My program project (patterns give prediction).
Hello guys,
I am posting on this forum to talk about a project I’ve been working on. I’ll try to be short and concise.
I’ve been thinking of a solution to anticipate in the best way possible the movement of the market. As a lot of you probably have, I have tried several different combinations of indicators, I have tried signals services, I have tried applying some strategies of my own… In the end, my ITM ratio is never satisfying enough to me.
I don’t want to be a big trader. My trading strategy is placing maximum 5 big takes a day, take the money and call it a day. I don’t spend my life trading, I am not interested in doing so. With that in mind, I need a simple yet accurate predictive algorithm to help me place 5 trades tops.
The solution I have thought of is based on patterns. At each new candle on a timeframe, the algorithm looks a couple candles back (say 4 to 10), saves the pattern, and searches for all the occurrences of this pattern for a period back in time (say a year). It looks up how many times this pattern was followed by a red candle, and a green candle. Then it compiles the statistics.
To make it simple to understand, i worked on what the graphic interface of the software could look like.
The image is low def and watermarked by me to avoid copying. As you probably guessed the screenshot is just a demo, the numbers are not real, it is just to show what the software could look like.
I have a background in coding and I am able to code such a tool, I want to make it great, I am really excited about this. But this requires a lot of work, and I want to make sure people are interested before I start working on it for good.
The software would work with a membership plan. There would be different status. Normal, Gold, Platinum, Ultra…
Here is a non exhaustive list of the options I have thought of for the program, that would be included depending on the membership plan:
- Number of pairs available
- Number of pairs displayed at the same time
- Loopback period. The bigger the period, the more samples, the more accurate the prediction
- Number of timeframes available
- Possibility to link real or demo trading account to copy trades
- Multiple timeframe signals for one pair on the same screen
- Load .csv databases for extra analysis
- Personal assistance
- Graphic interface style
- Running time of the software
…
These are the options I can think at the top of my head. Exciting huh?
The plus factor of the program+algorithm is the graphic interface, with all the extra options it brings to the table. It is simple and direct to understand, and can be tweaked easily.
The software doesn’t exist yet, I just wanted to get any feedback here on this forum to know what you guys think.
I don’t know yet how much the prices for the membership plan would be, it will depend on how many people are interested. I do want to keep the membership costs as low as possible. I don't want to be part of all these programs that are supposed to make you a lot of money but cost 300 and end up not working. I want my software to be efficient and cheap, with a lot of users, a community around it. This sure requires a looooot of work for me, as I work alone. I am also open to collaborations with designers/coders on this. The subscription could be cancelled any time, no subscription of minimum 3 months required or anything like that. You don't want it anymore you just stop paying 
Anyways, this is a quick preview/draft of the project, waiting for feedback and see who is interested!
Bye!