Technical Perspective: A Very Overbought Market

By Carl Swenlin, Decision Point

One of the general rules of the stock market is that things will get as good (or bad) as they can get, then prices will start moving in the other direction. This is another way of describing “regression to the mean”.

This is the reason that we technicians have our indicators — so that we can get an idea when conditions have reached extremes that could cause prices to start moving in the opposite direction. One of the indicators I like is the Percent of PMOs (Price Momentum Oscillators) Above Zero because it is smoother and has less noise than other intermediate-term indicators.

The second panel on the chart below shows the PMO for the S&P 500 Index, which is the price index just above it. The bottom panel shows the percentage of individual S&P 500 stocks that have PMOs above the zero line.

As you can see, the indicator has recently topped at a very overbought level. In similar cases noted on the chart, half were absolute top pickers, and, while the other half announced an internal peak in strength, they arrived well ahead of the price peak. But, even though they were early, the indicator peaks in early 2011 and 2012 were ultimately followed by price declines that sent prices lower than they were when the internals peaked.

While indicators may be topping in very overbought territory, a price top is not guaranteed. Nevertheless, conditions are less than ideal for making new commitments to the long side, and increased caution is warranted.


Got a comment or question about this post? Feel free to use the Ask Cullen section or leave a comment in the forum.

Decision Point provides investors with stock market indicators and timing tools. Our charts and daily reports are organized to help make a quick assessment of market trend and condition, and to quickly identify trading and investing opportunities.

More Posts - Website


  1. Of the three previous peak points highlighted with the red arrows, the indicator was useless in pinpointing a significant top. Like most indicators that I have extensively studied, this one is right 50% of the time.

  2. As with all indicators showing overbought and oversold – you use them as one piece of the puzzle. I watch for topping patterns, consolidation, indicator divergences, volume and other things to time a reversal. So when combined it can be better than 50/50.

  3. In other words: sometimes it works – sometimes it doesn’t. So what’s the point? Valueless. Excuses, like those provided by a commenter above, illustrates the point. If we say “they are a piece of the puzzle”, then we can just be subjective about it and make up whatever we want. T/A is useless. I studied it and used it for decades. I know. We retrofit everything because we want to believe. Like the Elliot Wave loonies. You see what you want to believe. If T/A provided any edge, how long would it take for everyone to use it and render it useless? Yet, its disciples believe. It is religion. Not science.

  4. “Technical analysis is a windsock, not a crystal ball.”
    — P. Arthur Huprich

    “All through time, people have basically acted and re-acted the same way in the market as a result of: greed, fear, ignorance, and hope – that is why the numerical (technical) formations and patterns recur on a constant basis.”
    — Jesse Livermore, “How to Trade in Stocks”

  5. T/A is extremely useful. I’ve studied it and have used it for decades. I know.

    T/A provides an excellent edge, when used properly and with many other tools, including risk management. It is science, not religion – except for those who can’t or don’t know how to use tools correctly.

  6. Any attempt to use TA to predict the future will result in failure. I can’t predict market direction, but I can find good r/w prospects within current direction using TA. It works especially well as a steering mechanism. Around Jan 1, I was about 20% invested and due to the quantity and quality of bullish set ups, that number grew to 50%. In the last couple weeks, due to regular profit taking,an increase in stopped out positions, and a lack of quality breakouts, the number is about 25%. If the market continues to deteriorate holdings will naturally drop and bearish set ups will come in to play. Simple. Not clairvoyant.

  7. That chart would be a lot more engaging if you stretched it to a 10-15 year perspective, at least. If the indicator is “early” it’s nearly useless from a practical viewpoint. So, I’m not very excited by the little that you’ve shown me. That said, you can make a strong technical case to suggest we’re likely to see a significant correction and I wouldn’t disagree.

  8. Romeo, and great info. How are you running those back tests if you don’t mind me asking?

  9. Sentiment is what matters, and right now Investment newsletters are the most bullish in 13 years (since 2000) and advisor sentiment has been >51.1% bullish for over 5 weeks, with bears under 23% (and now <21%).

  10. ThomsonOne allows you to export chart data to Excel, which is a useful tool. More often than not, I’m using a Bloomberg terminal to find data, but for less arcane studies, Thomson One is good.

  11. There’s only one arrow that could be called a failure (the second one), meaning you couldn’t buy back at lower prices within a few months, all others did give you a lower entrypoint

  12. I just ran a 1 month backtest on periods when the S&P is up 11% in a 60 session period and found the following. The S&P is up 64% of the time and generates a 1.3% return in the coming month. Not as conclusive as yours, but not favoring the contrarians as much as I might have expected….

  13. It’s interesting, isn’t it!? What’s even more interesting, methinks, is that te same counterintuition doesn’t apply to oversold markets–at least, the outcome is far less homogenous per historical precedent.

  14. As with ALL technical models – they work only to the extent that very few people know about them and use them.

    Most new techniques back test well – as they were not in use when that data was being accumulated. Back testing is a fools game!

    If the model had been in existence and widely used during the time those market data were being accumulated – the data would have looked far different.

    Churn baby, CHURN!

  15. I have been using John Bogles simple conclusion regarding overall time trends. Sell when you are up 10%, buy when they are down 20%. It has worked well enough for me, but I am not claiming superior results, just don’t be greedy.