The Case for the Zero Blitz

The Case for the Zero Blitz
The Case for the Zero Blitz
Photo: USA Today Sports Images

Guest column by John Shirley, Sports Info Solutions

Why Teams Should Blitz More

Blitzing allows defenses to play downhill and dictate to the offense instead of sitting back and being forced only to react. Blitzing is also the quickest way to increase pressure on the quarterback. The following chart illustrates this, showing that pressure rate understandably increases as more rushers are added. * All data used is courtesy of Sports Info Solutions and from the 2016-2018 NFL regular seasons, unless otherwise noted.

The more defenders rushing the passer, the more pressure a defense gets on opposing quarterbacks. And pressure is the defense's best weapon against this era of pass-happy offenses. The following graph shows just how drastic of an impact pressure can have on how offenses perform by Expected Points Added (EPA).

Just hurrying the quarterback drops the average EPA allowed per dropback from 0.21 to 0.04, with increasing pressure levels dropping it even further.

The Zero Blitz

Given the above reasons to consider aggressive blitzing, we can show why the Zero Blitz (Cover-0 Blitz) in particular is something that defenses should utilize more often. For this analysis, blitz schemes were split into three different buckets: Zero Blitz, Zone Blitz, and Man-Free Blitz (Cover 1 coverage with 5+ rushers). These three blitz types were compared by EPA per Dropback, Positive% Allowed (the percentage of plays in which the defense allowed an EPA above 0.0), and Boom% Allowed (the percentage of plays in which the defense allowed an EPA of 1.0 or more). All following statistics exclude plays from within the defense's 10-yard line since strategy changes heavily once a defense is backed up against its own goal line.

Defensive Performance by Blitz Type
Blitz Type Plays Avg.
Rushers
EPA Per
Dropback
Positive%
Allowed
Boom%
Allowed
Zero Blitz 614 6.0 -0.09 40% 24%
Man-Free Blitz 7896 5.3 -0.01 45% 25%
Zone Blitz 3041 5.2 +0.07 48% 24%

As shown above, the Zero Blitz is generally the most effective blitz scheme of the three categories. It has the lowest average EPA Allowed and Positive% Allowed. The most surprising part is that it also has a comparable Boom% Allowed. The Zero Blitz is the only blitz type that has no deep safety, so it would be assumed that it allows a higher percentage of big plays. However, it actually allows basically the same Boom% as any other blitz. This absence of any deep safety allows for Zero Blitzes to average about one extra pass rusher per play. As shown earlier, more pass rushers equals more pressure on opposing QBs.

The graph above shows just how much more efficient at creating pressure Zero Blitzes are when compared to other blitz types. While Zero Blitzes do have a slightly lower sack rate, they have a higher percentage of every other pressure type and an overall higher pressure rate of 50 percent.

Zero Blitz vs Man-Free Blitz

If the Zero Blitz is so effective, then why is it called so infrequently? It is consistently the least-called blitz (7 percent of all blitzes in 2018), while its more conservative man coverage counterpart, Man-Free Blitz, is consistently the most called (53 percent of all blitzes in 2018). This is most likely because coaches are generally risk-averse and blitzes with an extra rusher and no safety help are seen as feast-or-famine play calls. Man-Free Blitzes are considered a safer play call since they have a deep safety present who is there to clean up any mistakes and help in coverage. But does the presence of a deep safety actually help a defense? Based on SIS charting data, deep zone safeties were only involved in 8 percent of Man-Free Blitzes in 2018. Being involved includes making a play on the ball, helping in coverage on the targeted receiver, or being involved in a tackle. So, it seems that a single high safety is extremely limited in the amount of plays he can be involved in and might not be the safety net that he's assumed to be. Evaluating single-high safeties shouldn't be just about how many plays they can be involved in, though. The presence of a deep safety can also deter targets to the deep middle of the field (between the numbers and 15-plus yards downfield) and make life harder for opposing quarterbacks. This is proven partly true in the following table, which compares Man-Free Blitzes to Zero Blitzes on pass attempts only (excludes scrambles and sacks).

Zero Blitz vs. Man-Free Blitz Performance
Blitz Type Attempts Avg. Throw
Depth
EPA/Att Positive%
Allowed
Boom%
Allowed
Deep Mid
Target%
Zero Blitz 560 10.4 +0.05 43% 25% 7.7%
Man-Free Blitz 7055 10.4 +0.13 48% 27% 6.4%

It is true that Man-Free Blitzes do a better job at deterring targets to the deep middle part of the field, although it is only an improvement of 1.3 percentage points. It is striking how well Zero Blitzes stack up even when sacks are excluded, and those extra deep middle throws may not be such a negative.

