Creating turnovers is usually more random than fans realize.
Turnover Margin
Sit through five insufferable minutes of NFL coverage and you’ll hear somebody harp on the value of turnovers at least once. They’re important, and good teams do tend to win the turnover battle, but there’s also a certain amount of randomness and variance from year to year. Most notably, teams at the extreme ends of the turnover spectrum are unable to sustain that level in consecutive campaigns.
How did it project 2013? Very well. The two teams that stood out like a sore thumb at the bottom of the turnover rankings were the Chiefs and Eagles; they were each at minus-24 in 2012, and nobody else was worse than minus-16. In 2013, much of their success was driven by a dramatic shift in this metric. The Chiefs had the second-best turnover margin in football at plus-18, while the Eagles were fourth with a plus-12 figure. Kansas City’s improvement of 42 turnovers is the most anybody’s improved in a single season since the strike year of 1987. The Jets (minus-14) and Lions (minus-12) stayed roughly the same, but the Cowboys and even those absurdly lucky Colts — granted, in the one category where simple regression might have expected them to improve — saw their turnover margin improve by 16 or more from 2012 to 2013.
The five teams at the top of the turnover charts were unable to repeat in 2013, each falling off by at least 15. That included the Patriots, Bears, Giants, and Falcons — but the most interesting case was Washington. A huge chunk of its 2012 success came by avoiding turnovers on offense; Washington turned the ball over a league-low 14 times in 2012. It hit that figure by Week 8 of 2013, and by the time Kirk Cousins had inflated his trade value in December, the D.C. team had turned the ball over 34 times, tied with the Lions for the second-worst rate in football.
I could look up the top teams of 2016 and compare where they ended up in 2017 but I'm too damn lazy.
http://grantland.com/features/nfl-stats-predicting-success/