Forum Archive Index - February 2003

[Travis Morien <travismorien@yahoo.com>]

--- Phaedrus <Phaedrus@techemail.com> wrote:

> <<<< The counterargument would be to show the chart
> of a stock that didn't stop at the same point twice,
> and there are thousands of these.>>>>
>  That is not a counterargument at all! A stock that
> is continuing to make new highs doesn't stop at the
> same point twice, and as you rightly observe, shows
> no evidence of sustained resistance at any point.
> Most people call that an uptrend and there are
> indeed thousands of examples of these. Your
> counterargument is completely specious.

Support and resistance, whether it exists or not, is a
virtually useless concept.

What it basically implies is that a stock will either
go higher than a previous high - or it won't.  Or it
may fall below a previous low - or it won't.

No way of knowing in advance.
It doesn't work far more often than it does.

What exactly are we supposed to *do* with this
marvelous insight that stocks have support and
resistance anyway?
> 
>  You seem to have misunderstood the concept of
> resistance. It is NOT something that is always

Nope, i understand it completely.  I was a trader for
many years and have a deep understanding of technical
methods.  I don't use them because I realised that
they don't work, and have never worked.  Successful
traders use money management and risk management
systems designed to work for their trend following or
reversal systems.  I've never seen any evidence that
any form of technical analysis works.

In fact academics first got the idea of market
efficiency by scrutinising technical analysis. 
Unfortunately they figured that if TA doesn't work
that implied that FA didn't work, and it was 20 years
before the CAPM and beta and all that other rubbish
was dumped in favour of the new model (much like the
old model) Fama/French Three Factor analysis and
studies of regression vs momentum.

> present. It means something if it is present, and it
> means something else if it is not. It is of
> particular significance if a level that has
> previously been respected many times is breached. To
> me it is self-evident that this indicates a change
> in market sentiment. 

It works unless it doesn't work.  A trend will tend to
keep going until it stops.  A stock will go up unless
it goes down.

Technical analysts have developed over many years a
sophisticated Delphic phrasing system that enables
them to sound like they are deeply insightful on a
stock's progress when in fact almost everything they
say is an each way bet.
> 
> <<<<Any method of analysis that works only in
> hindsight and does not indicate much of any
> consequence in advance is of very limited
> usefullness.>>>>
>   Travis, that's three mistaken assumptions in a
> single sentence!
> (1) The trendline break Sell signal shown on the
> chart was evident in real time. When TLS price
> action gives a Buy signal by breaking above the
> current downward trendline, that, too will be
> evident in real time. I will post it here for you
> when it happens, if you like - no hindsight
> required! This trendline has been in place for 3
> years now, and has been confirmed on many occasions.
> Where, exactly, do you see hindsight as having been
> used here?

Stocks on average have about a 50% range between their
high and their low in any given year.  Autocorrelation
studies of the stock market show that trends in stocks
are no more common than one would expect them to be in
random walk data.  trends are impossible to tell apart
from minor movements until the trend is already well
established.

However a value investor trying to estimate future
return on equity, sales growth , profit margins and
growth in book value could see that Telstra was
overpriced a long time ago.  My analysis shows that it
is *still* overpriced. (though that is relative to
what discount rate one adopts)

> (2) This system indicates NOTHING in advance, and no
> claim of indicating anything in advance was ever
> made. I do not believe that anyone or any system can
> do that. This is a trend-following system - note how
> the exit was made late and below the peak. When the
> Buy signal comes, rest assured that it will be late,
> and above the preceding low.

What good is a system that indicates nothing in
advance?

Trend following systems have the fatal flaw that they
result in very high turnover, which wrecks tax
efficiency and adds massively to trading expenses.  I
gave up trading and swore off all technical methods
the day I did the maths working out what kind of
pre-tax return I'd need to get to compensate for my
trading expenses and short term CGT events compared to
a buy and hold investor.  My failure to find a single
role model of a famously successful trader that didn't
go broke eventually was also powerful incentive to
stop speculating.

