Statistical analysis: Playing with the lead
As time expired last Sunday, Philadelphia's Jason Avant couldn't hold onto what would have been an amazing, game-winning catch. In addition to the meltdown against the Texans, the Redskins are four ounces of pressure and one holding call away from squandering two other second-half leads. It's a razor-thin line between starting the Mike Shanahan era 2-2 and starting it 0-4.
It's easy to knock the defense for failing, or almost failing, to hold those leads, but the offense is as much to blame. The Redskins have managed to add only three second-half points in the three games in which they have had leads. And that was a 49-yard kick.
It's all too predictable: Once a team has a lead, its offense suddenly becomes overly conservative, running for two-yard gains and then throwing short dump-off passes on third down. Opponents who have their backs up against the wall increase their risk levels and are able to come back to win more often than most fans (and coaches) think. Offenses need to keep scoring to keep a lead safe. In today's pass-happy and offense-friendly NFL, it's unrealistic to expect a defense to shut down an opponent all game long.
Too many predictable runs and too many short passes: That's exactly what I expected to see when I looked at the Redskins' second-half stats in the games against Dallas, Houston and Philadelphia. I was prepared to write all about how Shanahan's conservative mindset put his team's leads at risk, so I was surprised to learn this isn't the case.
Shanahan's play calling has been slightly more aggressive than the league average for teams in similar situations. For teams with similar leads in the second half, their run/pass balance is typically 55/45. Shanahan has called for more passes than runs, for a run/pass balance of 47/53. Donovan McNabb's reads have been fairly aggressive as well. About 19 percent of his attempts have been "deep," defined by the NFL as greater than 15 yards through the air. That's right in line with the league average.
I think it comes down to simple execution, and there's really no systemic problem. The offense hasn't performed well in those six quarters, and it's as simple as that. Their success rate during those periods (percentage of plays that help move the chains and score) has well been below average at 40 percent. Just a slight increase, up to 45 percent, would mean several more first downs, more time of possession, and more points.
Performance varies from game to game and from quarter to quarter. There's no law that says teams will have all their success spread out evenly among all their games or within any single game. Players can't control when they perform well and when they don't. If they could, they'd save up all their successes for the moments when it matters most. Consistency isn't something that can be controlled directly. It's a product of overall ability and good fortune.
Brian Burke is a former Navy pilot who has given up his F/A-18 for the less dangerous hobby of football analysis. He is the creator of Advanced NFL Stats, a Web site about football, statistics and game theory.
| October 6, 2010; 2:59 PM ET
Categories: Statistical analysis
Save & Share: Previous: Devin Thomas: 'I just really don't know what's going on'
Next: Clinton Portis out four to six weeks with groin injury
Posted by: rich20ssu | October 6, 2010 3:03 PM | Report abuse
Posted by: ThisGuy | October 6, 2010 3:03 PM | Report abuse
Posted by: beep-beep | October 6, 2010 3:07 PM | Report abuse
Posted by: mattymatty2000 | October 6, 2010 3:07 PM | Report abuse
Posted by: zcezcest1 | October 6, 2010 3:08 PM | Report abuse
Posted by: alex35332 | October 6, 2010 3:10 PM | Report abuse
Posted by: beep-beep | October 6, 2010 3:12 PM | Report abuse
Posted by: Poopy_McPoop | October 6, 2010 4:45 PM | Report abuse
Posted by: emc7x | October 6, 2010 4:56 PM | Report abuse
Posted by: McLobo | October 6, 2010 5:38 PM | Report abuse
Posted by: laserwizard | October 6, 2010 5:53 PM | Report abuse
Posted by: VegasJim | October 6, 2010 7:20 PM | Report abuse