In this week's Varsity Numbers, Bill Connelly takes a page out of baseball's playbook and attempts to isolate power from efficiency.
03 Sep 2013
by Rivers McCown
In the past few years, a new generation of quarterbacks gave NFL fans an introduction to the read option. Cam Newton, Colin Kaepernick, Russell Wilson, and Robert Griffin had the unique talents necessary to make this schematic option work. They also had coaching staffs that would embrace their gifts rather than shoehorn them into running only a "modern" refined NFL passing game.
The next era of offensive evolution? It could be a trend towards the no-huddle offense.
Smoke signals out of Denver indicated that the Broncos were tinkering with more no-huddle looks this offseason under new offensive coordinator Adam Gase. Obviously, there's a whole lot made of Chip Kelly's offense in Philadelphia and how that will shake out, mostly referring to the tempo. Mike Mayock was interviewed by MMQB recently and had this to say:
So for me, the whole tempo thing is going to be a storyline this year. Chip Kelly is going to try to play at warp speed. The Patriots got seven or eight more snaps per game than any other team in the league last year. I think other teams want that. Every camp I went to has a version of that warp-speed tempo. With all these teams trying to run it, I’m really interested to see how the defensive coordinators handle it. One of the things that system does, if you move fast enough, it can take a talented defensive coordinator out of the game. I hear about all the zone-read stuff, but I think a bigger story is tempo.
Meanwhile, perusing Chris Brown's The Essential Smart Football will lead you to this section on the no-huddle offense in general (and Tom Brady in particular):
Given Brady's success this season -- not to mention the success of Peyton Manning's no-huddle Colts in past years -- I expect the no-huddle offense will continue its resurgence. It's worth pondering, though, why NFL teams have been so slow to incorporate something that seems intuitively to be so much better than the alternatives.
Brown goes on to note that teams such as the late 80's-early 90's Bills and Bengals used plenty of no-huddle -- as it is with writing, nothing you can put on an NFL field hasn't been done before -- but ultimately the blame seems to rest on elongated play calls that would actually seem to require a huddle.
So, in unpacking the data we have on the no-huddle offense, there are four things we need to keep in mind.
1) The small sample size. Last year included the most plays marked no-huddle in our database yet, but 2,180 plays out of 32,636 plays is still roughly 6.6 percent of all plays. This gets progressively smaller as we look further back, up until we get to 2002, where less than 100 plays were marked as no-huddle. So, as with most football statistics, small sample size rules are in effect.
2) Inconsistent play-by-play scoring. Moreover, because the NFL's play-by-play scorers are so inconsistent at actually marking no-huddle situations -- especially in the earlier years we've sampled -- there's the possibility that some of this is off by varying degrees. We'll soldier on with what we've got.
3) The Tom Brady and Peyton Manning effect. How much of our numbers are colored by the fact that these guys were the quarterbacks of a plurality of no-huddle snaps throughout the league over the past 10 years? They're only two of the greatest quarterbacks in NFL history, after all. So, I listed separate DVOAs with Brady and Manning (and their backs) and without Brady and Manning (and their backs) in the following table.
4) Running a no-huddle offense versus running a no-huddle offense at the end of a half. While coaches may have "packages" designated for end of half situations that will take advantage of certain no-huddle ideas, there is a big difference in the psychology of a team that willfully chooses to run a no-huddle and one that runs a no-huddle because they need to get down the field as fast as they can. To save myself from going over line after line of data to try to figure out which team was doing which, delaying this piece until March 2015, I simply listed a DVOA for plays marked no-huddle in the first and third quarter and listed how many no-huddle plays came in the second and fourth quarter of games.
