The Funnel Defense Effect
While I was in college, a funnel was a way to drink a beer in three or four seconds. Who am I kidding? It still is.
But, in fantasy football, the ‘funnel’ is a defense that many people target in an attempt to get a boost in quarterback scoring. In theory, when a team has a much stronger rushing defense than passing defense, it will ‘funnel’ the opposing team into passing the ball against a weak secondary. In theory, this increased passing volume against a weaker secondary should lead to enhanced fantasy scoring for the quarterback.
Scott Barrett of Pro Football Focus examined quarterback streaming in detail this offseason and found that bottom-10 defensive units give quarterbacks a 1.5 points-per-game boost on their average scoring. If the funnel defense theory works as expected, then quarterbacks should see a PPG increase greater than 1.5 points against these types of defenses.
How Do We Find a Funnel Defense?
Football Outsiders each week ranks defensive efficiency against both the pass and the run with their Defense-Adjusted Value Over Average (DVOA) metric. The 2015 end-of-season rankings are as follows:
Source: Football Outsiders
Since we are hoping to identify a select group of defenses to target, we can identify the six defenses whose 2015 passing ranks are at least eight lower than their rushing ranks. For example, Tampa Bay ranked 9th against the run and 26th against the pass — a difference of 17 spots. In addition to Tampa Bay, these defenses qualify as funnel units: Jacksonville (14), Baltimore (13), Miami (9), Pittsburgh (8) and the New York Jets (8).
Quarterback Points Against
These six defenses ranked 26th (NYJ), 15th (MIA), 14th (PIT), 10th (TB), eighth (BAL) and fifth (JAX) in points allowed to quarterbacks over the 2015 season. At first glance, this is a pretty wide range and it seems to identify only two defenses that generated extra QB scoring. At the same time, this accounts for only raw QB scoring and does not adjust for the strength of schedule.
Taking this a step further, we can use our Trends tool to compare quarterback scoring against these defenses relative to their season-long averages. To do this, we will examine Weeks 1-17 against each of these defenses and then subtract the season-long average of the quarterback from the points he scored against that particular defense. Then we’ll total them. See Tampa Bay below. (Using DraftKings scoring).
After adding up the 16-game differences for the defense in question we can see how many points these quarterbacks as a group scored over the season. (The data also accounts for injuries by combining quarterback points in instances when two quarterbacks both played significant snaps in a single week.)
This leaves us with four defenses that provided a significant boost to quarterback scoring and two defenses that didn’t.
Also, it’s notable that the four funnel defenses to provide a boost to quarterback scoring also were exploitable on both DraftKings and FanDuel, per our salary-based Opponent Plus/Minus metric. Targeting these defenses last year added value to DFS rosters.
Conclusion
I hoped to find more conclusive data on the funnel defenses but I think this is a good start. To be perfectly honest, I’m not 100 percent sure that funnel defenses are truly exploitable. I think that it is first important to note that New York boasted a passing defense that ended the season ninth overall. While it was more exploitable than the rushing defense, it was still a very strong unit.
Also, it’s important to note that Miami’s points allowed average in the chart above was heavily affected by Tom Brady’s 5.26-DK point game against them in Week 17 — more than 18 points below his season-long average. Then again, even if we remove that game, Miami’s defense still allowed QBs as a group to score only 0.66 points per game more than their averages.
In the end we are working with 16-game samples in the NFL and there is often a lot of noise. And there are still a lot of questions that remain:
• How much did Pittsburgh and Jacksonville’s explosive passing offenses affect the scoring?
• Does Vegas account for these types of defenses in their team totals?
• In a sport in which quarterback production is so heavily weighted on touchdowns, is this just noise?
Ultimately we can see that four of these defenses did provide a significant boost in quarterback production, but I don’t think that we can prove that the funnel defense was the reason. And, given that, if a time comes this season when people start relying heavily on the idea of funnel defenses, they might very well be the chalky plays to fade.
