On Sunday, I looked at turnover rates for every year in the NFL since the merger. Today, I want to re-examine turnover data but in a different light. In 2014, the average team committed 23.7 turnovers. As you might suspect, there’s a strong relationship between turnovers and winning percentage, with a correlation coefficient of -0.56. This says nothing about causation, of course, and the causal arrow does in fact run in both directions (committed fewer turnovers leads to more wins, and winning in games leads to fewer turnovers).
Here’s another way to think about the relationship between winning percentage and turnovers. The Patriots were responsible for 4.7% of all wins this year and committed 13 turnovers; as a result, when calculating a weighted league average turnover total, I made New England’s 13 turnovers worth 4.7% of that total. Meanwhile, the Buccaneers and their 33 turnovers were only worth 0.8% of the weighted league average turnover total, since Tampa Bay was responsible for just 0.8% of all wins.
Using this methodology, the weighted league average turnover total in the NFL was 22.5 per team, or 95% of the unweighted league average. I used that same methodology to calculate the percentage of “weighted league average turnover total” to “unweighted league average turnover total” for each year since 1960. In the graph below, the blue line represents the NFL ratio, while the red line represents the AFL ratio.
I believe that possession of the ball has never been more important in the NFL than it is today. Given how efficient teams have become at scoring, simply having the ball is extremely valuable. Think back to the early days of the NFL when teams were not very good on offense; the NFL truly was a field position game at that time, and teams would even punt on 3rd down in order to flip the field. In a game that might end 6-0, possession wasn’t quite so important. In today’s NFL, it’s vital, which is why coaches who are passive on 4th down are more harmful than ever.
As possession becomes more important, you would think that the ratio depicted in the above graph would be on the decline, and that’s somewhat true. The effect isn’t all that strong, and part of that is because turnovers have always been correlated with winning. But there has been a noticeable trend downwards since the 1978 rules changes. Curiously, that doesn’t continue once you go back to the ’60s, but scoring was higher back then, too. Back then, turnovers weren’t quite as important because just about every team was operating with a high-risk/high-reward passing game. Joe Namath and Daryle Lamonica were aggressive passers, and back then, a few interceptions were the cost of doing business for 50-yard touchdowns. In the short-passing, low-risk, horizontal offense of today — one that is highly dependent on YAC — things are very different.
Anyway, those are just my thoughts from an interesting graph. What do you guys think?