value of non-discretionary trading
this is my set of beliefs as far as the relationship between macro events and asset prices (stock prices):
1. Predicting macro events is impossible. Majority of economists and experts in this field are unable to do so. Majority of great forecasters are close to 50% so not very far from random. In the middle of 2008 we were told that subprime did not matter, at the beginning of 2009 that the future is dire. Since really smart people can't predict the macro i don't think i can do it.
2. Direction of asset prices can not be predicted. Also the relationship between stock prices and real economy is not stable - sometimes the market leads the real economy by many months sometimes it does not. Even the best traders/investors can not predict these relationships (even Mr.Morales quoted in this thread had substantial losses in 2009 as he was not well positioned at the beginning of the bull move). Its just really, really hard to do it consistently.
3. One can try to evaluate the market in probabilities (similar to weather i guess - we don't know if it will rain but when we see heavy clouds it does make sense to take an umbrella). Mechanical system might be a great help in assessing probabilities. "Robot" seems to me an excellent system (in fact one may even argue its Sharpe ratio will have to decline in the future it's so incredibly good). It is important to remember that losses are inevitable and can not be eliminated from the system (if there is a system without the losses or with very few of them it usually means it was curve fitted). What matters most is how large the losses are in comparison with the gains. Percentage of wins to losses is secondary (it does, unfortunately affect the psychology). Since i can't control the economy or predict prices i try to focus on what i can do. Proper position sizing, minimizing loss size and not overtrading.