Quote from peterfigliozzi:
It depends on what you are doing.
If you have discretionary entries, like I do, then having your head straight (being objective) is of utmost importance.
If you are executing a mechanical system, you have to be able to take the drawdowns and stick with the system, and resist changing things on the fly.
I like the concept of a trading model where you have mechanically triggered entries (and you must take each one), but you manually manage the exit. There is a basic profit objective and stop loss, but discretion can be used to take profits later than the objective or to take losses earlier than the stop. In other words, you use your discretion to cut losses short or let profits run.
This type of trading model psychologically has some benefits:
1) the entries are system / method generated and as such you may be less inclined to become attached to the position.
2) You employ your discretion to increasing profits and decreasing losses.
3) It is easy to benchmark your actual performance against the system/method's ideal performance when not using discretion to see if you are really improving or hurting the overall system results.
One of the downfalls is that there is a tendency to not want to take all the signals for one reason or other, particularly after a string of losses, which signals may turn out to be the more profitable ones.
You can also make a good case for discretionarily trading both exits and entries around a mechanical system's basic rules. Of course the shorter timeframe that the system uses, the more likely you are to "miss" entries and exits by waiting to execute at a more favorable price, but like the above method - this method can be easily benchmarked against the ideal system's results.
And then if you trade discretionarily, it comes down to the inner game much more because every decision is your decision, and breaking up (you and your decision) is hard to do.