^misantropia^ wrote:You picked a bad example. S&P 500 is a by-committee index and not necessarily neutral in the stocks they pick (and companies on the list profit from the 'follower funds' effect). Rather, look at the US stock market as a whole and tell me what ROI to expect.
Ah, but selling at a loss is not really a sustainable long-term strategy, is it now? /cheek
But seriously, is there ever a time when it's a good idea to cut your losses and walk away? This is a hotly debated topic among the traders I know (I hesitate to call them friends, traders don't have friends). The 'hang on for dear life and hope for better times' attitude is most common but it's by no means prevailing. I'm on the fence, convince me!
Yeah, those things are true, plus it probably suffers from the bias that companies will be excluded before they go bankrupt. Then again.. so what? You could buy an index fund that tracks it and make the same return, plus the index is highly diversified across companies and markets (clients). I see no issues with using it as a benchmark, but you're welcome to provide me with total return indexes (means incl reinvested dividends) that have performed much worse (and please no cherry picking a la Japan).
I find that remark by traders a bit unusual at the very least. All the succesful, disciplined traders that I know employ stop-loss orders: when things aren't going their way, the position gets closed automatically because of the risk that things will get even worse. History is filled with traders bankrupting themselves or even banks trying to make up the loss by fighting against the market and not getting themselves out of the position. When considering selling, you should only consider the prospects of the company compared to the price NOW, and not get caught up in your purchase price through anchoring bias.
It's a poor analogy on many levels, but think about it this way: if you flip a coin 5 times in a row, does the fact that it came heads every time change the probability that the next time it will come tails? Similarly, purchase price has nothing to do with how much something costs now or will cost in the future, past performance is not a good measure of future performance, etc.
