|
PureBytes Links
Trading Reference Links
|
At 06:43 PM 3/24/98 -0000, Bill Eykyn wrote:
>I cannot see how a large time frame indicator is going to help you on the
>intraday trading. And not to take a trade because the intraday movement is
>in the opposite direction of the daily indicator direction, doesn't stack up
>for me. The market may well be going up and the indicator showing this -
>but current swing could well be in the opposite direction.
>
HERE HERE! I heartily agree. I believe that much of the "lore" around
trading exists because it sounds so believable in the "learning to trade
101" books that abound. "Only go long intraday if the weekly trend is
up" - sure, that sounds reasonable. However, when you are looking at
1 minute, or 5 minute bars, and trying to capture 10% of the daily
range per trade, the weekly trend is pretty irrelevant. I will agree
that if you are entering the market randomly, you have a better chance
to succeed in a long position on an up day. BUT, you can also get
killed taking a long position on an up day (even one occurring in an up
week) if you get in just before a correction.
>I only wish I could find an indicator which told me direction in a timely
>and honest fashion, just before and not just after, happened. If you are
>position trading over several weeks or months, I am sure it is another
>matter, but in time frames of minutes and quarters of an hour, for me (and
>my eyesight!) the indicators all lag - and lagging is no use to me.
>
I strongly doubt that such an indicator exists. Here's my non-technical
explanation as to why:
We encounter physical systems in everyday life, and these shape the
way we (erroneously) view the market. By physical systems, I mean
things that obey Newton's - or at least Mandelbrot's - laws. Systems
like the weather - sure there is some uncertainty, but you can pretty
much count on a day in July being warmer than a day in January. You
have a higher liklihood of rain tomorrow if today is cloudy. Or
mechanical systems - broadside a Kia sub-compact with your 4-ton GMC
Suburban, and the odds are 48:1 that you will walk away, and the Kia
driver won't.
Unfortunately, we want to believe that the markets operate in a
similarly deterministic manner - that if the 5 period moving average
crosses above the 21 period m.a., signalling that the market HAS
moved upward, that it is going to continue to move upward. I would
postulate that if the market does indeed continue its move it is
more based on coincidence than on the predictive powers of the 5
and the 21 bar moving averages. The market just isn't a physical
system - oh maybe it sometimes has some of the properties of one -
I'm not saying that a clever, experienced pattern reader can't see
familiar set-ups, and trade them successfully more than half the
time.
What I am saying is that any indicator is just going to tell you
what the market has done. And, it's going to lag somewhat, so in
fact, it's telling you what the market was doing a few bars ago.
However, this provides no guarantees on what the market will do
next. Expressed differently, if Greenspan has a bad day, and
suddenly announces a rate hike, the market's reaction will not be
influenced by how recently the 5 bar m.a. crossed the 21 bar m.a.!
Well, I hope this sparks some discussion. I'm not going so far as
to say that technical analysis doesn't work. I'm just saying that
I doubt that anything that can be reduced into Easy Language can
predict something as complex and uncertain as the market's behavior.
Jay Mackro
|