The Redskins currently have the NFL’s worst third down conversion rate at 23%, but this is a widely misunderstood statistic. And that misunderstanding is the basis for some poor offensive tactics throughout the league.
Third down plays have enormous leverage in the outcomes of drives. First and second downs are important, of course, but third down is the make-or-break play that decides whether a drive will continue or stall. Third down play outcomes are magnified in their impact on the game. Teams rise and fall on how well they do.
Consequently, coaches and fans alike see that third down success correlates strongly with overall success. Coaches focus on making sure that their third down plays are as successful as possible, which means using the first two downs to gain relatively assured bits of yardage using running plays to set up third and short, rather than third and long, situations. This mindset is intuitively seductive, but it’s ultimately self-defeating.
Offenses are better off thinking of their three downs (and fourth when the situation requires) as isolated opportunities for ten-yard conversions rather than stepping stones toward what coaches call a “manageable third down.” The best third down situation isn’t third and 1 or even third and inches. It’s converting on first or second down, before ever reaching third down. Rather than seeking a short third down situation, offenses should be avoiding third downs whenever possible.
Running to move the chains makes an offense less likely to convert. When offenses passed on first down in 2011, they eventually converted on that series of downs 72% of the time. When they ran, they converted 66% of the time. (I’m not referring just to immediate conversion on the first down play itself, but conversion on any subsequent down as well.) The league-wide conversion rate when passing on first down has been consistently climbing over the last 12 seasons, while the conversion rate when running has been relatively steady.
But should a first down pass attempt fall incomplete, it certainly makes sense to follow with a run on second down to avoid a third and ten, right? No. On second and ten, passes still lead to successful conversions more often than runs by a margin of 57% to 46%.
Strangely, on third and ten, passing is no more successful than running at coming up with a first down. Teams run so infrequently in that situation that defenses are less prepared. According to the central tenets of game theory, this suggests that offenses are choosing the two strategies in the optimum proportion, at least when conversion is the measure of ultimate of success. So why are offenses consistently correct on third down but off the mark on first and second downs?
I think the answer is that the results of each strategy are immediately apparent to coaches and coordinators on third down but not on first or second down. The feedback is not a statistical abstraction like ‘future probability of conversion’ or some advanced stat like Expected Points, but rather an instantly concrete result. Either the punting team is jogging onto the field or your offense just successfully moved the chains. It’s much easier for coaches to statistically, or even intuitively, find the optimum mix when the results are so immediately apparent.
Third down percentage has a large impact because of its leverage, but as a statistic it can be misleading. Despite the fact that the Redskins are at the bottom of the league in that category, their overall series conversion rate is perfectly league-average at 68%. This means they’re converting on first and second down, which is the right approach. To be clear, this does not imply that offenses should pass on every first and second down. What it does suggest is that offenses should use a pass-heavy mixed strategy of plays designed to immediately convert, rather than chip away at, the distance to go. Seeking a manageable third down is a recipe for defeat.
Brian Burke is the creator of Advanced NFL Stats, a Web site about football, statistics and game theory.