Technical indicators like the RSI, that suggest periods when a stock or index is 'overbought' or 'oversold' are one thing, it's another when we add a definite buy or sell signal. Such crossover signals occur with the stochastic indicator.

Before I launch into today's general technical topic for this Trader's Corner, here's a mini-update on this week's technical action:


In my most recent (Sat, 9/23) Index Wrap I discussed how tradable tops or bottoms can tend toward equal percentage moving average envelope extremes above and below a centered moving average; e.g., a 21-day average. After hitting the upper 4% envelope line, prices have dipped and this may mark at least a temporary top. Since I always want to look at more than 1 key indicator, I've noted the 9/20 extremes (with the red down arrows) in both the RSI and my sentiment model; both extremes also supporting the idea of possible top. Of course we don't know yet if we're seeing just a temporary pause in the advance or the beginning stage of a deeper correction; e.g., back to the 1100-1080 or lower.

Extremes in the Bollinger Band (BB) indicator can also signal topping or bottoming action. This indicator is seen on my second chart, that of the Nasdaq Composite.

My reservations about using Bollinger Bands is that it can be hard to pinpoint a specific buying or selling area, as these bands expand or contract according to market volatility. This is both, I suppose, the technique’s strength and weakness. Unlike moving average envelope lines that are a FIXED percent above or below the moving average, Bollinger bands are plotted TWO standard deviations above and below an unseen 'center' moving average of 20 days, as set by John Bollinger.

The telling pattern I've found in the BB indicator, is that after upper or lower volatility extremes are reached and when prices go SIDEWAYS after that (e.g., for 2-3 days or a greater number), trend reversals tend to occur. This pattern (sideways after intraday lows or highs reach the lower or upper Boli bands) is seen in all the numbered examples below except #1. The sideways trend seen in example #6 was premature in suggesting a 'final' bottom. I am assuming that the recent BB extreme may mark a tradable top but not enough subsequent price action has occurred to say definitely; a top IS consistent with historical patterns.


On my next chart, that of (again) the S&P 500 Index (SPX), I have applied the slow stochastic technical indicator set to a length setting of 13. On the daily chart, this model then always uses the prior 13 days of Closing prices for its calculation. I've noted on the price portion of the chart (with the red down arrows) the last 5 downside crossovers of the slow stochastic which constituted sell 'signals'. Conversely, there were 2 upside crossovers of the two lines composing the slow stochastic model that constituted buy 'signals'. The crossovers are always AS OF the Close. Note that some downside crossovers occur ABOVE the upper level line (set to 80) and some upside crossover occur BELOW the lower level line (set to 20). These are the crossover signals to pay attention to. A fuller explanation follows.

Now that I hopefully have gotten your attention with how accurate these crossover 'signals' appear to be in identifying tradable tops and bottoms, there are background explanations needed as to what is seen above.

Indicators or technical 'studies' can be grouped into trend following type indicators like moving averages, or into indicators of a type called oscillators.

The central concept of the stochastic model or what it is attempting to measure, is that in an up or down market trend for any number of trading periods (e.g., 13-day or 21-hours), prices will at some point move away from the lowest low made during that period OR the highest high, at an increasing rate. This 'rate' or speed of price change is what the (slow) stochastic oscillator is showing visually. There is also a fast stochastic; more on that shortly.

An indicator which has a formula such that its scale is between 0 and 100 only, like Stochastics or RSI, will 'oscillate' back and forth between the low and high end of this fixed scale. Hence this "class" of indicators or technical indicators are called OSCILLATORS. The MACD (moving average convergence-divergence) has an 'unbounded' range as it can have values below 0 and above 100.

Oscillators, such as the Relative Strength Index (RSI) and the Stochastics model tend to of most use to shorter-term traders such as those trading options. The related concept of 'over-bought and over-sold' is one part of how we use such indicators. There are other concepts that suggest possible tops or bottoms; i.e., the concept of bullish or bearish divergences.

'Overbought' or 'oversold' should always be thought of in relation to what time frame is being considered. Traders are usually most interested in short to intermediate-term reversal points; e.g., 2-3 days to 2-4 weeks. Concepts of overbought or oversold relate to a potential vulnerability to a reversal. However, of course, stock or major market index can stay in areas considered 'oversold' or 'overbought' for a relatively long period. It is also true that rapid and steep advances or declines are not usually sustainable for an unlimited period; the exceptions are mini or major 'bubbles'. A market will correct at some point, even if just to go sideways for a time after a steep rise or fall. WHEN such a correction will occur is an open question in a strong trend. Less so in more of trading range market.

There can be a tendency to think that the first time or two that an oscillator reaches an extreme reading, whether using an hourly, daily or a longer-term weekly chart, it is time to exit a particular stock or index option. However, there are many instances where the trend takes prices significantly higher or lower before there is a correction. And a 'correction' may just turn out to be a sideways consolidation period, before there is yet another push in the same direction as before.


