Interesting article about handicapping football games. The guy is spot on with regards to the fact the turnovers are sometimes random events that need to be ignored when handicapping a team, but often the turnovers are not random and must be considered in handicapping.
However, he makes a mistake in his last paragraph. He tries to prove that turnovers are leading indicators but only shows that turnovers in the past are highly correlated with past performance. DUH! He needs to show a point in the season where Team A lead the league in turnovers and then show how they performed ATS (that's against the spread in case you didn't know) from that point forward.