Stock Trading Strategies

The Trader’s U E-Letter: Issue # 172

Wednesday, January 25, 2006

Stock Trading Strategies: It’s No Myth — Round Numbers DO Matter… And Can Give You An Edge

By D.R. Barton, Jr. Chairman, Trader’s U We’ve done our share of myth busting here at Trader’s U. We’ve busted myths about short selling, buy-and-hold investing, and the January Effect.

And just yesterday, I got into a discussion about the “myth of round numbers” with a new friend and trader from Indiana who you’ll hear more about in the future.

Round numbers are, in general, numbers that end in zero. And there are many variations on what different traders and investors classify as round numbers.

The myth is this: “Prices gravitate toward or have special respect for round numbers.”

Today, I’m proud to report that this is no myth! Numbers do actually move toward and/or respect round numbers.

And while that information may be useful at your next cocktail party, let’s find out how you apply this knowledge to your stock trading strategies and put some extra money in your investment portfolio.

Cross-Eyed For the Sake of Knowledge

How can we know if prices tend to move toward or stop at “round numbers”?

First, we need a definition of “round numbers.” There are a number of ways to look at these. My earlier definition of “anything that ends in zero” is a good starting point. (We’ll concentrate on stock prices here, but I have seen data that show similar effects for futures and currencies.)

I landed at this question: “How often do prices stop and reverse at a 10-cent or dollar increment versus what we could reasonably expect statistically?” I had the idea to use weekly data — the high price and the low price for a given stock each week. If the high and the low prices hit exactly on dime or dollar increments more than an average amount, then this “round number” myth could be real.

And then I threw myself physically at the problem. Pouring over literally thousands of data points, I sacrificed my eyesight just for the sake of you, the loyal readers of Trader’s U. But enough about my selfless acts. Let’s get to the research…

Compelling Evidence That Round Numbers Matter in Your Stock Trading Strategies

So here’s the hypothesis: If a weekly high or low fell on a dime increment more than 10% of the time (if prices are purely random, then the highs and lows should fall on 10-cent increments 10% of the time), then we have a good chance that traders favor round numbers. Likewise, if weekly highs or lows fall on dollar increments more than 1% of the time (random prices should land on whole dollars only 1% of the time), then prices really do favor round numbers.

To do the trading research, I looked at 250 weeks’ worth of data for each stock. Decimalization — the pricing of stocks in one-cent increments — became a regulatory requirement almost exactly 250 weeks ago. So we have 250 weeks of good data where prices were quoted in terms of pennies, not in eighths or sixteenths. Since I looked at the high and the low for each week, that gave me 500 data points per stock, which should make a statistically significant sample.

I choose two heavily traded NYSE stocks and two heavily traded Nasdaq stocks so that pricing could not be influenced by light trading volumes. The results: