In the early 1980’s I subscribed to a number of stock market newsletters. My favorite was Investors Intelligence, and to their credit they are still in business.
I loved receiving their letter. It contained a summary of other newsletters and introduced me to the concept of sentiment and “point and figure” charting.
If you have never seen or thought about “point and figure” charts, they are curious and interesting collections of X’s and O’s, and they look completely different than bar charts. PF are only concerned with vertical movement in a stock, and time (or horizontal movement) is disregarded. This can have certain advantages over conventional charts.
They don’t contain any points or figures at all. A brief history can be seen here: Wikipedia - Point and Figure Charts
What I liked about them was that they filtered out the smaller fluctuations in price movement and they generated numerical buy and sell signals. I hand plotted numerous PF charts of the various stocks and futures I was trading and I was able to make profitable buy and sell decisions using them. Unfortunately, I didn’t always follow the signals generated by the charts - a common weakness of all traders, and often a financially fatal weakness.
The Apple computer was around in those days and IBM came out with its PC and PC Junior in the early 1980’s, but stock market charting programs were not available. So, hand plotting was my only option. And, I read some years later that doing the work by hand was preferred by many traders since it gave a more “intimate feel” for each stock or commodity. And, while that may or may not be true, it is true that hand plotting does take a lot of time and is prone to mistakes. And, when you are trading you don’t want to make mistakes.
The longer I worked with PF charts the more I thought the “normal” method of plotting them needed improving. First of all, the box sizes were fixed, then at some predetermined and arbitrary level they changed. Well, I thought that this was ridiculous, so I began plotting the price scale as a log scale, so that every box became an equal percentage of the previous box. When I did this the buy and sell signals that the charts generated improved.
A few years into my charting, and after I had a computer, I wrote a little utility program that allowed anyone to print a Log Point and Figure chart on any dot matrix printer, and I uploaded it to Compuserve for anyone to use.
I also began to pen the up columns in green and the down columns in red. I added little green and red flag symbols at the buy and sell points, and eventually I got totally sick of drawing all the X’s and O’s so I just put a dot at the top and bottom of each column and drew a quick vertical line to fill it in. When you’re hand plotting charts the last thing you need is more aggravation. Later I used colored pencils to shade the columns either green or red. I was having fun, but this was very time consuming.
Eventually stock market charting programs became common, yet there was no way to computer backtest a PF chart system. So, I started writing one, but never finished it.
I eventually came to the conclusion that any system is better than “seat of your pants” trading, and that a PF chart system was just another way of filtering price and bringing me to a rational buy or sell signal. From that point on, it was up to me. Without an automated trading method, a “hand’s off” method of placing my trades, it still came down to me. Should I take the signal or not? The chart says sell, but what do I think? Should I think? What do “the experts” think?
I found out that this sort of questioning leads to losses, and while a system seldom produces more winning than losing trades, that system, in the long run, is going to make more money than “do I really want to do this” trading. I found out that I needed a system, and I needed to know if the system worked over time, and on which stocks.
And this is still my preferred method of trading.
Here is a link to one my early hand drawn P&F charts of PWJ (Paine Webber). It shows PWJ moving up strongly from the summer of 1982 at the start of the five year bull market that began on August 13th. There was only one sell signal generated.
Recent Comments