I like the format and concept of this book, and I purchased this book (which I regret) however I have some major complaints:
1) Chernev's analysis can be bad or even awful. I worked through his games while running Stockfish 2.1.1 in the background crunching numbers, and Chernev's advice/insight is frequently awful and just plain wrong, steering you into believing bad blunders that lose pawns or pieces are merely just less optimal moves (or even the most optimal response). And then the moves he recommends for the losing party as best, the computer given minutes to think says would also be blunders and frequently has an alternative not in the notes that would get them out of the jam unless the winning party kept up with absolutely ideal play. I've stopped being able to trust his notes.
2) The games appear to be between masters from the 1890s to the 1940s and amateurs or club players, rather than between two masters. So one side tends to really dominate, and has more precise play that the computer tends to agree with. The other play tends to constantly do something different than the chess program recommends, always to losing results. I am not sure how educational it is to watch such lop-sided contests, especially with frequent bad advice from Chernev when checked against a chess program.
3) The openings used are not common ones seen today. So forget Sicilians and King's Indian Defense, you just get a lot of Giuoco Piano's and QGD.