markets are efficient is poor. He argued that if I could develop a winning system on a computer, so could others, and
we would all cancel each other out.
What is wrong with that argument?
Because people develop systems and people will make mistakes. Some will alter their system or jump from
system to system as each one has a losing period. Others will be unable to resist second-guessing the trading signals.
Whenever I go to a money management conference and sit down with a group to have some drinks at night, I always
hear the same story. "My system worked great, but I just didn't take the gold trade, and that would have been my
biggest winner."
There is a very important message here: People don't change. That is why this whole game works. In 1637,
tulips in Holland traded for 5,500 florins and then crashed to 50, a 99 percent loss. Well, you might say, "Trading was
relatively new then; these people were primitive; capitalism was still in its infancy. Today we are much more sophisti-
cated." So you go to 1929 and find a stock like Air Reduction which traded at a high of $233 and after the crash fell to
$31, a decline of 87 percent. OK, you might say, "The Roaring '20s were crazy times, but now things are surely
different." Move ahead to 1961 and you find a stock called Texas Instruments trading at $207. It eventually dropped
to $49, a decline of 77 percent. If you think we have gotten more sophisticated in the 1980s, all you have to do is
look at silver prices, which in 1980 reached a peak of $50 and subsequently fell to $5, a 90 percent decline.
The point is that because people are the same, if you use sufficiently rigorous methods to avoid hindsight, you
can test a system and see how it would have done in the past and get a fairly good idea of how that system will
perform in the future. That is our edge.
Do'stlaringiz bilan baham: