That's a link to my CAPS blog from about 2 years ago - 8/10/10. On that date I sort of shifted my CAPS strategy. I realized I wasn't going to get anywhere pursuing more accuracy points - with a 77% accuracy, I was almost 'topped out' in terms of the percentile rankings - and in order to start coloring up my hat, I was going to need score points.
I started spending a lot less time on looking for junk to short. Not that there isn't plenty, always - look back through my last year of picks to see some obvious tech names. But, instead, I started putting some effort into finding some value plays - good businesses, no debt, growth prospects, priced 15% under - what? Well, I'll be honest. Not 15% under "what I thought the company was worth." 15% under what the market was valuing similar stocks with similar prospects. The process is something I call 'turning over rocks' - eventually you find something interesting under there.
Mission accomplished: my score is +3000, my rating is 99.55, up from 97.46 two years ago. Gained about 2500 score points; lost 3 accuracy percent. It's hard to find good stocks; sometimes when the market undervalues companies, the market is actually right.
The latest triad of stocks I am thinking about are INTC, CSCO, and PM.
PM is the ultimate widget seller. They have a product that behaves like a widget - demand for it is hugely inelastic, supply is extremely consistent and predictable - and they are quite efficient, due to decades of practice and good management practices - at distributing it across the world. Despite their huge cash flows they keep their operation streamlined and efficient - they take a certain pride in it.
INTC is the opposite. They are best of breed in their field, not by saturating the market with advertising to create a false sense of their product's superiority as PM does; but, instead, the reason they are best of breed is that their capital expenditures for R+D exceed a medium sized country's GDP. They manage and execute impeccably - but in a different way from PM. They're not building widgets. They're crafting masterpieces, and they roll out a new set of masterpieces, predictably, every two years. No one can keep up; no one has ever kept up since 1976. If you are a consumer and you need processor, sooner or later, you will consume Intel. I don't think the top brass at INTC ever worries much about a percent or two of margin; I think they worry most about keeping the masterpieces flowing. And it shows.
CSCO is an interesting company. I am going to sum them up very concisely: they are, in reality, a widget company that is built on the INTC model and still thinks they're crafting masterpieces. The recent kerfuffle where they hacked their routers to try to enforce their customers to sign on to their information-snooping cloud service is the best case in point. The consuming public won't let CSCO, a commodity maker of hardware, pull this stunt. They want, instead, to install CSCO commodity hardware in their homes, businesses, green boxes, and data centers, and then let people connect AAPL, GOOG and MSFT products to those routers and let THOSE companies snoop their data. But CSCO is not permitted in that club. I would short them to death from here.