Valuations matter...until they don't
The key to making money in the market is to buy a stock when it's undervalued and to sell when it's overvalued. Or at least that is the blanket statement used by the so-called professionals.
As any blanket statement, it's a mixture of common truths and damned lies. First of all, one must define what is meant by "undervalued" and "overvalued". If you have a vague idea of "fair value", then what percentage above or below that number should be considered over- or undervalued? Taken to its logical extreme, this definition would suggest that a stock that's fairly valued at $100 must be bought at $99.99 and sold at $100.01, limiting your capital gains from the trade to 2 cents per share.
You would say, quite correctly, that nobody trades that way. True, but then how do you choose the upper and lower limits of your trading range? Can you say, for instance, "I will only buy it below $90 and sell above $110"?
The trouble with that is, stocks don't normally trade in such a range. Speaking practically, your stock is very likely to spend 17 years (the average duration of a secular bull market) far above $110 and then spend the next 17 years far below $90. Apply your trading rules too strictly, and one likely outcome is that you will buy your stock for $90 in year 1, hold it for 8-9 years as it lingers in the $30-40 range, hold it for another 7-8 years as climbs back to $90, then sell it for $110, and then wait another 17 years as it skyrockets to an astronomical height and then gradually comes back down to $90 - hardly a successful trade by any standard. That by itself wouldn't be so bad if you could just make your quick $20, say goodbye to that stock and start trading the next one which has just approached the target price. But what are you going to do if ALL of them trade synchronously, so that at any given moment you find the ENTIRE MARKET grossly over- or undervalued?
The reason investing is so difficult is that more often than not, rational analysis involving PEs and cash flows sets you up for failure. We had one enormous secular bull from 1983 to 2000, yet if you had followed your rational instincts you would have to sell in 1994-1995. We now know in hindsight that the correct answer was to hold until 2000, but not everyone could do that. For that, you would have to be crazy enough NOT to sell in 1996, NOT to sell in 1997, NOT to sell in 1998, and NOT to sell in 1999. Logically, it would make no sense whatsoever. You could be forgiven for not selling in 1995; perhaps, you could even be excused for holding through 1996; but by the time we get to 1999, your determination to hold defines you as a total lunatic. When they talk about fortunes made in the dot-com bubble, you have to see the irony in that prudent investors actually made very little money; those who struck it rich were the lunatic types who thought that a stock priced at PE of 80 was a hold.
Now, look 20 years back from that point, and you'll see the same story but in the inverse mode. Back then, you had to be a lunatic NOT to buy in 1972, 1973, 1974, 1975, 1976, 1977, 1977, 1978, 1979, 1980, and 1981. Only the most paranoid permabears sat tight until 1982, which we now know was the right time to enter the market. They must have been the Alstryans, talking about Alstrynomics, zombulators, 70% unemployment, and the coming end of the world. No other rational excuses would justify betting AGAINST the market that hadn't made any headway since 1969.
So, how crazy do you have to be in order to correctly identify the entry point or the exit point by ignoring all logic? And more importantly, do you need to be a nut to the second power in order to correctly identify the entry point AND the exit point? If you think the market can't go higher than 1100 or lower than 666, consider that the entire previous history suggests otherwise.