Zero Blitz vs. Man-Free Blitz Performance on Deep Middle Throws
Blitz Type Attempts EPA/Att Positive%
Allowed
Boom%
Allowed
On-Target
Throw%
Zero Blitz 43 +0.04 33% 28% 51%
Man-Free Blitz 453 +0.26 46% 41% 68%

There isn't much of a sample size for deep middle throws against Zero Blitzes, but the few results we have compared very well to those same throws against Man-Free Blitzes. It would be interesting to see how a larger sample size might affect these numbers to see if Zero Blitzes would continue to have more success.

Criticisms of Increased Blitzing

The criticisms of aggressive blitzes are the risk of allowing a big play (which the data so far suggests to be a false assumption) or that a quick passing game can nullify the pass rush. It has been shown already that Zero Blitzes allow a similar Boom% to other, more conservative blitzes, but how do they compare to other defenses when they fail to create pressure? As it turns out, about the same. On 269 clean-pocket pass attempts, Zero Blitzes averaged +0.24 EPA allowed and a 54 percent Positive%. This is basically the same as the average clean pocket attempt against all defenses -- +0.23 EPA allowed and a 55 percent Positive% allowed. A quick passing game seems like a logical counter to the Zero Blitz, but the numbers don't quite bear that out. On 42 pass attempts faced from a zero- or one-step quarterback drop, Zero Blitzes averaged -0.14 EPA allowed and a 43 percent Positive% allowed in 2018 (drop type data only exists for 2018). Once again, the Zero Blitz is extremely effective, but also in a small sample. To increase the sample, let's look at how Zero Blitzes performed against short passes (5 or less yards downfield). On over 239 short pass attempts from 2016 to 2018, Zero Blitzes averaged only -0.05 EPA allowed and a 48 percent Positive%. It seems that a quick passing game isn't quite the counter to the Zero Blitz that it's expected to be.

Conclusion

Hopefully this analysis shines a light on just how effective blitzing—the Zero Blitz in particular—can be at disrupting an opponent's passing game. Unsurprisingly, creating pressure on the quarterback is a major factor in this, and the Zero Blitz has shown to be the best at doing so, since it allows for an extra pass rusher. The fascinating aspect is that this aggressive blitz isn't as risky as it is assumed to be, allowing a similar big-play percentage as its more conservative counterparts. If defenses want to start dictating play to the offense, the evidence points to further utilization of the Zero Blitz.

Comments

19 comments, Last at 16 Jul 2019, 12:11am

#1 by crw78 // Jul 01, 2019 - 12:59pm

Seems entirely possible to me that the reason the zero blitz is more effective is because it is not called as often. Due to this, there's an element of surprise the offense may have more trouble adjusting to. If the zero blitz was called as often or more often than the others, my guess is that offense's would adjust accordingly, and the numbers wouldn't be as good.

Points: 0

#15 by PTORaven // Jul 08, 2019 - 12:47pm

if i had a nickel for every analysis that fails to take scarcity into account, i'd have a lot of nickels. doesn't mean that the analysis itself isn't worthwhile, but yeah, i wouldn't expect the zero blitz to be as effective if a team started calling 50% more of them. same way a pitcher's 3rd pitch wouldn't be as effective if he started throwing it 20% more often.

Points: 0

#2 by Will // Jul 01, 2019 - 2:02pm

Disregard :D

Will

Points: 0

#3 by robbbbbb // Jul 01, 2019 - 2:09pm

You don't tug on Superman's cape
You don't spit into the wind
You don't pull the mask off that old Lone Ranger
And you don't zero-blitz Russell Wilson with the game on the line

Points: 0

#4 by Todd S. // Jul 01, 2019 - 2:23pm

When I clicked on this, I figured it was a guest article by...Gregg Easterbrook! Always read past the headline, kids.

Points: 0

#9 by Dave Bernreuther // Jul 01, 2019 - 5:59pm

Easterbrook is an advocate of almost never blitzing, so unless you thought the headline was sarcastic....

Anyway, aside from the rare chaos theory 0 blitz against any QB, it seems to me that these are a terrible idea against an elusive and/or experienced QB but a great idea against overmatched QBs. (This is hardly a new idea.) The Ravens won a Super Bowl and thus guaranteed Joe Flacco extra shelf life by zero blitzing the hell out of Kaepernick inside the 10... because for all that he did well (and people forget about), he wasn't quite prepared to handle that much pressure that quickly on a short field where the receivers were easier to cover.

Which is another situation-dependent argument in favor of 0-blitzing, of course...

Points: 0

#13 by LionInAZ // Jul 02, 2019 - 7:17pm

Yes, probably because Kaepernick was an unlikely candidate to QB a Super Bowl team. 

