I use this same system (got from a book in '83 or '84 using the USFL) along with another system based on the all the teams strength of schedule along with my ratings that I adjust weekly.
I saw the same flaw you mention. So, if my strength of schedule system show 20-17 and the YPPT system shows, for example 31-14, I know that team won with turnovers, kick returns, etc. and maybe the other team lost because of them. That's why I went with Oakland over Philadelphia 2 weeks ago. That's my Rule 13 play. Hits about 65%.
This system predicted that Oak would cover the 14. What is your point?