|No-Huddle Stats, 2012-2003|
|NH Passing DVOA||7.6%||7.2%||1.0%||4.3%||1.0%||5.3%||7.9%||5.0%||-0.9%||17.6%|
|NH Passing DVOA w/o Brady/P.Manning||4.1%||0.4%||-5.4%||-1.0%||-0.3%||1.4%||5.1%||-1.3%||-10.8%||20.9%|
|2nd/4th Quarter NH Passes||975||946||739||811||646||632||536||257||183||184|
|1st/3rd Quarter NH Passing DVOA||8.7%||24.0%||-7.0%||-6.1%||11.6%||2.9%||28.7%||8.1%||-15.2%||-26.9%|
|NH Running DVOA||11.3%||16.9%||13.9%||3.0%||11.3%||4.1%||1.5%||11.3%||36.1%||10.8%|
|NH Running DVOA w/o Brady/P.Manning||13.3%||16.7%||16.0%||3.9%||5.7%||5.2%||-1.3%||9.0%||6.5%||16.4%|
|2nd/4th Quarter NH Runs||348||252||186||251||164||198||198||94||44||42|
|1st/3rd Quarter NH Running DVOA||9.9%||15.8%||-0.6%||-2.6%||-5.0%||10.2%||5.7%||19.5%||19.3%||24.6%|
Running sure has been more effective than passing, hasn't it? While I think some of this can be explained just by zone schemes and ideas breeding lighter offensive linemen that would be built for quicker mobility, I think what we really may have here is a clash of ideologies. NFL defensive positions are not often built with the same versatility that offensive skills are. In a pass-happy league, no defensive coordinator is going to eschew an edge rusher like rookie year Bruce Irvin because he can't set the edge. Instead, he'll just try to spot him in passing situations. Take away the ability to substitute, and suddenly you aren't able to hide weaknesses that easily. Players with versatile all-around skill sets become more important.
So, naturally, you put a defense on the field to stop a no-huddle -- predominantly thought of as a passing offense, especially when you figure the end-of-half situations it plays against in a lot of these downs -- you're going to sacrifice to stop the pass. That's not to say that this is all end-of-half situations: Ray Rice had a 32.6% DVOA in 40 first or third quarter carries in Baltimore's no-huddle last season.
Generally, these numbers are bereft of huge negative DVOAs. Take Manning and Brady out, and you see a few less appealing numbers -- but you can say that about just about any stat if you cherry pick things enough. If this were more of a fad offensive idea, you'd expect to see the numbers take some wild turns toward the negative with the small sample sizes. I think the fact that you don't is indicative that the no-huddle has potential as more than just a change-up for teams, so long as they build their identity and offense around it.
What the hell happened in 2009? The entire chart seems to flow smoothly and evenly as far as percentage of no-huddle plays run until we get to 2009, where there were more no-huddle run plays than any year asides from 2012 and the run DVOA is pretty low.
The two teams that significantly shook this tree are the Browns and the Bills. If you're like me, your first thoughts might be "Oh! Chan Gailey and Rob Chudzinski!" Nope. Try Alex Van Pelt and Brian Daboll.
The Bills actually fired offensive coordinator Turk Schonert before the start of the 2009 season, after he'd spent the whole offseason preparing a no-huddle offense. Schonert graciously disagreed with Dick Jauron's assertion of the situation, saying "[Jauron] wants a Pop Warner offense. He limited me in formations and limited me in plays ... he's been on my back all offseason."
For the first five weeks of the season, Van Pelt ran some of the installed plays, giving Fred Jackson 51 carries out of a no-huddle attack. He had just two more no-huddle runs for the entire season, and finished the year with a -45.3% DVOA out of the no-huddle. (Marshawn Lynch added 18 carries of -24.8% DVOA for good measure.) After Week 5, Trent Edwards (-34.9% DVOA in 101 no-huddle snaps) was deposed. Ryan Fitzpatrick and a slower scheme were installed. Jauron was fired for ever going along with the idea that a Trent Edwards no-huddle scheme would be a good idea.
Meanwhile, in Cleveland, Daboll had Jerome Harrison and Chris Jennings run out of it 77 times in the last seven weeks of the season -- they accumulated 16 DYAR for their effort, and both had negative DVOAs. Brady Quinn and Derek Anderson combined for a -16.0% DVOA in 148 attempts. It should be noted that during this span, the Browns did finish on a four-game winning streak to lift themselves out of "worst team in the NFL" status. Unfortunately, in a copycat league, teams tend to copy things that are actually successful, and "this lead the Browns to 5-11!" isn't an inspiring rallying cry.
So, with those efforts on tape, it does make a little more sense why the NFL suddenly backtracked a bit on the no-huddle in 2010. They were probably filing it away under "underdog strategies," "nonsense," or "Seriously, Trent Edwards in a no-huddle?!?"
14 comments, Last at 15 Oct 2013, 11:34pm by saramyne