On a side note, this site has become very slow since the upgrade just completed.

Points: 0

#14 by Todd S. // Jul 08, 2019 - 8:07am

As my brain scanned the headline, it interpreted "Zero Blitz" as "never blitz" - not "Cover-0 Blitz."  

Points: 0

#5 by techvet // Jul 01, 2019 - 2:33pm

Any Packers fan who hears "Cover-0" immediately thinks of the 2014 NFC title game between the Seahawks and the Packers, where the Packers were in Cover-0 for the winning TD in OT.

Points: 0

#12 by Vincent Verhei // Jul 02, 2019 - 4:36pm

If it makes you feel any better, that wasn't the first or last time that happened. I don't have access to the numbers, but as someone who watches a lot of Seahawks football, it sure seems like Cover-0 against Russell Wilson leads to a touchdown way more often than not. 

Points: 0

#6 by Harris // Jul 01, 2019 - 2:58pm

All of this ignores the fact that big blitzes are only called when 1) the offense is facing a very long conversion they're unlikely to make or 2) the red zone, where the defense doesn't have to worry about getting beat deep. I don't know how effective a 7-man blitz would be on 3rd-and-four at midfield.

Points: 0

#7 by Pat // Jul 01, 2019 - 4:23pm

The other issue is that the teams that call zero blitzes are likely the ones that have the personnel to handle them, and possibly only versus teams that they know that they'll work against. I mean, you know this is true to some extent: there are obviously receivers that get double coverage on an extremely high percentage of plays, in which case you just don't have the flexibility to call this at all.

But specifically because a zero blitz is a situational play, you'd expect there to be a strong bias in how well the average performs because the teams that are *capable* of doing it the most also happen to be better defenses, period.

Points: 0

#8 by Joseph // Jul 01, 2019 - 5:13pm

Looking at the numbers, they state 614 instances across the 2016-2018 seasons--in other words, 768 games total (not team games!). In other words, in a given game, there is an 80% (614/768) chance that one of the teams will call a zero-blitz once. But, just as you say, it is more likely that the distribution is not very even--some teams will call it more, and some opponents will see it more against them. There are just certain QB/WR combos that you don't want to blitz, period.

Also, for the author--why weren't turnovers or at least turnover rate included in this study? I know that they can be kind of random--but over a 3 year time period, surely that would even out, right?

Points: 0

#11 by GwillyGecko // Jul 02, 2019 - 3:45pm

614 plays, not 614 games where at least one zero blitz was called

Points: 0

#17 by Joseph // Jul 10, 2019 - 5:02pm

Exactly my point! 614 plays in a sample size of 768 total games analyzed!--In other words, a pretty rare coverage play.

Points: 0

#10 by billprudden // Jul 01, 2019 - 6:35pm

Sirs -

I wonder if an unemployed blitz-phile, perhaps one of the Ryans or Jeff Fisher, would write a companion piece?

For all of their faults, they'd certainly bring a perspective on this matter. Can't hurt to ask...

Love the content, and thanks!

Bill

Points: 0

#16 by Aaron Schatz // Jul 09, 2019 - 11:58am

Apologies to everyone who posted a comment in the past week. We just discovered that our new back end caught those comments and didn't publish them. I've just published all the waiting comments and we're working to fix the issue.

Points: 0

#18 by gomer_rs // Jul 12, 2019 - 1:04pm

While managing to hit upon it in the picture, strangely enough. As a Seahawks fan got to see perhaps the best QB against the zero blitz in the NFL. But, the way it was used was almost as a "prove it" tool. To play QB at reasonable level you have to deal with the zero blitz well, and once you prove your ability, you don't face it.

Points: 0

#19 by speedegg // Jul 16, 2019 - 12:11am

Interesting article, but brings up more questions.

Usually, the offense knows a blitz is coming, so the QB will audible to 7-man protections and 3-step throws/routes, but the corners MUST cover for 3 seconds and a blitzer MUST get to the QB, so if both sides know a quick throw is coming is that why EPA goes down?

Or, say it's a Gregg Williams defense where they go Cover-0/check downs a lot would Cover-0 be the result of pattern-match scheme, would the WR routes force a defense out of their zone assignments and switch to man? Also, when a receiver stays in to block I think the general rule is the defender blitzes in an attempt to overwhelm the pass protection. So it would be good to know is Cover-0 the called play or was that a check the defense executed as a result of a change in formation or routes run?

Points: 0

Save 10%
& Support the Writers
Support Football Outsiders' independent media. Use promo code WRITERS to save 10% on any FO+ membership and give half the cost of your membership to tip the team of writers.