I was profitable as a trader, it worked while I was at
it.  (At least until the point where I bought index
put options just before the bear market began - a
month or two too early and thus lost money).  I gave
up trading because I realised that bigger money was to
be made as an investor, with less risk.

> (3) Very limited usefulness? I disagree, strongly.
> This chart is an excellent example of just how well
> TA can work. I built my TLS stake over the week or
> so when it first listed (acting on the basis of my
> trading rule #28 - buy anything the government
> sells). I sold on the trendline break, as charted. I
> have not held TLS for nearly four years now, but
> will seriously consider buying again when price
> action breaks above the current downward trendline.

Why?  Do you have reason to believe that the present
value of the future profits of Telstra exceed the
price of Telstra stock, or are you just hoping that
everyone else does and you'll be able to ride on the
backs of investors?

How will you know if the upward break is not just a
false signal, and how will you estimate your future
profit?

> I know people that are still holding, having given
> most of their gains back to the market. Other poor
> souls have bought high, and been buying all the way
> down. Averaging down, Dollar-cost averaging. Very
> expensive mistakes here. How have you fared with
> this stock? 

Got out pretty much at the peak, aeons ago. 8)

I more or less completely got out of stocks in 1999 in
fact.  At the time I was using technical approaches in
addition to valuation approaches, but valuation
approaches gave an earlier and more reliable signal.

It wasn't until mid 2002 that I really seriously got
back into the market on the long side.  Until recently
most of my recommendations were sells (CSL and BIL
were doozies!)

>  I was somewhat amused by your choice of quote from
> Ben Graham - from a TA perspective, you could hardly
> have chosen a better one to illustrate exactly what
> we are discussing. :-
> "Market movements are important to (the investor) in
> a practical sense (CHARTS!!!) because they
> alternately create (depict) low price levels at
> which he would be wise to buy (SUPPORT!!!) and high
> price levels at which he certainly should refrain
> from buying and probably would be wise to sell."
> (RESISTANCE!!!)

Support and resistance levels rarely ever coincide
with levels of intrinsic under and overvaluation.  If
they do, this is entirely coincidental.  What Graham
is saying is that if a stock is sold so low that it is
undervalued it then becomes a good buying opportunity.
 If the stock is overvalued, you might consider
selling it.  Support and resistance don't come into it
because support and resistance levels are set by
speculators with no concept of value, only price.

The next bit says "Conceivably they may give him a
warning signal which he will do well to heed - this in
plain English means that he is to sell his shares
*because* the price has gone down, foreboding worse
things to come.  In our view such signals are
misleading at least as often as they are helpful. 
Basically, price fluctuations have only one
significant meaning for the true investor.  they
provide him with an opportunity to buy when prices
fall sharply and to sell wisely when they advance a
great deal.  At other times he will do better if he
forgets about the stock market and pays attention to
his dividend returns and to the operating results of
his companies."



>  Why do you think TLS stopped rising at the price it
> did? Maybe there were enough followers of Ben Graham
> selling, at a level where they now considered TLS to
> be overvalued, to stop the meteoric rise of TLS in
> its tracks. Whatever the cause (and I believe that
> nobody could answer that question with any
> certainty) the collective market opinion was
> unchanged ten months later. At 916, TLS again hit a
> wall of selling. In a year, TLS had moved from being
> an undervalued stock to being an overvalued stock.

The stock was overvalued long before it got to those
levels.  Telstra had ceased to be an investment and
quickly became a speculation only a short time after
it was floated.  Once it reached the point where
reasonable estimates of the present value of future
earnings were well below the selling price value
investors had no further influence on the stock.  Why
should they, value investors make up fewer than 5% of
market participants?
 
