The limits of "buy what you know"

Peter Lynch famously advised investors to "buy what you know". For most of us, that generally means looking at companies that produce the consumer products we buy, but for those with specialized knowledge, it opens up other possibilities. One challenge with that is that the company with the best product isn't always the best investment. I was reminded of that by an e-mail this week from occasional commenter Y./The Rivers.

A fellow shareholder in Alloy Steel had mentioned another holding of his to me, a development stage biotech company, so I asked Y., a microbiologist/biotechnologist, his opinion of it. After offering his assessment of the company, Y. added this cautionary note,

I'm a pretty bad judge of stocks, particularly in the biology realm. a few years ago, I bought some shares in Affymetrix (AFFX) because they had the best microarray platform hands-down, and were about to acquire a big new customer (Merck). Then they made some bad business moves and got way behind the up-and-coming tech (next-generation sequencing, NGS). Their main competitor, Illumina (ILMN), with a much inferior microarray platform, did everything else better, and is now the biggest NGS company, I think. If you compare their stocks over the last few years, it's quite disparate.