Do automated strategies that can generate consistent profits really exist? worth?

Do automated strategies that can generate consistent profits really exist? value ?

Some say it doesn't exist,//also correct. After all, 99.9999% of strategies are losing money and are rubbish. Automation will only accelerate the loss of money.

Some say if it existed it would be an ATM machine, worth a fortune.//also correct.

If so, then there will be a logical paradox here.
Why would someone with this strategy look for a partner? or Is it possible that someone wants to buy it at a high price?
From the perspective of unbalanced resource allocation, it is possible. If someone with this strategy does not look for a partner for selling or money, such as other private things or resources that he does not have,

Is it possible that someone wants to buy it? Theoretically, it is impossible, unless he is a master who develops reverse strategies, that is, specializes in studying the commonality of loss strategies


So, if you are not God, please don’t be 100% sure about this, because the market is much broader than our knowledge.

 

1. Nothing works forever. If you have limited capital then by the time you scale it up, it will stop working. Better to have a big lump sum ahead of time. The conundrum is that we don't know when it stops working.
2. Automated strategies that work currently and into the foreseeable future exist. They are effectively ATMs for the owners. From my own experience, every niche gets more crowded over time as others discover your "edge" or whatever you want to call it. Eventually the edge gets so worn out that it doesn't exist. Bigger players might still be using it by having access to better technology or better facilities (ex. more shares to borrow, exotic instruments).

 

This is a profound topic.
The "effectiveness" of a strategy must be defined. If it is a tick strategy for intraday trading, it can make money in half a year. There are 500,000 automated strategies in the world. There are 1,000 strategies that can make money in 1 year. There are 100 strategies that can make money in 5 years. There are 5 strategies that can make money in 10 years.
If it is a 1-minute cycle strategy, professionals usually require 30 years of data to be called "effective".

2. If it is a high-frequency strategy that relies on the advancement of computer hardware technology, as you said, there is a "bank run" phenomenon in the market. If it is based on the correctness of logic, it is generally impossible to make money now if the logic is correct, and suffer losses in the future if the logic is correct.
3. This is indeed a problem for large funds, and I have not found a solution yet. Just like the "market projection" phenomenon mentioned by Soros, your participation itself is the influencer of the market, and it will have a reverse impact.

 

B A C K T E S T is your answer.

 

It is not that simple.
Backtesting may "cater" to the market.

So professionals will propose "double-blind" testing to prevent this.
Blind selection of time periods ；
Blind selection main contracts.
This harsh destructive test must be carried out.
This is just the first step...

 

I love seeing stuff from someone who knows it all. All 99.9999% of it.

 

Where do all these numbers come from and why do you believe them?

 

Short answer, yes. But there's alpha decay which is the consistent part

 

Hello Amos.davis,

I believe these type of algos do indeed exist in the markets and they are certainly not for sale to the public.

I have no proof, but I am unable to create an algo that consistently make money month to month or year to year GURANTEED.

 

I have an automated strategy that makes money 100% of cases. And it is historically proven strategy, it has been profitable for years.