NFL Offenses Facing 5-Year Downward Spiral on Third Down

By Scott Kacsmar

Cold Hard Football Facts’ Comeback King (@CaptainComeback)

It is one of the greatest cases of “perception vs. reality” in the NFL today. The game of football feels so much more catered to the offense, but the data says scoring is not significantly higher than it was in seasons before the Super Bowl even existed.

In trying to pin down some of the areas of the game that might show an increased advantage for the offense, let’s focus on third down conversions.

Of all the records set last season, one that was overlooked was the team offensive record for highest third-down conversion percentage in a single season. The New Orleans Saints converted 118 of 208 attempts for a rate of 56.73 percent, which tops the previous record (56.15 percent) set by the 2006 Indianapolis Colts in their Super Bowl season.

That was great by New Orleans, but how about the rest of the league? It turns out despite all of 2011’s absurd offensive numbers, offenses converted third downs at the lowest rate since 2004.

To this point in 2012, offenses are below last year’s rate. If it should continue, this will be the fifth consecutive season in which the league-wide third-down conversion rate decreased. So much for the offensive domination, right?

Settle in for some cold, hard data mining.

Third Down League-Wide Averages

The importance of third downs is unquestioned. It’s a drive extender. Teams who convert a high rate often score a lot of points. The correlation between third-down conversion rate and points scored is strong at 0.67 from 1991 to 2011.

Third-down data only goes back to 1991, which is a pivotal year for the NFL in terms of what data they began publishing in mass volume. That would include things such as time of possession, fumbles, and fourth-down attempts. The NFL “officially” kept track of third downs starting in 1981.

These numbers are for the regular season only. You can see the change in conversion rates from year-to-year in the final column. The 2012 season is updated through the Arizona/St. Louis game in Week 5.

Some people prefer charts, so here is one of those.

The yearly rates make for a fun looking rollercoaster ride, but we are actually staying pretty close to a 36-40 range, so the peaks and valleys are not what they seem.

The average is 38.26 percent. The range of conversion rates has a minimum of 36.46 percent (1999) and a maximum of 40.13 percent (1995).

It is likely no coincidence the season with the most pass attempts per game in NFL history (1995) is the season with the best third-down conversion rate, as third down is often dominated by the passing game. Other than 3rd and 1, teams would rather throw the ball in these situations, even if other research definitely suggests 3rd and 2 should be a running down as well.

You start getting a steady climb after 2004, which was the season the league put an emphasis on illegal contact in the passing game. The passing numbers also exploded in 2007 without any major rule changes that would have caused them.

That is where the third-down rates peaked in recent years. Ever since 2008, it has been a steady decline each season, which just has not been the case going back to 1991. Even though offenses appear to be more efficient in the last few prolific years, it has not shown up on third down.

Is a 0.20-0.35 drop in percentage points significant? Not so much. If the offenses in 2011 converted just 16 more third downs (0.5 per team), then they would have had the same rate of conversion as they did in 2010.

Extending 16 drives might only produce an extra 28 points in a season, or less than a tenth of points per game for the league.

But in 1996 when the rate fell by nearly 1.7 percent, that was the equivalent of about 111 extra third downs, so it was a bit more significant. Not surprisingly, the scoring decrease from 1995 to 1996 was the second largest the league has seen since 1978.

Scoring has not suffered the last few years, but the trick with third down is always that you can ease it or simply avoid it by being better on first and second down.

Year-to-year Correlation

Great offenses still tend to be the best at converting third downs. There has been research by others to suggest third down success does not repeat itself well year-to-year.

The year-to-year correlation coefficient for third down conversion rate is 0.39, which would suggest neither a strong nor weak ability to repeat your performance. A team like the Colts showed they could sustain this. Peyton Manning helped the team lead the league in third-down conversion rate in five straight seasons from 2005 to 2009.

Like we said, the passing game strongly dictates how third down goes, and the quarterback of course controls the passing game.

Here are the year-to-year correlations for third-down conversions for all teams.

The higher the number means the more likely teams were able to maintain their rate – good or bad – the following season. Notice the highest correlation is the most recent example: the 2010-to-2011 seasons (0.68). That might be why Football Outsiders mentioned the “third-down rebound effect” has seemingly disappeared on offense in recent years.

This also might explain why we keep seeing minor league-wide decreases in recent years. There are not as many teams going through wild swings from year to year. Sure, an injury to someone like Manning hurt the 2011 Colts (fell from 44.61 percent in 2010 to 34.58 percent last year), but it has been less common.

Finally, let’s utilize some of the power the new Game Play Finder has at Pro-Football-Reference. We can look at stats on all third down plays (even ones including penalties), which is exactly what has been summarized in the following table.

For those two years in 2007-08, we saw the average third down attempt fall under 7.0 yards to go for a first down. That explains why those seasons had the second and third highest rate of conversion since 1991.

Last season, the average was at nearly its highest (7.29), and teams went with a pass on a century-high rate of 78.0 percent of plays.

Not to make any more sweeping generalizations over this data set, it is a good one to have for reference.

Conclusion

Should anyone be concerned offenses are converting 37.8 percent of their third downs instead of 39.0 percent? No, of course not. The 2012 season is still on pace to be the highest-scoring season ever, even if things still look strange to this point.

It just gets back to the point that no matter how many records we keep seeing broken each year, it makes you wonder what all these performances are really producing in the end when the offense statistically is still not dominating the defense.

As much as people want to say the NFL is transforming into the Arena Football League where it’s scoring drive after scoring drive, the numbers do not support it.

You would think with all of the supposed offensive advantages that we would have a league where teams would average more than the typical 20-23 points per game we have seen decade after decade.

Did you just watch Sam Bradford throw 12 straight incompletions and Kevin Kolb take 17 sacks in the last five days? Defense is not dead yet, and when push comes to shove on a critical third-down play, the defenses are winning just as often as they did over 20 years ago.

Scott Kacsmar is a football writer/researcher who has contributed large quantities of data to Pro-Football-Reference.com, including the only standardized database of fourth quarter comebacks and game-winning drives. You can visit his blog for a complete writing archive. Please send any questions or comments to Scott at smk_42@yahoo.com, or you can follow him on Twitter at @CaptainComeback.