New Smart Money article

This month's issue of Smart Money has an article which references The Little Book That Beats the Market. This is significant to me, since that's where I learned about it in the first place. Jack Hough, the Stockscreen columnist, comments on how good his picks did from the 3/06 issue (it actually came out in February. I know because I was tracking those picks, which were based on January stock prices). They returned 33% vs. an S&P return of 22%. The main thrust of his article was against diversification, saying you should just pick 5 - 7 "good" cheap stocks and you won't be dragged down by the losers you might have in a diversified portfolio. I'd link to the article, but it's not up yet at SmartMoney.com. He suggests using the Magic Formula to select the stocks.

There are some significant problems with his article. The vast majority of investors are not smart enough, lucky enough, or do not have enough time to really identify which are the ones that are most likely not to decline. And, my own tranche investing has proven that there are plenty that decline. Fortunately, in the top 25 there are a lot that go up significantly as well, including ones that more than double (PNCL, WNR). But I've also had some that have lost half of their value. Could I have avoided those with some research? Maybe...but I also would have missed out on some of the winners also. Look at WNR even now. It's up 154% since September 2006. Would that seem like a prime candidate to buy into now after that run up? Yet many people, and the MFI website, still say it's a buy.

Or let's take a big loss, MTEX. I haven't sold the position, but I'm sitting on a 41% loss since May. Would I have had any clue that the Texas AG was going to sue them for improper marketing techniques? Maybe, but I doubt it.

I'm willing to say that if this was my career and all I did was track a few companies, I would be able to do well, but even then I would be exposed to much more risk in such a non-diversified portfolio. Joel Greenblatt says that if you can closely identify a smaller number of good companies, you can drop the number of companies you invest in, but I hold that most people can't do that, which is why I think this was an irresponsible article.

Other issues the article doesn't address:

1. When should you sell? The concept is still value investing, so when you hit the point that the stock is "fairly priced" do you take your gains?

2. Beyond simply running a stock screen (such as MFI), how do you pick the 7 best? No discussion of this at all.

It was a highly simplistic article, and quite a shame since I usually find that column fascinating. I'll stick to diversification.