
It's amazing how Graham's methods were derived almost half a century ago but most of his predictions were correct and direct parallels over what he saw during the Great Depression occurred during the tech bubble. Graham goes for reliable growth over long time scales so you'll never pick multi-baggers or hear about your stocks in the news.
My takeaways are to look for stocks that are undervalued (ie having a low Price to Book value) and show consistent growth over a number of years. Investing with Graham's methods would have you avoid high growth (high P/E) stocks like Yelp in exchange for slow growers like Caterpillar. I could only imagine what he'd think of Amazon.
Ensco (ESV) and Rowan (RDC) are two stocks I picked using Graham's methods. At the time Ensco had a P/B of 0.93 (below 1 is what you're looking for) and a P/E of 8.33 and Rowan had a P/B of 0.79 and a P/E of 15.32 (below 20 is preferable). These stocks have actually turned out to be very profitable in the short time I've held them and I'm only now starting to hear them mentioned in the news.