≡ Menu

NFL Average Plays per Team Since 1950

In light of the Patriots nearly breaking the NFL record for plays, and the promise of up-tempo offenses in Philadelphia (under Chip Kelly) and Denver (Adam Gase), it’s easy to think that the number of plays run per team is about to reach historic levels. But that seems really unlikely.

The graph below shows the number of plays run per team game for each season since 1950. The total number of plays run is in blue for the NFL; I also added the same data for the AFL in red. As you can see, the number of plays per team game has been relatively steady over the last 64 years, but the high-water marks were the early ’50s and most of the 1980s.

In addition to plays run, the graph also shows:

  • The number of rushing plays per team game in green, a number that’s obviously on the decline.
  • The number of completions per team game in black, which has risen as the number of runs has declined.
  • The number of incomplete passes per team game in orange. Incomplete passes stop the clock, so I thought we might see something interesting there. How’s this for trivia: there were 13.5 incomplete passes per team game in 2012, the same number that existed in 1948! While 1948 is off the graph, you can see that the number of incomplete passes per game has been remarkably consistent throughout NFL history. In fact, the average from 1950 to 2012 is 13.5 incompletions per game, and the league average was 13.5 +/- one incompletion in over 80% of the seasons since 1950.
  • The number of sacks per team game is in purple, a number that has also stayed very consistent over time. Only three times since 1950 has the league average been less than two sacks per game or more than three sacks per game.


  • Richie

    The incomplete passes data is fascinating. You can see a little uptick in the 90s, which makes sense because I think teams were passing more in the 90s but hadn’t yet gotten to the point where QBs were getting really accurate.

  • Danish

    So if plays, sacks and imcomplete passes have stayed the same, running plays have been replaced by completions and completions only, right? No wonder offenses are getting more efficient. I took a course in timeseries analysis once. It seems like the variation in completions would be well explained by the change in runplays – perhaps with a lag of a year or two. I wish I hadn’t forgotten everything i learned during that course.

    The spike in completions starts exactly in 1977, which was the year the Mel Blount rule was introduced. Surprised to see such an immediate adjustment in playcalling. Was there a similar rule change around ’92?

    • Pyper

      The rise in completions seems to match up well with the spread of Bill Walsh’s West Coast offense which put an emphasis on the short “efficient” passing game. These days, all offensive systems, whether they’re west coast or not, feature a strong emphasis on the short passing game.

  • Jason

    @Danish – In 1993, the dead-ball-to-snap play clock was reduced from 45 seconds to 40 seconds (the “post-administrative stoppage” play clock remained at 25 seconds following the placement of the ball). I think that would account for some increase in total plays.
    Curiously, I remember many arguing at the time (and a few years thereafter) that they shorter play clock resulted in less total offense and less scoring, because teams didn’t have as much time to get the play called and do pre-snap adjustments. No idea if any data bears that out.

  • NOS 1.0

    Fascinating. Great analysis!!!!!!