SeriousTrade, I was right where you're at like less than 2 months ago. As they say, everything works until it doesn't. So what I've found is, having a fair knowledge of price action, I had to create a simple strategy that worked (until it didn't). That's where I'm at now. Entry, exit, scratch, filters to keep me from trading. Every day I take notes on all trades, and then after seeing what trades work more than not, I adjust and improve my strategy. Continual refinement will hopefully improve results.
I am (day-) trading most only my own automated systems, so I can spend all my time doing trading R&D ... I concur that most setup ideas fare no better than random entries., and those (relatively rare) which are better than random, remain subject to significant variance in win% I do believe that discretionary trading can be better than automated trading, but it can also be worse ... let me explain: - if one defines a pretty rigid trading plan, rigid to the point it could be automated, and then trades it manually, then the results are going to be worse than what the automated system would do ... 2 reasons for this, 1st is without automation, there is no possibility to do enough backtesting to validate every aspect of the system, hence some rules will actually be detrimental, unknown to the trader, 2nd reason is manual execution is inferior to automated execution (from taking a break to judgement errors, through fat fingers & what not) - to the contrary, if the trading plan is more guidelines than rules, and the trader remains immersed in price action, then he has a lot of room to trade what he sees happening now, and with enough screen time & practice this will work more often than not - but not for an automated system.
75% of my signals are counter-trend in ES cause this is what this market does the most, reverse. Whereas in other markets like currencies, Russell and Crude Oil is more likely to trend. You have to know the market you trade well, watch what happens when other markets open and close, there are relationships, often times of late the ES and Crude Oil trade in same direction. Or watch all the Indexes to find out which is the strongest/weakest. Reports are often initially wrong, find the divergence and take a scalp of a couple points. Some markets blast thru S/R only to retrace a few minutes later to take out the protective stops of those who took the breakout, then goes back in same direction-get in on 2nd attempt is a great entry and when little traders jump back in, they push market even further. Trading is usually the same overall, find where the little trader is going to get screwed and trade against them. But the more you trade, the more experience you have, eventually you notice where you get in is the least part of the method and needed many more money management rules for the profits.
I agree with this. I haven't been daytrading long, but after my first 6 months of trading one very specific rules-based plan manually, I compared my Ws and Ls to an automated backtest of my trade plan over the same 6-month timeframe. After 6 months, 47.7% of my 606 manual trades were winners. Compare that to 44.1% winners for backtest of the strategy. When I get my signal, I don't usually enter automatically, although sometimes I do if the timing looks right. I usually wait for a pullback, which usually comes, but sometimes it doesn't. If I miss it, oh well, just wait for the next trade. It's a trader's patience that can make all the difference vs just trading an automated program.
This is how I tend to frame my trades. Do I have too much company on my side of the trade? What is the makeup of the people on my side of the trade? Is it dumb money? What possible exogenous events have not been priced in by the financial models that I can make a profit? Everybody is using their own models to forecast. I speculate they probably are spread out in a normal distribution in terms of signals trade. Thus I find it best to be contrarian in nature. To see things just before the herd does.