Nick Karad's  Instablog

Like many others, I enjoy watching Jim Cramer's Mad Money show on CNBC and try to catch it whenever I can. He's a terrific showman, and regularly accomplishes the difficult task of making stock trading interesting for Main Street. Although some may question his investing skills and especially his recommendations, many of the underlying explanations he provides are both accurate and informative.

When asked about entry and exit points on the show, Cramer regularly repeats the following standardized approach: he suggests finding a stock that is on a clear uptrend, and then setting up smaller trades around a core position. So for example, you could buy 300 shares of a stock 'X'; whenever the price of X goes down by 10% (or some such fixed percentage), you buy another 100 shares, and when it goes back up by 10%, you sell 100 shares to take some profit off the table. This allows the trader to make some smaller, profitable trades on the side, while the core position slowly appreciates.

This approach made sense to me, so I decided to try it on a couple of positions, and discovered a serious flaw in this approach. Or perhaps it needs some modification that Cramer has neglected to mention on the show, a recovery action in the event of an unlikely but nevertheless very real possibility.

The flaw is this: buying on the dips and selling on the bumps is all very well as long as the underlying stock price continues to rise. What happens, though, when the stock flattens out or completely changes direction, even as you're buying on the dips? It's the old fallacy of trying to catch a falling knife!

This is exactly what happened to my test cases. I tried Cramer's approach recently with Dollar Thrifty Automotive Group (DTG), which has been on a nice uptrend since mid-April. I opened this test-case position in September, and at first, the scheme worked beautifully. Each time the stock dipped, I bought some shares; each time it increased over the threshold, I sold some shares. I took profit multiple times.

But then something changed: at the beginning of November, DTG went low, beyond the expected level of support, and then stayed low all through the month of November. For the first couple of dips, I followed the strategy faithfully and bought more - but then got stuck as it stayed down and even went down further.

Finally, good money management principles prevailed, and I decided to put a stop-loss below the next threshold. In the meantime, luckily, the trend reversed course and the ensuing uptrend rescued me.

I saw another example of a stock abruptly changing direction: Alpha Pro Tech (APT), which was on an uptrend since June. There was nice money to be made buying when it went low and then selling when it went high. But the last time it went down at the beginning of November, it stayed down; again, I was stuck and the losses started to mount, albeit slowly.

On balance, I think I favor the Lefevre approach better (from the classic book "Reminiscences of a Stock Operator"): once you identify a stock with a long-term uptrend, buy it progressively as it goes up, but with reasonable stops underneath appropriate support levels. If the trend suddenly changes, you automatically exit with a profit; if it keeps going up, you continue to load up and build more profit.

This approach certainly saved me last year, in September '08 - as the whole market collapsed, my stops got hit one by one and I got out of several positions with relatively minimal impact, considering the scope of the market decline.

To summarize, Cramer's approach of buying on dips works well as long as the underlying stock doesn't suddenly give up its existing uptrend.



Disclosure: Long DTG, Long APT at the time of writing