Original E-mail

-----------------------

Hi Joel,

   I visited your website and just wanted to make a few comments especially on 
the mechanical hibernation. It was a Motley fool book that got me hooked into 
investing and trading, and from that viewpoint also helped me lose a lot of 
money, but I lost it, they didn't. On the otherhand the mechanical investing 
community helped put me at the beginning of a correct path (only right as long 
as the accounts are). If a strategy for stock selection has never worked in the 
past it is unlikley to work in the future, so mechanical backtesting can at 
least provide an idea of what really seems to work consistently.

I used the backtester site www.backtest.org to develop my own screens traded on 
monthly schedule. (I don't trade on a monthly schedule, I just use the monthly 
time frame as an indicator for finding stocks that are likely to perform well 
over a short period of time.

"It is really "hibernation investing". That is, you buy something, ignore it 
for a fixed period of time, and then wake up from your hibernation to re-
evaluate the situation. There is no inherent reason why one should invest on 
the first Monday of the month, or the last Friday, or any other day."

This is true, but in a very noisy system, it is a perfectly fine methodology, 
as days average out over time. the screens can be tested for different monthly 
start dates for annual hold screens. It is unfortunate that you cannot vary the 
day of the week for the start of a screen.

As the interval of time that you trade mechanically goes to zero, then most of 
the screens especially the best performing ones actually reduce to: cut your 
losses short and let your winners run. (The main drawback of the backtest site 
is that you cannot reduce the holding period to less than 1 month.) That is 
most of the best screens are based on relative strength over some interval of 
time so some of the folowing statement is in error.

"While many of them say that they say that they do not use technical analysis, 
in fact they do. However, the technical indicator that they use is the weakest 
and most useless technical indicator of all: IBD's relative strength." 

Not the first part of course, many of their screens use technical analysis 
because they use stock returns over 4,13,26, and 52 wk periods. They do not 
primarily use the IBD rs ranking which is basically useless. Many screens have 
4wk Returns, 13wk Returns, 26wk Returns, and 52wk Returns. All of screens I 
have developed using the site use some combination of these returns to rank 
stocks.

They also use relative strength indirectly thru the valueline timeliness 
rankings. I do not use the valueline rankings though because what good are they 
for all of the other stocks in the stocks universe that they do not cover.

I won't say the other comments are off base, because they are largely correct. 
Market timing is possible its just not perfect.

However,
"The energy thay spend on developing new systems, or on tweaking the old ones, 
would be better used learning how to trade instead of just re-arranging the 
components of the screens like the deck chairs of the Titanic. "

Technical analysis boils down to the same thing for many people, "just re-
arranging the" techincal indicators "like the deck chairs of the Titanic."

Anyway just wanted to make a few comments. Best, Allan

------------------------

My Reply (> stuff) and his Comments

------------------------

Joel,

>I have not looked at the screens in detail for some
>time, and my recollection was that they relied more on the IBD ranking than
>you say.  

Nope most of the screens don't use the IBD rank. In fact, at the 
www.backtest.org site, under standard screens their is list of screens 
developed by MI community. At least 20 use RS directly and only one uses IBD-RS.

There has been an interest in using the WER (the ibd weekend review) list as a 
mechanical screen in recent times which uses the IBDRS rank, but this is just a 
few screens out of a large number of screens which use more recent stocks 
returns for prediction.

>However, trades should be based on events, such as a
>technical indicator firing - e. g. the famous "cup with handle" that the
>CANSLIM people look for.  That is a lot of work.  I did it for a while, with
>decent results, but it is extremely time-consuming.

I agree and it takes forever to enter a position if you only use the CANSLIM 
entry points, various technical indicators provide other entry points, which I 
prefer to use.

Actually, it is really funny that some MI's say they don't use techincal 
analysis, everyone of the highest return screens they developed uses either 
various returns or timeliness rankings which are also dependent on relative 
strength. last time I checked RS is Technical indicator.


> Would you object if I put your message on that page? 

No.

> I would not include
> your E-mail address, unless you wanted me to.

probably best if you didn't.

Best, Allan