|
PureBytes Links
Trading Reference Links
|
|
Superfragalist,
Best post I've read in five years....maybe
ever. I endorse every word. Everyone should read this once a
week. One small point: it's unfair to just hang these trading
misconceptions on newbies. I know, and share research with, a bunch of
engineers that are some of the biggest offenders of the most obvious system
development sins. Poor babies, they always want to hang "just one more
variable" on an approach. Will they ever learn?
Thanks again....
Steve
----- Original Message -----
Sent: Friday, January 21, 2005 3:44
PM
Subject: [Metastockusers] Free Code for
the best possible entry
Someone asked about a system recently and said it
seemed to work well
when the markets were trending upward, but not so good
when they
weren't.
No surprise there. Most newbie's don't
understand how the
trend really works, and how it impacts systems, both
for trading and
for development, so I thought I would share a little
experience and
opinion.
I'm very sure there are some of you who
won't believe what I
write, and that's fine. You're free to do things
anyway you
want. After all it's your money.
I don't want to write a
book in order to educate the one or two
people who will read this so I'm
going to be as brief as I can
and then you'll have to figure out the rest
based on what I say in
here. Also, I'm not going to put any lace on what I
have to say. I
usually don't anyway, but from now on I'm going to post a
disclaimer
saying you need to pay attention to information and not the
style. If
you are easily irritated by the style stop reading now, and
don't
waste your time (or mine ) with posts or emails about being
politically correct. So here we go.
Newbie's are very
concerned about finding a system that works
across a long time frame, etc.
They are especially concerned about
entries. They want to hit the entry as
close to the turning point of
a trend as possible. Both of those issues
are a complete waste of
time. The reality of it is when a market is
trending either up or
down, it really doesn't matter much what you are
using for
entries as long as you are entering in the direction of the
trend.
(Yes, this is true. Later I'll give you some more facts you
can
research on this.)
A few simple trend indicators are all you
need to see when to enter a
trade. Entries during trends require almost no
brain cells. You can
see this from the nineties when everybody was a
wizard trader, and
then got wiped out in 2000 on. They just didn't pay
attention to
what I'm explaining to you in this post.
Once we have
our entry, the problem becomes one of exit strategy and
money management.
You should develop an exit strategy that works separately
from the
entry indicator. If you don't know how to do this, you
can read many good
systems development books by Chande, LeBeau and
Kaufman. They discuss
exits in great detail. They will also educate
you on money management.
Exit strategies are too numerous to discuss in a post.
However,
there is not such thing as the BEST exit. There are exits
that work for
your style of trading.
To start the process of systems development,
throw a few
trend indicators on template and look at some charts. The do a
few
tests WHEN THE MARKET IS IN AN UPTREND. Most newbie's
test
everything across a broad range of all kinds of market conditions.
That's doing things the hard way. They've heard this is a good way to
test systems. Wrong! The only people who need to test a system that
way are the people who are trying to get you to pay $3000 to $10,000
for one. They want it work at least a little when the market is not
cooperating so they don't have hear a lot of complaints. (If you
want
to read an interesting history of trading systems and the people
who sold
them, Bruce Babcock has a history section in his book The
(Dow Jones)
Irwin Guide to Trading Systems. After you read that you
want ever consider
buying one. It's got all your old favorites in
there Gann, Wilder,
etc)
Since you're an individual trader and you've studied the
systems development books I've recommended, educated your mind
and
finally understand how trading actually works, you can change
what you are
doing according to market conditions. When that's the
case, there is an
easier and more profitable way.
First figure out the trend. There are
several ways to do this
and they're all easy. On a weekly chart you can
use a 10 and 40
week moving average. When the index of choice is above the
10 week
moving average the market is in an uptrend. When the index is in
between the 10 and 40 week moving average, it's whipping
around,
consolidating, and screwing you out of money. That's what it's
doing
in there. When it's below the 40 week moving average it's in a down
trend. That's the easiest way to define things.
You can also look
at some set of moving averages on a daily
basis, like the 4, 9, and 18
triple MA or something else close like a
5 and 20 EMA or SMA. It doesn't
matter that much. Pick one.
