Oscillator characteristics

Any oscillator essentially measures the rate of change, or momentum, of price by comparing the current price to price at a certain point in the past. Typically, the difference is the current bar’s closing price minus the close x bars ago; alternately, the current price can be divided by the price x bars ago.

Momentum calculations basically “detrend” price data (that is, they minimize trend influence longer than the look-back period), plotting price swings above and below a horizontal line that represents the trend. Some oscillators (such as the RSI and stochastics) use calculations to place their values between fixed upper and lower levels (e.g., zero and 100).

It shows a hypothetical price series with a two-bar momentum calculation (current price minus price two bars ago) plotted below. When price is falling at a steady pace, the momentum calculation stays at -20 points (A), meaning the current price is 20 points below the price two days ago at each of these bars. As the rate of descent of the price begins to slow (B), the momentum reading begins to climb (although it is still negative) before the final bottom (C) occurs in the price series. When price turns up, the oscillator crosses above zero into positive territory (D).

The oscillator continues to rise as the price climbs (E). The momentum calculation peaks at F and then declines, which means that, although price is still increasing, it is gaining at a slower rate. For a momentum calculation to keep rising, the price gains must continue to increase in size; if the gains are the same from day to day, the momentum indicator will be flat. Price itself does not peak until point G. This “leading” characteristic of oscillators is the primary reason traders gravitate toward them.

But this leading characteristic can also place you on the wrong side of the trend. For example, the oscillator bottoms (A) ahead of price (B). The market rises and the oscillator peaks (C), but price continues to a new high (D). The oscillator peaks a second time (E), but price continues to advance. The oscillator is “diverging” from price, which in this case is a false indication of a reversal. The rally might be losing momentum, but price is nonetheless moving higher — and it could continue to do so for an indeterminate amount of time. Comparing the price action can explain why the oscillator looked beneficial in the former and not the latter. The calculation is detrending the price action and showing the current cycle of the market. If that cycle happens to fit the oscillator’s look-back period then the oscillator will lead the tops and bottoms just as the trader hopes.

If it doesn’t, the indicator will give false indications of peaks and bottoms. This occurs because oscillators typically use fixed look-back periods and do not adjust to the market’s ever-changing cycle length.

Some technicians use sophisticated mathematics to determine the current dominant market cycle and adjust their indicators on the fly. A simpler solution is to track an oscillator’s range by comparing its current value to a certain past value — in other words, create an oscillator of an oscillator. Comparing the current oscillator reading to the range of its readings over a given period creates a new indicator that is more in tune with dynamic market action. For example, when the market cycles become short and cause the oscillator to turn up from a higher level than the previous oscillator bottom, a dynamic oscillator will pick up on this event. The stochastic-RSI is a hybrid indicator that accomplishes this goal.