When a fantasy football player is preparing for a draft, he or she will often consider a player's schedule before selecting an offensive talent. After all, the defenses that a player faces over the course of a season will impact how well he'll perform on a week-to-week basis.
But do we ever consider the impact a defense has on the offense of the same team? The NFL is a game of three phases - offense, defense, and special teams. And a failure in any one of these areas can have an impact on a team's chances of success in other phases.
Crunching the Numbers
When we consider numberFire's data on NFL offenses and defenses since 2000, we can learn a lot about the connection between the quality of an NFL defense and the corresponding quality of the offense.
In fact, we learn that there's no connection at all.
One of the claims many make about teams with bad defenses is that they run more plays on offense.
But when testing the connection between Adjusted Defensive Net Expected Points (adjusted for strength of schedule) and offensive plays ran for every team since 2000, there is zero correlation between the two statistics. Not a figurative "zero" meaning next to nothing, but literally no correlation whatsoever. There's no math that supports a claim that a bad defensive team runs more plays than a good defensive team.
Furthermore, there is virtually no correlation between our defensive metrics and our passing metrics. Bad defensive teams aren't more efficient or productive at passing the ball, despite being behind and needing to throw the ball more often to catch up.
Simply having a bad defense doesn't make an offense run more plays or perform better through the air. So is there anything we can take away from the link between bad defenses and fantasy production for offensive players?
The correlation between a defenses and pass-to-run ratio is strongest mathematical relationship, as the two variables have a much stronger correlation than any of the aforementioned combinations. There's a mild relationship between how good a team's defense is, and how often they pass the ball. Bad defensive teams pass more, and good defensive teams pass less, as you might expect.
Using this Data
So how can we use this data to our advantage? Let's consider what we know.
1. Having a bad defense doesn't give an offense more plays.
2. Having a bad defense doesn't make an offense more effective.
3. Having a bad defense causes a team to pass at a higher rate than a team with a good defense.
That means that running backs who are limited as receivers and play for teams with bad defenses are in a disadvantageous position, while runners from strong teams will have a chance at more carries and more overall production for fantasy owners.
Who does this impact? Toby Gerhart is one of the more notable players whose value will take a hit thanks to his team's defensive standing and likely need to throw the ball more often on offense. The former Stanford back isn't Chris Ivory-level bad coming out of the backfield, but he's had very low per-target NEP numbers as a receiver since entering the league in 2010.
And since Gerhart isn't expected to contribute heavily in the Jacksonville passing game (we only project him to pick up one touchdown through the air on fewer than 40 receptions), his ability to provide steady contributions as a runner may be put in jeopardy by his team's poor defense.
The only defense last season worse than the Jaguars' was the Atlanta Falcons, and they feature another back whose value may be limited by his team's bad defense. Steven Jackson had an awful year last year as a receiver, finishing as the sixth-worst receiving back in the NFL on a per-target basis among players at the position with 20 or more targets.
Playing on a team with younger, more capable receiving backs behind him on the depth chart does nothing to help his cause, and neither does the Falcons' league-high pass-to-run ratio last season. Jackson is already on shaky ground as a productive fantasy player thanks to his age and career workload, and this revelation about his team's defense won't help matters either.
Other backs who may be impacted by their defense providing them with fewer opportunities are Eddie Lacy, Maurice Jones-Drew, Darren McFadden, Trent Richardson and Ben Tate. These players will still get chances to run the football, but they play for teams with bottom-10 defenses last year, and may not have the a consistent workload this year as runners.
Poor defenses aren't all bad news for running backs, however. A back capable of catching the ball out of the backfield will get more chances to ply his trade on a team with a bad defense, something we saw with Matt Forte last season, and will likely see again as the Bears will struggle on defense as they did last year.
But players lower on depth charts who excel on third downs, like Jacquizz Rodgers of the Falcons and Danny Woodhead of the Chargers, also figure to get more opportunities due to deficits their offenses will likely be in thanks to poor defensive performances.
Another Noteworthy Trend
Since 2000, there is a less pronounced but still noteworthy trend of good defensive teams having productive rushing offenses according to our NEP data.
The problem here is the classic "correlation does not imply causation" debate. When it comes to bad defensive teams throwing more often, we can agree that it makes perfect sense for a quarterback to throw more frequently with a bad defense on his side, as he'll likely be behind in games more often.
But is there any logical causation between a strong defense and a productive rushing game? Which one predicts the other?
We could say that this is the flip-side of the logic used to support the claim that bad defenses pass more often, as good defenses run more often, and as a result, are more productive overall.
But simply running more isn't enough to impress our metrics. You have to run efficiently.
So unfortunately, we can't predict how good a back will be based on how good his defense is. But we can use this information to predict running backs who will get more opportunities and may have a higher chance for increased production and efficiency.
This means that Marshawn Lynch (with the top-rated Seattle defense in 2013), DeAngelo Williams and Jonathan Stewart (with the third-ranked Carolina defense), and Jamaal Charles (with the fourth-ranked Kansas City defense) figure to get plenty of chances this season, and will each produce in line with their career norms, given the defenses perform well again.
But it also means that Rashad Jennings may have more chances than expected, as the Giants' defense was ranked in the top 10 within our metrics last year, and should improve as the team rebounds from a mistake-ridden 2013.
None of this data is strong enough to vault a player up your draft board by multiple rounds, or take a player out of consideration. But it's a useful tiebreaker, and something to consider when making a touch decision between two players.