John Bollinger has done a lot of work discussing the ramifications of the width of Bollinger Bands. In short, if the Bands are too close together (too compressed), then volatility is “too low,” and the market is due for an explosive move – probably to the downside. Conversely, if the Bands are quite far apart, then volatility has gotten “too large” and a contraction in volatility – and probably a stock market rally – is at hand.
The CBOE introduced the Volatility Index ($VIX) in 1993. The calculation of $VIX has changed a couple of times over the years, and due to the complexity of those calculations, $VIX itself cannot be traded. However, in 2004, $VIX futures were listed, and in 2006, $VIX options were listed. $VIX futures are the underlying instrument for all of the Volatility ETN’s and ETF’s that exist today (VXX, for example).
There is a seasonality to volatility that has persisted over the years. Not every year is the same, of course, but the general pattern is similar. This can sometimes be useful in helping one determine whether to expect increasing or decreasing volatility during the life of positions that are being established.
Many volatility traders – we are among them – complained about the lack of response by volatility derivatives during last fall’s market decline. That was especially true in the downward thrust in December. $VIX itself managed to put together a decent move, as it rose from 16 in early December to 36 on Christmas Eve. But one cannot trade $VIX; only the $VIX derivatives are available for trading.
We have written repeatedly about the similarities between the markets of late 2000 and early 2001, ascompared to late 2018 and early 2019. Those comparisons are still valid.
Last Friday (January 18th) the “stocks only” breadth oscillator stood at +854.04 – the fourth highest reading of all-time. This is extremely overbought, but is not a sell signal (“overbought does not mean sell”). In fact, in the past, extremely overbought readings have often led to much stronger markets in the short term.
Last year (2018) was a very interesting year in a number of respects. One of those was the behavior of volatility and especially the behavior of volatility derivatives. Since one cannot trade $VIX but must instead trade one of the listed products – $VIX futures, Volatility ETN’s or ETF’s, or options on those instruments – there are some nuances involved. Since all of those instruments are based on $VIX futures1 , that is where we’ll concentrate this discussion.
We often talk about how the stock market usually trends in the opposite direction from volatility ($VIX). But how do we really measure the trend of $VIX? Often, we use the 20-day moving average, but that is a very short-term moving average that changes trend easily. In a longer-term bull or bear market, we would not want to swing in and out of a “core” position when we are trying to stick with the trend.
If you enjoy seasonal trading patterns, they abound from the end of October (the “October Seasonal,” which was strong this year), through the beginning of the new year (the “Santa Claus” rally). Over the years, we have combined three different late-year seasonal patterns into one trade.
The three patterns are as follows:
Several times, we have mentioned the fact that in a bear market, there is usually selling in October, followed by a strong October Seasonal rally, and then a failure of that rally in early November. If it is truly a bear market, new lows are made in November or early December.
