What makes a good EA? In the opinion of this writer, it is the following.
An EA needs three legs and it needs all three legs all the time. Here is a description of each leg.
Leg 1. A reason for entering a trade.
Leg 2. A definition of what price action will tell you are wrong
Leg 3. A definition of what price action represents a likely end of move.
Lets measure most EAs against this list. The reason to enter is often weak, but even weaker is the assumption that the same StopLoss amount and the same TakeProfit amount will work time after time. And how many EAs are grinding away with the optimizer searching for the 'right' StopLoss and TakeProfit levels.
The EA must contain the definitions, let the computer apply the definitions and develop the StopLoss and the TakeProfit for every trade. It will always be different. Whe would bet me that the same answer is the best answer or even a good answer?
How to test for this? Look at a chart of your last three trades. Is that where you would have entered manually? And after the position was on, was that the best place to place the Stop and the TP? Not likely. Write down why and where you would have placed them and that is the start of your definitions. And you are now on your way to a better EA
Each EA has it's own personality..
they don't even need any good reason to enter a trade because reason is relative, it's depend on how you trade..(EA Logic to enter trade)
The most important part of an EA is not only the logic, but also money management or risk management..
My point is how an EA minimize the risk when it's red and how they maximize profit when it's on the green condition
I think the weakness of an EA is... it only use technical analyze to trade, without any fundamental analyze..
I think that's the strenght of an EA. My ideal EA doesn't know what's going to happen and doesn't care either. I don't know what's gonna happen, so why should my EA?