"Averaging down" as a trading approach regularly causes controversy. While difference of opinions is always good, let's have a deeper look into it to make sure that opinions are informed and that we are talking about the same thing.
There is averaging down and averaging down. Not all of them are created equal. I'd break them down by two kinds.
1. A trader buys, position goes against him, he fails to cut his losses, sees them growing and getting out of hand. Eventually at some point he adds to his position following the logic "If I liked it at $20, it should be even better at $10" and/or "it can't go any lower". Both are false: anything can and often will go lower (no lack of examples of that over last year, eh?); and who is to say it was any good at $20 to begin with? And is $10 a better price or simply a proof that $20 was a mistake? This kind of averaging down is a "bad" one; it's done out of frustration, and it adds to a mistake. More often than not it increases eventual loss. In most cases what follows is: your position does recover some, by some magic stalling right under new breakeven level ($15 in our example). This gives you a chance to exit with a small loss but you don't take it - after all, recovery has started, you are looking at possibility of nice profits now (and on double size, no less). Sure enough, stock reverses and drops under $10 where you either exit with even bigger loss or put it in your long term portfolio, a.k.a. Grave of Short Term Trades Gone Bad. Another frequent scenario is, stock dives briefly under your second entry level, you sell your second position for a small loss, and that's where stock reverses and goes back to that 15... you curse your decision to cut losses on second part and don't sell first part - after all it's cutting the loss that killed your chance to get out even, right? Sure enough, it reverses down and you are looking at ever-increasing loss again.
Those rare instances when this strategy works only reinforce the idea of it being a viable approach, eventually provoking you to employ it again and again, until it leads you into a loss exceeding anything you saw in your worst nightmares.
2. Averaging down is a part of planned strategy. When a stock comes into your target zone but there is a lot of uncertainty in the markets, you don't feel confident enough to fully commit and don't want to stay on a sideline. You break your purchase in parts and plan a strategy for those parts. This strategy includes various scenarios of building up to full position in a case of further drop, in case of reversal, in case of stall. It also includes an "uncle point" - event or scenario which proves that the whole idea of entry was an error, so whatever is accumulated up to that point is being dumped. There is nothing's wrong with this kind of averaging down - it's done by a design, to minimize exposure at the uncertain time and increase it as events develop in a favorable way. I wouldn't even call averaging down but that's a matter of semantics.
As we see with many other things, there are no absolutes in trading. There is, however, need in clarity, in straigforward well-designed and thought through plan. Such plan, among other things, wil include definitions - as we mentioned earlier in this blog, "as you name the boat, so wil it float".