We are open to sharing the system with anyone willing to pay the price. However, we've spent years developing this system, and at this point, we don't even want to think where we'd be if we hadn't. It is our prized asset and our financial future. So, we are pretty hesitant to give it away. But, if someone is willing to pay $150,000 for everything we have, then, maybe we might consider it. Currently, we have a deal on Quora for a fraction of that price. Stay updated on our profile for further details.
Would you collaborate with a programmer if it meant sharing your system/research?
- Thread starter Laissez Faire
- Start date
Quickly learn to program with enough knowledge and or expertice is BS. It takes a lot of time.
I don't.
Time comes at a premium these days. If my life situation was different, I'd probably put everything else on hold and go all in to learn to code, but I have a hunch I won't get anywhere fast with the current time available. There's an alternative cost for sure.
I've done some basic programming in MATLAB and it didn't come naturally to me. A lot of work just to do some basic charting and solving of differential equations. I've also done some brief stuff in Python many years ago now.
Based on that experience it doesn't seem like something I'll learn very quickly. Of course, I only dabbled with code and it's a different matter if I went all in and had proper guidance.
In relation to the solution discussed in this thread: If a professional programmer says he needs 3 and even 4 weeks of full time work (maybe it's an exaggerated quote) - I wonder how long it will take me to learn the necessary code AND get the work done myself. I imagine it will be years...
No, you're not crazy and it is a controversial matter where people usually are in disagreement.
Let's be clear, though. If you want to make a prediction, you need to base it on data. Looking at a chart and saying "This looks bullish" is not a prediction. A lot of the time when people express their opinion about where the market is going it's just based on hunches or what they THINK will happen.
I spent many years just assembling the data I think are relevant for making predictions. And even then - I would have to understand how to best use that data. Like baking a cake. I'm at a point where I'm still solving this, but certainly arrived at some interesting answers and have a good base understanding of how the markets generally move.
There's also a misconception that once you make a prediction you'll stick to that no matter what. Most certainly not. If you're driving in full speed on a road and your GPS/chart says it's a clear road for miles ahead you'll be a fool to not stop if you now see that the road is blocked or there's a turn that's not on the map.
A good trading model whether it's predictive or not is updated as more information is known. For example, you may have an idea about what happens tomorrow based on what happened today. By tomorrow, the market is opening up or down and that's new information. After 2 hours of trading "THIS" happened and you now know more. When there's one hour left of the day only so much can happen MOST OF THE TIME.
If you study the markets in depth you'll see that it's basically the same days repeating over and over again with a rough variation. I firmly believe that there's nothing that will happen tomorrow that didn't already happen in the past. Of course, some of the days are less common than others.
Now, even if we all agree and acknowledge that the same patterns keeps repeating over and over again it's still not clear what can be predicted or not.
This is a bit off-topic, though, so maybe better to move it to a separate thread.
Well, I don't agree with that, although I generally look at the most recent data which I agree is most important for sure. I do like to have a good historical database for reference. Particularly since I also look at weekly and monthly data. A 3 year period may contain only a certain set of market conditions. Also keep in mind that I trade stock indexes and not single stocks.
If your data set only contains a low-volatility grinding period you'll be in the dark when volatility spikes and we're entering a different regime.
But I suppose it all depends on how your trading model works. Clearly we have different approaches.
I'm in/have been/rotating around *exactly* the same kind of dilemma as you for years.
- First I ended up farming out code to different devs and bringing it together myself which works, but is an utter sh*t show when you want to make changes and you're back in line of their other clients work
- Second I looked at coding myself, and it's a non-starter. I simply don't have time to learn to the level of proficiency I need
- Thirdly, I looked at the joining up with a dev partner as you're doing, but even with an NDA, the chances are if it's going to used elsewhere it'll happen without your knowledge and you'll be none the wiser.
This might be worth a look, i only came across it this am so just investigating myself :
https://www.1station.app/
- First I ended up farming out code to different devs and bringing it together myself which works, but is an utter sh*t show when you want to make changes and you're back in line of their other clients work
- Second I looked at coding myself, and it's a non-starter. I simply don't have time to learn to the level of proficiency I need
- Thirdly, I looked at the joining up with a dev partner as you're doing, but even with an NDA, the chances are if it's going to used elsewhere it'll happen without your knowledge and you'll be none the wiser.
This might be worth a look, i only came across it this am so just investigating myself :
https://www.1station.app/
Well, I've been spending my whole life building a trading system trying to decipher HFT bots (i.e. big money, smart money or whatever you call and you guys can just laugh after reading this.)
It's nearly impossible to do so, but I managed to get some lucks in recent years. Now, I'm reaping the rewards from that success with sky is the limit.
I don't see how I'll share anything openly on here to everybody, but I've been thinking of giving away my trading system to one or two people whom I can trust. A lot of things about my trading system that I still don't understand, especially swing and position trading styles since I only concentrate on scalping and day trading presently. Having additional brains analyzing my trading system might benefit us all, but so far I can't find any trustworthy person yet lol.
It all comes down to how much in depth you understand PA of the instrument that you trade. Making money is easy after that.
Unless one has very profitable setups working in all kind of market conditions, automate your setups will waste a lot of your time since market condition can change in any minute, especially sideways. Also, when you're dealing with volatile futures instruments, you still have to constantly adjust your position size to ensure you don't expose to too much risks and etc....
Besides, if someone spends a lot of time trying to help me like doing coding stuff, I won't feel good to cheat on him and etc... So, I definitely will reject his offer to help me out.
Anyway, good luck.
Maybe instead of giving it away, you can invite some of us to watch you trade it live?
I would be perfectly happy with live calls for a week. Your results are stellar, but the posting frequency is low. What would a typical week look like for u given all the different types of days and times for an entire week of trading?The approach I would take is:
1. specify what you want him to reconfigure/develop/build. You will need to be able to do this whether you code it yourself, use the programmer that you're considering, or use another.
2. optionally, split programming the different pieces between different programmers. This doesn't sound like a good choice in this specific instance though
3. retain the key variables which are likely to comprise your IP/secret sauce/edge for yourself.
e.g. if using a moving average crossover, specify that after moving average (x periods) crosses above/below moving average (y periods) then do something. Specify that x and y are variables (that can obviously be varied). Test it using variations of x and y. When happy that the programmer has delivered what you have specified, run it with x and y using your specific backtested variables.
1. specify what you want him to reconfigure/develop/build. You will need to be able to do this whether you code it yourself, use the programmer that you're considering, or use another.
2. optionally, split programming the different pieces between different programmers. This doesn't sound like a good choice in this specific instance though
3. retain the key variables which are likely to comprise your IP/secret sauce/edge for yourself.
e.g. if using a moving average crossover, specify that after moving average (x periods) crosses above/below moving average (y periods) then do something. Specify that x and y are variables (that can obviously be varied). Test it using variations of x and y. When happy that the programmer has delivered what you have specified, run it with x and y using your specific backtested variables.