“Overbought!”

“Oversold!”

There’s not a day that goes by without a pundit using one of these powerful words to support an existing bias. Most interesting, some pundits use the term “overbought” to further a bullish outlook while others use it to justify a bearish outlook. The same is true with the term “oversold.”

Unfortunately, there’s never a definition of overbought/oversold provided by these pundits and no analysis accompanying these predictions. They are merely forms of entertainment, but regrettably are not treated as such.

The truth: there’s almost an endless array of ways to define a market as oversold or overbought using an endless array of indicators and an endless array of time periods. If you analyze these variations going back in time, each will have a different outcome.

In the past few months, we have seen both extreme oversold and extreme overbought conditions, a function of the violent sell-off in the first few weeks of January (worst start to a year in history) followed by the equally violent advance that took place from mid-February through the end of last week.

On January 20th, the day the S&P 500 hit an intraday low of 1812, I tweeted that the percentage of stocks above their 50-day moving average had moved down to 9% (an extreme seen less than 2% of the time going back in history). You tend to see bounces from such extremes. This is on average, of course, for there is no holy grail in markets. (Note: many a perma-Bear used my tweet to argue that “breadth was weak” and therefore the market was headed for further declines.)

By February 22, the market had rallied a bit and another indicator, the McClellan Oscillator, was flashing an “extreme overbought” reading (seen less than 2% of the time going back in history). I tweeted out the following chart and noted in a subsequent tweet that you “tend to see strength” (over 1-12 months) after such readings. The S&P closed that day at 1945. (Note: many a perma-Bear used this tweet to argue that the market was extended, “due” for a pullback, and soon headed for lower lows.)

During the next few weeks, the S&P 500 would continue to move higher. On March 16, the S&P 500 was recording another “overbought extreme” reading, this time in the percentage of stocks above their 50-day moving average (for our new research on moving averages, click here). It had moved above 90%, a long way from the opposite extreme seen in January. Interestingly, I noted, you tended to see above-average forward returns (1-12 months) from such extremes in the past. The S&P 500 closed that day at 2027.

(Note: many a perma-Bear used this same chart to argue that the market was extended, “due” for a pullback, and soon headed for lower lows.)

Fast forward to today and the S&P 500 has moved up to 2072. Over 93% of stocks in the S&P 500 are now above their 50-day moving average, one of the highest levels in history. (Note: many perma-Bears are shouting once more that this is an omen of impending doom.)

Gurus, Probabilities and the Great Paradox in Markets

What will happen from here? I can’t tell you that; in fact, no one can. While I play one on TV at times, I am not a Guru – far from it. Indeed, there is no such thing when it comes to investing because we are forever dealing with probabilities, never certainties.

When I wrote the “Silver Linings Playbook” post at the February lows, I did so not because I possess some magic crystal ball but because the probabilities (looking at extreme bearish sentiment and oversold indicators) favored a bounce. The S&P 500 has since rallied 15%, but that outcome was far from a certainty.

Anything could have happened at the February lows and anything can happen today. The best we can say is what is most likely to happen, and that those odds are forever changing. While not nearly as entertaining as a “call’ with an exact time and price, it is more likely to be correct.

If I am “right” more often than your average pundit in forecasting, then, it’s not because I can predict the future. It is only because I am more apt to let the probabilities guide my outlook than some agenda or bias. And I’m less likely to make an extreme, outlandish black swan prediction with little chance of success.

But I digress.

When the market has been this extremely overbought in the past, we’ve seen above-average returns going forward. Again, this is on average, which is very different than always.

If it doesn’t sound intuitive that an “overbought” market would be bullish, that’s because it it’s not intuitive at all.

It is one of the great paradoxes in markets that both extreme oversold and extreme overbought conditions can be followed by above-average returns. Why? Because extreme strength begets strength (momentum) while extreme weakness does the same (mean reversion/value). Momentum and Mean Reversion are the most powerful forces in markets, and exist because investor overreact and underreact to new information, again and again.

But just because they are powerful, does not mean they are always “working.” As I wrote about value and momentum in the past year, they are cyclical and prone to inevitable periods of underperformance. The same is true of overbought/oversold indicators, sentiment indicators, and any other indicator you can think of. There is no holy grail in markets with perfect foresight, only probabilities. The sooner you can accept this truism, the sooner you can begin your journey down the road to investment success.