First off I'm not trading live, but I am in the final stages of developing an automated breakout strategy that I will begin to trade live soon.
My quick and dirty answer to your question is this. If you think discretionary trading is hard, evaluating and implementing automated trading strategies is much harder.
I think trying to evaluate tens or hundreds of EAs is a waste or time. If you are passionate (a prerequisite IMO) about automated trading then I would suggest learning basic C#, reviewing the source code of example trading strategies, and using a wizard provided by your trading platform to help get you started. You should be able to do all of these in a few weekends. If not then maybe its not for you. And thats just the first step, you'll need to avoid certain trading methods and techniques to eliminate overly curve fitting or vague ideas that are difficult to automate.
But don't expect to develop a profitable strategy in a few months. Its going to take a year or 3 of implementation and evaluation before you should feel confident in deploying your strategy with live money. It is very time consuming.