In general, I wouldn't. The only situation would be if you know you entered too early on a trendline break and you realize it , but your overall signals are still valid. Then you can take another position on the weaker retest (i.e. a right shoulder , double bottom etc). You will have to know when your trend signal has changed though, and get completely out.
I guess if I was trading someone else's money I would be more inclined to add to losers, and to hold on to winners and losers for a much longer timeframe. Since it's my own capital, I feel compelled to stay in the game and support my family.
It might be better to reverse a losing position as rule, who knows? There is no statistical test for randomness but to me its fairly plain that highly liquid markets are pretty random, in all but possibly the long term. If random then very difficult to trade successfully using any technical rules. There are no studies I know of which even confirm whether the 5-20% that seem to win in a given time period are consistent winners over the long haul. In short, most with any experience would say its a very tough game with no easy answers for winning. Not saying its impossible for one trader to be better than another.
the question doesn't make sense. each entry is its own trade, treat them as separate events with separate exits etc. you can't "add" to a position without a time machine, all you can do is establish another position in the same direction. the difference is not subtle. imo.
Many traders average their basis by adding to a losing position. I don't agree with it, neither do many, if any published trading experts. However, adding to a position has nothing to do with a time machine. Does your broker reflect your portfolio as separate positions for the same stock? No! It reflects one position. Adding to a position that is moving in your favor is commonly done. As far as separate entries and exits, with Cybertrader for example, you must specify whether you want positions closed as LIFO or FIFO. So in your strategy damir', while you might think that you are closing out the first trade, you might actually be closing out the second. Technically, you are accurate.... except for the time machine thing.
I suppose if you use some support level as a stop loss, your maximum % loss becomes reduced as the price approaches that level. That could be one reason to add to a losing position, but only if you are absolutely sure you would cut losses at that level.
a position is a combination of entry/exit/etc. unless you have a time machine you cannot replicate the same conditions. it isn't "adding" to a position, it is a separate position that happens to share a ticker symbol and a direction. this isn't nitpicking. understanding the distinction - really understanding in a shake-your-bones kind of way - is an incredibly eye-opening experience.
Bolts the max % would be reduced yes, perhaps making the trader feel better, but the dollars lost... which when paying the mortgage is all the matters .... is increased. damir' I don't disagree at all with your statement that each trade should be taken on its own merits. You are exactly right imo. But the common perception is that one is adding to the original trade and averaging the basis for the sole purpose of reducing the breakeven point. Again though, placing each trade on its own merits is the right way to go.. as you said.
I think if I were playing with a few million dollars, averaging in would be the best way to trade. -FastTrader