Yesterday’s steep sell-off was accompanied by a pop in implied-volatility readings. Gauges such as the VIX and VXN rose about 11% on the day. This breaks what had been a pretty steady trend lower over the past 4 months in which those measured have declined some 40% since the market made a low on March 9.

But even as the market continued to rise in late May and early June, the VIX became sticky, remaining near 30 even as the 20-day historical, or real, volatility of the S&P 500 declined from 28 to 24 during the past month. This has led some to the conclusion that 30 was an important level that indicated the market was poised for a pullback. Yesterday’s sell-off -- and the accompanying jump of the VIX back above that round number of 30 -- seems to confirm this theory.

Using the VIX as a Predictor

One problem with this thinking is that the VIX is a statistical measure, and it's the result, not the cause, of any market move. Additionally, because it's a statistic, applying technical or charting analysis typically used on stock, isn't very effective.

But this isn't to say the VIX can’t be used as a predictive tool.

Most people tend to use it as a contrary indicator. That is, like most sentiment readings, when it gets relatively high -- an indication of fear or caution -- it's taken as a bullish sign. If it's low -- which is interpreted as complacency -- it's regarded as bearish.

The problem with this approach is that the VIX not only can stay low for extended periods -- it was at decade lows for most of 2006 as the market hit all-time highs -- but it can also rise during bull markets, such as during the tech bubble.

Probably one of the best ways to use the VIX as a predictor would be to use it for short time frames, when it doesn't respond as it should to the underlying market conditions.