The Redskins have five games to improve some lagging aspects of their game to avoid either another one and done appearance in the playoffs or not making the postseason at all.

One area where they have had issues all year long is in third down defense, particularly against the pass. And while anyone watching the games knows that they have had problems, pulling out the numbers gives a shocking look at the extent of the difficulties.

On third down, opposing quarterbacks have completed 71 of 105 passes (67.6 percent) for 933 yards with six touchdowns and no interceptions. Plugging those numbers into the passer rating formula of 114.5. If you assigned that rating to one quarterback for the season he would be second only to Tom Brady in the NFL.

It’s not surprising to see that quarterbacks like Aaron Rodgers (151.8 third-down rating) and Ben Roethlisberger (137.5) did well on third down against the Redskins. But rookie Cody Kessler was 11 for 11 on third down with a 128.5 passer rating. Sam Bradford had a respectable 95.8.

That looks bad but it’s worse than it looks on the face of it. For the NFL as a whole, the passer rating on third down is 88.6. That’s a touch under the league rating on all passes, which is 89.5. And that makes sense. Teams expect the pass on third down so they should be able to defend it better, or at least as well as they do on other downs. But that’s not the case with the Redskins.

 

What makes this puzzling is that the Redskins are pretty good defending the pass on downs other than third. Quarterbacks have completed 66.4 percent of their passes against them with a passer rating of 85.9.

The bottom line is that opponents convert on third down 43 percent of the time when they pass, 28th in the NFL. The better teams in this category have conversion percentages in the lower 30’s; the league average is 38 percent. If the Redskins can get their conversion rate to the mid-30’s and give their potent offense another possession or two per game they will be in better position to record another win or two.