Creating a trading robot - page 15
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
That's right (let's keep it on a first name basis).
Just look at Peter Konov - how many people help him?
I needed help in testing individual TCs from my League - only two people responded.
Now - I'm putting League TS in Shared Projects - anyone who joined will have in the tester all 672 current TS, and for the promise to open a free signal - he will be open to work for a demo or real. Do you think many people will be interested? The maximum - they will install, run a couple of TS in the tester, and that's all.
Everybody wants to know less and do something, but at the same time - get more results. Alas, it will not work...
It's not even out of thirst for freebies.
An arbitrary group is incapable of creating a product, or the likeness of a product, simply because there is no pressing need for it.
It is unlikely to go any further than talk because of the blurring of purpose, disagreement on all issues and loss of interest.
All such "shall I sing a song..." style threads
Can look at any open/free projects - there is a core of 1-2 maximum 3 people who strictly need the project, there are committers who use it from time to time and have the qualifications + patience to patch and 99% of all others.
So successful projects are the result of activities of individuals or tight groups. And only then can they grow
That is, successful projects are the result of individuals or tight groups. And only then can they grow
I agree. I read a good article on this very recently on Habra...
Well... let's see, maybe here - something will be different...
I agree. I read a good article on this very recently on Habra...
Well... let's see, maybe here - something will be different...
It won't. To sell/use the product, taking into account the interests of all participants, you'd have to register it and put in add-ons/server part that prevents any participant from leveraging it past the team. Just for the formalities will have to spend $ 5-10k beforehand. Although for 1000$ you can freelance order a lot of stuff from several people, or buy ready net-worker, and put rest in trading account with minimal risk. One example:
It won't. To sell/use the product, taking into account the interests of all participants, you will have to register it and put in the add-ons/server part that prevents any participant from leveraging it past the team. Just for the formalities will have to spend $ 5-10k beforehand. Although for 1000$ you can freelance order a lot of stuff from several people, or buy ready net-worker, and put rest in trading account with minimal risk. One example:
Here came the idea that it is possible to make a decent Martin. It is enough for the robot not to make more than 1-2 errors in a row. Quite an achievable task in my opinion.
It may be realisable, but it is not even a simple task. But if a solution was found, making the robot profitable by manipulating the lot is easy...
Not gonna happen. To sell/use the product, taking into account the interests of all participants, you will have to register it and put in add-ons/server part to prevent any member from leveraging it past the team. Just for the formalities will have to spend $ 5-10k beforehand. Although for 1000$ you can freelance order a lot of stuff from several people, or buy ready net-worker, and put rest in trading account with minimal risk. One example:
Thousands of quid?
What is there to spend thousands of quid on? You'd think someone would have really working ideas... In fact, everything has already been thought of a long time ago, and I'm sure no one in this forum will come up with anything more clever. All the techniques are freely available. The main thing is to apply the appropriate one in time. And it will be normal to get a few percent a month - which is quite good.
So here's an idea that it's possible to make a decent Martin. It is enough for the robot not to make more than 1-2 mistakes in a row. Quite an achievable task in my opinion.
This is NOT a solvable problem. You need"reliably no more than 2 errors in a row" - and no one will tell you the future.
I have many experts in the League who have only shown 1 consecutive SL on a two year history. However, they tend to have a very small TP and an impressive SL. And on the second SL in a row, when it is obvious that the system has failed, the slippage is already quite noticeable. The system shows a "Martin" type of curve without any martin. If we add Martin to it, the slippage will be huge on the second SL.
Is it worth it?
This is NOT a solvable problem. You need"reliably no more than 2 errors in a row" - and no one will tell you the future.
I have many experts in the League who have only shown 1 consecutive SL on a two year history. However, they tend to have a very small TP and an impressive SL. And on the second SL in a row, when it is obvious that the system has failed, the slippage is already quite noticeable. The system shows a "Martin" type of curve without any martin. If we add some Martin to it, the slippage will be huge on the second SL.
But is it worth it?
Overlaying the inverted, MathAbs(100 - RSI(14)) on the 'correct' RSI(14) is an option for the strategy. No?
It means SL should not exceed 100 points, and TP > 100.
Then 1 SL in a row doesn't come out.
It has long been known, and my League confirms it - on a trend TP/SL Ratio should be greater than zero. The classic trend is considered 3 (although, my experience puts it usually at 1.5-2). At 30% trend guessing. You see, with such a low number of guesses there is no way to get few stops in a row.
For a flat on the contrary, less than zero. Classic flat is 1/3 (in my experience 0.6-0.2). When guessing 80% of the flat. And here - you can easily have long periods without stops, and only 1 stop in a row. However, this stop will be very large. And if the system fails, the drawdown will be very noticeable.