The dailies will tell you when a correction
is taking place
in an up trend or down trend market. If the long term
market bias is
up, but there's a correction going on, you will want to
take
shorts only and expect to hold them for a very limited period of
time. This means use tight stops and don't let things run just to see
where they're going to go. You only do that when the daily trend
agrees with the long term trend.
Okay, now we know how to determine
the trend, what comes
next? As I said, stick a couple of trend indicators
on a
chart—your choice of which ones—they all work about the same. There's
no BEST one.
See how they look on a chart with a few
symbols---during a
period of time WHEN THE MARKET IS IN A DEFINITIVE UP
TREND. Then put
them in your systems tester and test them on a bunch
of stocks like
the S&P 500. Use the same trend indicator formula for
entries and
exits—reverse the entry please—don't worry about exits
at
first. If the trend indicator returns a reasonable amount of money,
has a rational number of trades, etc then you're in business.
Next
develop your exits. Exits need to fit your personality
more than entries.
What I mean by that is too many people look at
what makes the most money
and then they can't trade it because
the draw downs, trade frequency or
other problems cause them hysteria
in the knickers. Pick exit strategies
that you are comfortable with.
Look at them on a chart. If you feel good
about what you are seeing,
put them in the tester. If the tester gives you
reasonable results---
even if they're less profitable than some other set
of
conditions—you're in business.
One of the first steps to
becoming a pretty good trader is to
understand that you don't, shouldn't
and can't maximize
everything. Don't even try. If it fits and you can trade
it, you'll
learn to improve it over time. However, you won't throw it out
because you hate the way the system messes with your emotions.
Now
run the system over all the time frames that show the
market is in an up
trend and that you have data for. Break the
periods into subparts and see
how the system performs. If it does
pretty good during all the up trends,
then you've got your up
trend system.
Reverse the process for down
trends. Most decent trend
indicators identify down trends just a good as
they do up trends.
Remember to test your down trend system WHEN THE MARKET
IS IN A
DOWNTREND. Seems obvious, doesn't it. Okay!
If you
test you're up trend system when the markets are in a
down trend, guess
what—it will perform poorly. Now why would you
trade it during a down
trend. Well, here's a clue---don't
trade it during a down trend unless you
have a death wish, and some
of you do.
If you don't like shorting,
then stay out of the market when
it's not in an up trend. Let me repeat
that, STAY OUT OF THE
MARKET WHEN IT IS NOT IN AN UP TREND IF YOU DON'T
LIKE SHORTING.
Well, you've read that the markets only trend 30% to 40%
of
the time. So how is this good systems development?
If you test
your system during up trends and it has 6 winners
for every 4 losers and
it makes 3X the profit for each 1X the loss,
if you run the tests when the
market is in a sideways pattern, the
systems test results are going to go
down. Now it produces 4 winners
for every 6 losers and only 1.5X the
profit for 1X the losses. You
can trade the system during sideways markets
but get ready for more
losers with smaller profits on your winners. In
addition, you
won't be able to hold the trades for as long. Sideways
markets may
require tighter stops, and different exit conditions. Do you
know how
you figure that out? Well, it involves using those market
bias
charts I talked about earlier when the market is in the sideways
pattern.
You may find that of three up trend systems, one works
better
in sideways markets. However, it's not going to work much better.
Nothing will because sideways markets baffle everybody. Up two days,
down three days is hard to trade, period.
You may want to stop
trading in sideways markets. A lot of
people do. You may want to consider
a sort of market neutral strategy
where you are taking longs and shorts at
the same time. Just
don't expect to make as much money. It ain't going to
happen.
In a sideways market if you violate the rules of good money
management and exit strategies, you are going to pay, and pay and
pay.
This is the time when money management and exit strategy is
everything.
Now you're thinking that can't be true, he's saying
entries mean
little, and that I can use almost anything when the
market is in an up
trend or down trend. Yes, that is what I'm saying.
LeBeau, Van Tharp and
others have tested all kinds of random entry
strategies and random exit
strategies and guess what. During the
trending markets they made money
with all kinds of dart throwing
crap.
