I’m undertaking a 1000-day reinvention project, blogging here daily to track my progress. In Tuesday Book Club, I share an idea from a book.
From William J. O’Neil’s How to Make Money in Stocks: A Winning System in Good Times or Bad:
To detect a market top, keep a close eye on the daily S&P 500, NYSE Composite, Dow 30, and Nasdaq Composite as they work their way higher. On one of the days in the uptrend, volume for the market as a whole will increase from the day before, but the index itself will show stalling action (a significantly smaller price increase for the day compared with the prior day’s much larger price increase). I call this “heavy volume without further price progress up.” The average doesn’t have to close down for the day, but in most instances it will, making the distribution (selling) much easier to see, as professional investors liquidate stock. The spread from the average’s daily high to its daily low may in some cases be a little wider than on previous days.
Normal liquidation near the market peak will usually occur on three to six specific days over a period of four or five weeks. In other words, the market comes under distribution while it’s advancing! This is one reason so few people know how to recognize distribution. After four or five days of definite distribution over any span of four or five weeks, the general market will almost always turn down.
Here’s a chart from February. The thick part of the candles for each day show the opening and closing prices for the day. A green candle indicates close higher than open, and the open price is the bottom of the body. A red indicates the opposite — close lower than open, and the closing price is at the bottom. The skinny wicks indicates the highs and lows for the day.
This chart shows heavy selling on February 5th, where the price drops significantly on much higher volume than the 4th. On February 12th there’s another distribution day with a large red candle.
But this chart shows the S&P 500 already in a downtrend. It’s trading below its 20-day simple moving average, which is sloping downward. There’s no top signal forming here because the top signal must have already occurred, during the market’s earlier uptrend.
O’Neil suggests quick action if you see indications of a market top:
After you see the first several definite indications of a market top, don’t wait around. Sell quickly before real weaness develops. When market indexes peak and begin major downside reversals, you should act immediately by putting 25% or more of your portfolio in cash, selling your stocks at market prices.
Yes, he’s encouraging market timing. You can do market timing if you have a way to get back in. He writes, “Don’t ever let anyone tell you that you can’t time the market. This is a giant myth passed on mainly by Wall Street, the media, and those who have never been able to do it, so they think it’s impossible.”
I’ve raised a large amount of cash in my portfolio. I am still seeking opportunities in energy stocks and a few other areas (utilities, agriculture) but mostly waiting for the market to decide if it’s going to enter a real correction or crash or start a solid advance again.
How do you know when to get back in?
O’Neil has his own way of determining if a market rally during a downtrend won’t turn into a sustained uptrend:
You’ll know that the initial bounce back is feeble if (1) the index advances in price on the third, fourth, or fifth rally day, but on volume that is lower than that of the day before, (2) the average makes little net upward price progress compared with its progress the day before, or (3) the market average recovers less than half of the initial drop from its former absolute intraday high.
More specifically, O’Neil looks for a follow-through day. First, you need to see a major market average (e.g., S&P 500) close higher after a decline that happened either earlier in the day or during the previous session. This is the first day of the attempted rally. Then, on the fourth day to seventh day of the attempted rally, you want to see one of the major averages follow through with a big gain on heavier volume than the day before, 1.5% to 2% or more. O’Neil writes “the most powerful follow-throughs usually occur on the fourth to seventh days of the rally.” You need to wait after the rally begins to see if it is real not just a transitory bounce.
There are a few more criteria he suggests looking for. You want to see a broad-based rally, not one driven by just a handful of stocks or sectors. You want to see continued strength in the days that follow. And, according to his trading philosophy, you should always be selective with your stock picks, buying only those that show chart patterns that he suggests indicate stock price increases ahead.
Also, watch for distribution days on the days after the attempted rally first starts. A distribution day (lower price on high volume) on the second or third day of the rally means the rally fails almost all the time, so don’t keep watching for big advances on days four through seven if the you see poor performance on days two and three.
Yesterday the Dow was up 0.83%, the S&P 500 up 1.01%, and the NASDAQ up 1.22%. Today futures are negative, but we’ll have to wait and see what happens to decide if this ends the rally attempt with a distribution day.
Getting back in
The S&P 500 is still above its 200-day MA, so by at least one of my market timing rules, a real rally could mean getting right back in. But I’m thinking of following some modified version of O’Neil’s approach, where he buys stocks that show strong chart patterns indicating future gains, and then selling once a stock has moved up 20 to 25% (or selling if it declines by 8%, as a risk management practice). This might be called swing trading, where you aim to profit from price swings in stocks that play out over several days or weeks.
O’Neil views the boring market we have right now — not advancing, not really declining — as a time when the best stocks are preparing for future runups. So what I’m doing is constructing a watchlist of stocks I might want to buy, if they show me I should.