>  I am a simple, unsophisticated chap. To my mind,
> individual stock prices are, overall, either going
> up, down or sideways. I have found that if I buy and
> hold the ones that are in uptrends, I make money. If
> I buy and hold the ones that are in downtrends, I
> lose money. I have found that it is possible to make
> money by short-term trading stocks that are moving
> sideways between an upper and lower boundary in a
> trading range, though I do find this a little more
> difficult than trend following.

Have you ever calculated the before tax return you'd
need to get to compensate for your trading and tax
expenses compared to a buy and hold type?

I'm sure you are a very clever person but if you
really can achieve a return *double that of the
market, for extended periods of time* then you
shouldn't be mucking around here - go apply for a job
with the Quantum fund, you and George Soros have a lot
more in common than you might have supposed.

>>  Travis, it is apparent that we have some deep
> philosophical differences. In an attempt to
> ascertain just how deep, I would like to know if you
> accept/believe/acknowledge/admit/concede that stock
> prices sometimes move in sustained trends, and that
> sometimes they move sideways, ranging between an
> upper and a lower level.

Whether they move in exploitable trends that can be
recognised ex-post, often enough to pay for
transaction costs is debateable.

Momentum in stock prices is well known, see for
example Louis Chan, Narasimhan Jegadeesh and Josef
Lakonishok, "Momentum Strategies" The Journal of
Finance 51 (no. 5), December 1996 where both price and
earnings momentum was demonstrated on a time frame of
up to 12 months from portfolio formation.  However,
the momentum premium was below what one might
reasonably expect transaction and excessive tax costs
to be for a high turnover strategy, so momentum
investing doesn't really work in practice.

On the other hand, it is also well known that over
periods of five years or more a company's price will
move in almost perfect correlation with earnings.  If
a stock is genuinely undervalued (something that can
be identified for the more extreme examples of under
(or over) pricing) then the future movement of the
stock is likely to be biased in the direction of the
true valuation.  An undervalued stock is still going
to exhibit something close to a random walk over short
timeframes however on longer time frames trends may
manifest themselves.

If you are a chart reader, and only a chart reader,
you are probably too busy watching the monkey to
notice the organ grinder who is the one really calling
the shots.  You may well mistake numerous short term
runs with trends, and end up churning your account
when these runs revert to normal trading.  The normal
course of the market is for trends not to exist.  When
a true trend is in progress it is likely that the
majority of the movement will take place before it
becomes apparant that this is a trend and not a
retracement.

I liken my value investing to hedging.  When you buy a
futures contract you reserve the right to buy that
asset at some time in the future at the price you
bought the contract at.  A rise in prices won't hurt
you because you'll make a hedging profit on the
contract which you can use to offset the spot price of
the commodity you wish to buy.  

If the price falls then that needn't bother you much
as a hedger because your purchase decision was made
with the assumption that today's price was acceptable
to you and the chance of a fall in purchase prices was
less significant than the consequences of a rise in
prices.

As a value investor I seek to predict the future
return from owning a business, using mathematics
identical to that used to value a bond.  Based on my
assumptions of future profits, which for some
businesses are certainly more predictable than stock
prices, I can calculate with a reasonable degree of
certainty what the "yield to maturity" might be from
buying that business/bond at that particular price. 
If I find that a stock has a yield to maturity that is
very high, say 15 or 20%pa over a ten year period, I
consider this an outstanding opportunity and wish to
lock in that profit by buying today.

Like a hedger, I am happy with this return and try not
to become too concerned if my decision to go long was
a little bit early.  More often than not I'll buy more
stock if the price falls still further and I notice
that this business that I covet trades at still higher
yields to maturity.

Travis
www.travismorien.com


__________________________________________________
Do you Yahoo!?
Yahoo! Mail Plus - Powerful. Affordable. Sign up now.
http://mailplus.yahoo.com

----------------------------------------------------------------------------
To remove yourself from this list, please use the form at
http://www.sharechat.co.nz/chat/forum/