The up and downs on Wall Street have fascinated and frustrated countless investors over the years, rewarding only those with a keen eye for opportunity. Discipline has always been one of the key ingredients in successful portfolio management, although even the most seasoned of veterans have been rattled by the markets’ volatility at times. Arming yourself with the proper tools and strategies is only half the battle; knowing the context in which indicator or approach is most practical will arm you with confidence to pull the trigger while others are scrambling to react [see 3 Market Valuation Indicators ETF Investors Must Know].
One such approach that strives to bring clarity in moments of uncertainty is trading volume analysis, which is an age-old discipline embraced by fundamental and technical analysts alike. Volume is simply the measure of how much a given asset has changed hands in a certain period of time; volume data in itself doesn’t offer much insights, however, it does warrant closer analysis when we consider it in the context of price action [see ETF Technical Trading FAQ].
Does Above-Average Volume Have Predictive Power?
Investors and traders alike are generally quick to take note of above-average trading volumes, but do big trading days actually hold any predictive power for future returns? Below, we’ve compiled returns data based around above-average trading volume days; more specifically, we took a look at how the S&P 500 Index, as represented by States Street’s (SPY, A), performed on days when trading volumes exceeded the trailing one-month average by 50%, or a factor of 1.5x.
Since the start of 2010 through the end of 2013, there have been 67 instances where SPY’s daily trading volume has exceeded its one-month average by a factor of 1.5 times or more. When considering the above returns, it’s important to recognize the inherent bullish bias seeing as how the sample period spans the current bull market only [see 101 ETF Lessons Every Financial Advisor Should Learn].
Putting It in Context
Although above-average trading volumes don’t appear to offer tremendous predictive power at first glance, they do hold some valuable insights when interpreted in the right context. Firstly, note the market’s tendency to deliver positive returns in the five days following a volume spike; while every instance of this can be further scrutinized under a microscope, the general takeaway here is that above average trading volumes tend to attract more buyers in the days immediately following.
Second, and more importantly, it’s important to consider the markets’ performance leading up to an above-average trading volume day. If we only consider the volume spikes when the market was up more than 5% during the prior 20-day period, which has happened three times in our sample, we see that the 20-day returns after the spike are considerably smaller than the average 20-day return following all volume spikes [see also How To Swing Trade ETFs].
Even more noteworthy is the market’s average 20-day returns following a spike after being down more than negative 5% in the prior 20-day period, which has happened in 12 instances; this observation showcases the fact that the market has a tendency to deliver more positive returns following a surge in trading volume if it has been beaten down in the days leading up to the volume spike.
The Bottom Line
Above-average trading volumes hold little predictive power over future returns for the S&P 500 Index ETF. For the most part, above-average trading volumes tend to attract more buyers than sellers in the immediate days following. However, the longer-term trend that plays out after the above-average trading volume comes is more related to the markets’ performance leading up to the volume spike day. In other words, if the market has been falling for the past month and there is a volume spike, there is a much greater probability that the market will turn in more positive returns over the following month than if the market had been rallying prior to the volume spike.
Follow me on Twitter @SBojinov
Disclosure: No positions at time of writing.