Disliked{quote} You don't understand how robots work , I will explain briefly because each robot is coded with precise accurate execution. As an example I have two different systems , I have 75 of them. One buys at support and one buy a price action set up , the support EA has to be coded to buy at support , the support trade many not materialize for weeks or months depending on time frames and related moving averages + other conditions , but every few months I get a trade on the exact precise logic , same applies to price action + related entry conditions...Ignored
So, what's the correlation factor between them?
Mainly because if 2 EA are executing similar strategies, it's 1 huge chunk of risk, it doesn't really achieve proper diversification.
Because you really only need 2 EA to perform well, 1 trend following and 1 mean reverting, and adding a customized money management system according to runs of winners and resetting when hitting a losing trade. Something like a highly moderated version of Kelly criterion, but if it's used from the form of a logarithmic spiral of 0.1% factor increment in risk factor per successful trade tied to positive risk reward ratio.
So basically winner systems are allocated more money with each successful trade perhaps maybe if starting at 0.1lot then it becomes 0.01 capped at 0.15, then when it lose 1 trade it resets back to 0.10. so maybe the detection of runs can start from 2 or 3 trades successfully done.
Which is a play on the fact that markets have a tendency to persist on certain condition then switch. So while they are persisting in 1 condition, 1 system is making money, another is losing, the winner by winning gets more money, loser by losing holds the same.
You can cause the system to have a drift of profits from money management strategies.
By being able to negate correlation or even go opposite, you can diversify risk enough that correlation risk is not an issue. Then next is to figure out if both are long term profitable.