PRICE OSCILLATOR - S&P 500
APPLICABLE TIME FRAME(S):
Each weekday night by 7:00 PM EST
This indicator is an offshoot from a part of the 3-day cycle trading method outlined by George Angell in Winning in the Futures Markets. While the formula given in the book was part of a larger system and used on daily bars, I have found better use for it using 30-minute bars, and with a moving average to smooth out the fluctuations.
The formula is somewhat convoluted, but it does a good job at giving a "score" to individual price bars. Here is the formula itself:
((High - Open) + (Close - Low)) / (2 * (High - Low))
Let's look at an example. Say we have the following information for a particular price bar:
Then, that bar would get the following score:
((852 - 848) + (851 - 847)) / (2 * (852 - 847))
= (4 + 4) / (2 * 5)
= 8 / 10
Since the bar opened near its low and closed near its high, it received a high score. While this is bullish for a time, when it becomes "stretched", that's when we need to worry. By following this information on a 30-minute time frame and using a 13-period moving average (one full trading day), the data can be responsive to intraday movements without giving us too many whipsaws.
When the Price Oscillator reaches an extreme, it often marks short-term exhaustion in buying or selling pressure. We generally use readings over 59% to indicate an excessive amount of buying pressure (particularly when in a longer-term downtrend), and readings below 41% to indicate that the selling may be overdone (especially when in a longer-term uptrend). This indicator works especially well within defined trading ranges, and will give a false signal (likely becoming very extreme) when a trading range is broken and a new trend begins.
The chart below shows five distinct occurrences of the Price Oscillator reaching extremes. When can see that each time the market became overbought, indicated by a peak in the oscillator reading over 59%, the market had great difficulty sustaining any upward thrust. Conversely, each time we became severely oversold, indicated by a reading below 41%, the market rebounded or at least there was a letup in the selling pressure save for one instance. In the context of such a severe decline, oversold readings are less reliable, and it usually takes a more extreme reading to produce a change in trend (note the February 13th reading of 31% - more than two standard deviations from the average).
Although this is a real example and points out the value of following this information, we do not mean to intimate that the market ALWAYS peaks when the Price Oscillator reaches overbought, or troughs immediately after the indicator reaches 41%. It is a guideline and not a trading system unto itself.
*Standard Deviation. See below...
68% of readings (1 standard deviation) should be between 41% and 59%
95% of readings (2 standard deviations) should be between 32% and 68%
99% of readings (3 standard deviations) should be between 23% and 77%
In other words, we should expect a reading under 32% or over 68% approximately 13 times per year. Since such a reading would be unusual, it suggests that we are seeing an unsustainable trend. These figures assume a normal distribution curve.
© 2005 Sundial Capital Research, Inc. All Rights Reserved.