Stocks & Commodities V. 23:1 (30-35): Linking Point & Figure And Bar Charts by Robert Busby, Ph.D.
Ever thought of combining these two chart pattern techniques? Find out how.
Point & figure is one of the oldest western techniques of charting, tracing its lineage back to Charles Henry Dow. Large numbers of financial professionals rely on point & figure (P&F) charts for buy and sell signals. An even larger group of professionals look at financial data in the form of time series, and prefer bar, line or candlestick charts. Devotees of these two “parallel universes” are frequently at odds with one another, each asserting that only their version of technical analysis is correct. Of course, many people use both techniques; but there is at least anecdotal evidence of a sizable body of technicians who do not cross boundaries.
P&F VS. TIME SERIES
P&F advocates say that their charts contain all essential market information, filtering out market noise and simplifying decision-making. They assert that only market movement counts, not the time required to make the moves. But those who prefer time series methods say that time affects the value of money, closing price affects margin calls (and therefore volatility), and open-to-close trends confirm close-to-close trends. No such information is available from P&F charts.
Even those who use both methods tend to use them separately, and do not combine them. Analysts might like to combine P&F charts mathematically and graphically with other indicators, but they don’t generally have a method for doing so. They might want to test trading systems that combine traditional signals with P&F signals, but their software often will not allow it. For most market watchers, P&F charts are a technical end-game. Although many interesting patterns can be seen, further mathematical manipulation is not generally performed on them.
In addition, there are many methods for choosing box
height and reversal numbers in P&F charting. Some systems use closing prices, while others use highs and lows. However, questions such as which choices give the most reliable signals or avoid the most false signals are normally decided by trial and error — not explicit analysis.