The Stochastics indicator attempts to highlight 'momentum' strength and, in terms of momentum strength, areas where a stock or index has reached high or low extremes. The stochastics process (its first name) study was popularized by George Lane, who was involved in trading futures markets.

If I select Stochastics in a charting application I can typically edit 'length' (number of trading periods used for calculating the indicator) something called 'smoothing' of %K and smoothing of something called %D. The length setting is the one for sure you may want to vary. You can set smoothing % K to '1' so there is NO smoothing of this number and leave the default setting for smoothing %D at 3; i.e., this setting is the one that 'smoothes' out %K and that is the way this indicator was designed to be used.

The most common default setting for 'length' I've seen (the number of hours, days or weeks that the stochastics formula will reference) tends to be either 9 or 14-periods or bars. The right hand scale is 'normalized' or made such that the readings are always on a scale between 0 and 100.

The stochastics default setting for oversold and overbought levels is typically 20 and 80. The stochastics indicator tends to have wide-ranging fluctuations between 0 and 100; e.g., a low at 5 or 10, a high at 90.

The stochastics indicator is composed of two lines: a slower line called the percent D (%D) line, which is a simple moving average of the faster %K line. As with the MACD, the two lines of varying speeds lead to upside or downside crossovers that are thought of as buy or sell 'signals'. The following charts are historical examples mostly taken from my (Essential Technical Analysis) book.

NOTE: You can skip over this section of what the 'slow' and 'fast' elements are if you are only interested in knowing that you'll probably ONLY want to use the slow version of stochastics.


The Stochastic formula 'looks' at the current price in relation to the highest high or lowest low in the period being measured. The stochastics formula plots the current close in relation to the price range over the length set for this indicator and gives this a percentage value.

The initial calculations for a stochastic of 14-days (used in the above chart) are twofold as 'fast' and 'slow' lines are established. The fast line or %K formula is 100 – (the close minus the 14-day low) divided by (the 14-day high minus the 14-day low); i.e., the price range for that period. The slow line or %D (here called 'FastD') is equal to a 3-day average of %K. This first formula is referred to as the fast stochastic model. The fast stochastic lines react so quickly to price changes that it is mostly appropriate for very short-term traders. Its popularity was established in the major commodities bull markets of the 1970's.

The 'slow stochastics' variation of the basic stochastics formula is simply to take the 'FastD' figure and apply a smoothing calculation yet again, which results in another line which we can call 'SlowD', to differentiate the two versions of %D. The important thing to remember is not this alphabet soup, but the fact that the slow version of the stochastics oscillator (slow stochastics) is the version that is in most common use and is most likely what you will be using if you choose the stochastics indicator to apply to a price chart. Use of the slow stochastics indicator is the most appropriate for all around use.

If an upside or downside trend acceleration is especially strong, the two stochastic lines will reach a reading of 20 or below or 80 and above, on its percentage scale; 20 and 80 begin a level generally defined as oversold and overbought, respectively. In the overbought area at and above 80, the rate of price increase is thought to be too steep to be sustainable at the same rate of change; such an extreme suggests that there is an increasing probability that the stock or index will experience a correction, either by a sideways consolidation or a downward reaction.

At and below a reading of 20 in the slow stochastic indicator, the rate of decline is relatively steep and is also considered to be unsustainable. The assumption and much experience in coming to this conclusion, suggests that the market in question faces a significant likelihood of a correction, either a sideways consolidation or upward reaction.

Buy and sell CROSSOVER signals are considered to be optimal if they occur at or above the overbought (80) levels and at or below the oversold (20) levels, respectively. There will be instances of crossovers that occur in the middle of these ranges and these should not generally be traded on unless there is compelling other technical considerations that are guiding you; e.g., a break out above or below an important trendline.

The most frequent use made of the stochastics indicator is to serve as a model to show price momentum visually and as an overbought/oversold indicator. Stochastics, as well as the RSI and MACD (think of these as the 'big 3' of momentum type indicators), are also worth their weight in gold for the occasional bullish or bearish divergences; that is, when the indicator does not 'confirm' or make a new high or low along with a stock or index price.

I tend to use stochastics somewhat less on weekly charts. I do suggest monitoring stocks or other financial items of interest such as bonds, futures and FX markets, with at least one of the 3 major oscillators on a weekly basis.

Whether highlighted by use of Stochastics, RSI or MACD, once price action registers overbought or oversold extremes, there is simply a greater probability of a correction or trend reversal within a relatively short (upcoming) time frame. Keep in mind that there are of course exceptions to this. In very strong up or down trends, the other important consideration for staying in a trade is by use of trend following indicators such as moving averages.

Stochastics can be a very useful indicator, but its BEST use will tend to with stocks or indexes that are trending back and forth in a well-defined price range. That range can be a 'horizontal' one, such as seen in my next to last chart:

A beneficial use of stochastics can also occur with a stock or index where prices are trending lower or higher within a wide-ranging downtrend or uptrend price channel, as seen in my final chart: