intrigued by this data)))
I was advised to download SoftFXQuotesDownloader from server tpdemo.fxopen.com data is divided into archives and divided into subdirectories with the average duration of half an hour. It is rather time consuming to glue them into one file manually, may be someone knows how to do it automatically?
I would also like to get data from another source, please advise perhaps other servers for QuotesDownloader, or other ways to obtain such data.
Thank you.
Spread your lip, darling...
The topic of L2 history was brought up in a thread about HFT. As far as I remember it was suggested to learn C# for this purpose and write a parser.
On the other hand it is quite possible to get by and selectively glue manually from these archives into one table in a couple mio lines, it's quite fast and then spend more time on thinking about what in principle can be done with this data...
I'm sure that if you find some patterns in hourly L2 then 90% of these statistics will be correct on monthly and annual, assuming it's real data and not fabricated.
But for sure it is a very distorted, with bad intentions towards the client, flow and it will quickly drastically change the structure, which will prove to be a sham, because people can't all crowd drastically change their trading style.
IMHO about the throw-ins ProstoTak, he's obviously a Cossack (always glimpses his brokerage) with his 100%... it's bullshit. Even 80% is bullshit if you take a long sample.