In playoffs, it's all about defense (not QBs)

January, 17, 2014
Jan 17
11:30
AM ET
Much of the focus heading into Sunday’s Conference Championship games will be on the quarterback play, particularly with future Hall of Famers Tom Brady and Peyton Manning facing off for the fourth time in their postseason careers.

But if recent playoff history is any indication, it’s not the quarterbacks who will have the most impact on which teams get a trip to MetLife Stadium, but rather the defenses’ ability to stop those quarterbacks that will make a bigger difference.

Just looking at the past few Super Bowl champions, quarterbacks who rated outside the top half of the league in Total QBR through the Divisional Playoffs have led their teams to championships, including Eli Manning in the 2007 season, Ben Roethlisberger in 2008 and Joe Flacco just last year.

What those quarterbacks had in common was a defense on the other side that could control opposing quarterbacks, each ranking in the top quarter of the league in terms of QBR allowed. Even the recent Packers and Saints title teams, which had elite quarterbacks, paired them with Top-3 defenses in terms of opponents' QBR.

An examination of all playoff games back to 2006 (as far back as QBR goes) shows that although quarterback play seems to carry over from game to game in the regular season, that correlation decreases in the postseason.

And conversely, the ability of a team to contain opposing quarterbacks seems to have a greater impact on how quarterbacks perform – and as a consequence, who wins – in the playoffs.

The details of our study
This analysis looked at how two components entering each game – the quarterback’s QBR for the season and the defense’s QBR allowed on the season to that point – related to the quarterback’s Total QBR in the game as well as the final result.

To ensure each of those numbers were representative, the only games analyzed were those in which both the quarterback and the defense had at least 100 action plays entering the game, and in which the quarterback had at least 15 action plays within the game.

The first graph below shows how well a quarterback’s QBR entering a game does in terms of “predicting” QBR within that game, with separate trend lines for regular season and playoff games.


The regular-season trend shows a decent amount of regression to the mean, as quarterbacks with extreme QBRs entering the game have less extreme performances, on average.

But the general trend of good quarterbacks having above-average games and bad quarterbacks having below-average ones is clearly present with the upward trending line.

On the other hand, the trend line for the playoffs is pretty wacky. This is what happens when quarterbacks like Mark Sanchez, Tim Tebow and Joe Flacco have games of 90 or higher QBR, and Tom Brady, Philip Rivers and Matt Ryan each post games with a QBR less than 20.

Now, contrast that picture with the one below, which looks at how strongly the opponent’s QBR allowed entering the game influences the quarterback’s play in the game.

The regular season trend is almost flat, showing that defensive quality has minimal impact on how well a quarterback plays in those games.

But look at the postseason trend – while clearly not perfect, there seems to be a stronger impact on QBR by defense entering the game in the playoffs. The elite defenses mentioned above flexed their muscles in the postseason, limiting opposing quarterbacks as they had earlier on in the season.


A more rigorous analysis using multiple regression shows that the quarterback’s QBR entering the game is still significant in the postseason, but the defense’s QBR allowed becomes more predictive in the postseason. The small sample size of the postseason means that this pattern might be due to random fluctuation, but the trend is still something worth keeping an eye on.

We can go one step further and look at how impactful these pregame QBR values are in terms of actually winning the game. A look at the chart below shows that the pattern is similar to that with the actual QBR in the game.

Having a starting QB with a better QBR entering the game gives you a solid chance to win in the regular season, but the team with that advantage is just 44-40 in the postseason since 2006. Conversely, teams that have a better QBR defense than the opponent are 51-33 in those same playoff games, including 6-2 so far this postseason.

Looking ahead to this weekend, perhaps we can de-emphasize Peyton Manning’s 22-point advantage over Brady in QBR and instead focus on the defenses that they line up against.

As it works out, though, the Patriots and Broncos have very similar, mediocre opponent QBR values to this point: 48.3 for Denver, 50.5 for New England. So there isn’t any clear advantage there.

In this regard, the more interesting matchup is out in Seattle. Both defenses in the 49ers-Seahawks game rank in the top five in terms of QBR allowed, but the Seahawks are ahead of all other NFL teams at 29.5.

In each of their past seven games, the Seahawks have limited the opposing quarterback to a QBR at least 20 points below what he came in averaging entering the game, including their last meeting against Colin Kaepernick in Week 14.


If the Seahawks can keep up their own trend of shutting down opposing quarterbacks and the similar bigger-picture pattern that has emerged over the last several postseasons, they should give themselves a good chance to win this weekend.

And even though the road would go through Brady or Manning, the trend of strong defense carrying over more in the playoffs gives Seattle a pretty good chance of taking home the Lombardi Trophy in early February.

SPONSORED HEADLINES

Comments

Use a Facebook account to add a comment, subject to Facebook's Terms of Service and Privacy Policy. Your Facebook name, photo & other personal information you make public on Facebook will appear with your comment, and may be used on ESPN's media platforms. Learn more.