In the sideways markets,
very few strategies made money. You
have to scratch out profits where ever
and when ever you can find
them.
These are all the secrets you
need to know to be successful.
Okay, there's one more success factor
worth repeating. Quit
trying to maximize everything, Stop it, stop it, stop
it. Maximizing
will kill you. There is no one best method, strategy,
theory, etc.
There are one or more strategies that fit you and that will
allow you
to trade with enough success to make money. If you try to find
the
maximal money making strategy, it will wrap itself around your neck
like a boa and choke the life out of you as punishment for violating
the common sense rule that maximization only works in theoretical
mathematics and engineering classes. In real life, it is going to eat
your fruits and nuts until you starve to death.
If you take your
up trend system and run it when the market
is in a down trend, it's going
to look very, very bad, and it
should. If it didn't it wouldn't work worth
a crap in an up trend. So
don't struggle trying to fix it so it finds the
one long trade out of
the hundreds of short trades that are there.
Let's summarize. You have two systems—one for up trends'
which is
long only' and one for down trends, which is short only.
You use them
according to the market bias derived from the weekly and
daily charts that
I mentioned. You learn that almost any half decent
trend indicator will
work when the market is trending, so you
don't worry about the perfect
setup, etc. You simply take the trades
when the trend indicator tells you
to take them. You spend some time
finding both a money management and exit
strategy that fits your
personality but is not the optimal strategy for
making the most
money. When the market is moving sideways you use your up
and down
trend system, but you recognize that trades are going to be quick
and
you're only going to make a little money. You will not fall in
love
with semi-meaningless words like over bought and over sold
because you
understand there really is not way to determine that. You
will, however,
recognize that almost every indicator is right part of
the time. Your job
is to figure out which ones you LIKE and when they
are likely to be
right. You will understand which market conditions
cause your
favorite indicator to decline in its predictive abilities,
and you will
adjust as need be using the market bias trend detection
system. And
finally, you will erase from your mind the thoughts that
it is possible to
maximize or minimize any thing for any reason
regardless of your
educational back ground, profession or belief in
higher powers.
I
think I'm going to write a detailed article on how to make
all of this
work for Roy's newsletter. In it I'll explain
what the better trend
indicators are and how to use them, and I think
I'll give more detail on
testing and trading these systems.
Sign
up, I think you'll find it very enlightening.
www.metastocktips.co.nz
No, I
don't work for Roy. I don't get paid for writing
anything in the
newsletter. Roy lives halfway around the world from
me. So why do I
recommend his newsletter all the time. For the same
reason I recommend
system development books.
Because the newsletter is directly on point
with a whole lot of the
questions I read on the boards. If you won't spend
$120 a year to
get your questions answered, improve your trading systems
dramatically and learn how to code your own stuff, then why should I
spend my time answering your pleas for help on the boards. If you
won't help yourself why should anyone else bother with you.
You'll
notice I mostly recommend systems development books rather
than trading
books. Systems development books tell you what works and
what doesn't.
Most trading books talk about somebody's
personal trading system, or a
system that newbie's can't seem to get
enough of. Ninety nine percent of
the time, after you've spent a lot
of money and time learning some guru's
pet system, you'll wind
up giving it up and doing what I'm telling you to
do. You never hear
the guru's tell you their system only works well when
the market is
in an up trend. You know why they don't tell you that,
because you
wouldn't buy their system.
Do you know how many of the
guru's trade from a large capital
account—almost none of them. At
least a few of them admit it. Most
of their money is in mutual funds. They
move it in and out of the
mutual funds using the trend identification
methods I've told you
about. They don't trade with their serious money.
Rather than follow the guru's, develop your own simple methods.
It
will serve you much, much better as will learning how to see the
market
bias without Gann or Elliot or some other complex as hell
method.
Have fun!
No virus found in this incoming message.
Checked by AVG
Anti-Virus.
Version: 7.0.300 / Virus Database: 265.7.0 - Release Date:
1/17/2005
Yahoo! Groups Links
|
No virus found in this outgoing message.
Checked by AVG Anti-Virus.
Version: 7.0.300 / Virus Database: 265.7.0 - Release Date: 1/17/2005
|