Fantasy Theory: Do teams with a bad defense pass more?

Sometimes for fantasy purposes it can be quite interesting to have a QB or WR on a bad team that gets behind a lot, and is forced to throw a lot. But is this somewhat predictable? Can we find a correlation?

Forum user Fort4242 decided to investigate correlation between passing attempts and a number of defensive stats, ie does a team pass more depending on the defense? Unfortunately none of his stats led to any significant result.

I went a step further and decided to match all offensive and defensive stats for 2018 on Pro Football Reference to see if there was ANY kind of correlation to be found. First the spoiler: NO.

(but it looked for a bit that I could indeed find something. The text below is the description of my wild goose chase)

I calculated the correlation of each pair of columns, and then used conditional formatting to highlight the fields with high (blue) or low (red) correlation. Horizontal are offensive stats, vertical are the defensive stats, in all cases for the categories:
- points and yards
- plays, turnovers and 1st downs
- pass stats
- run stats
- penalties

In my mind, there are 2 clear areas with a lot of correlation:

- it seems that good pass defense and good pass offense are correlated

- on the other hand, good rushing stats and rushing defense have a NEGATIVE correlation.

Neither of these statements make a lot of sense to me, so let's investigate these.

Pass defense and offense

Let's zoom in on the numbers for the pass offense vs pass defense.

Some of these correlations make very little sense to me. Offensive passing TDs correlate with allowed pass completions and allowed 1st downs? That just doesn't make sense. Therefore, let's focus on Pass yards vs pass yards allowed; and pass 1st downs vs the same allowed.

Horizontally we have the pass yards and 1st downs allowed; vertical we have the scored yards and 1st downs. So now we see the same names appear in both graphs at more or less the same places (which makes sense):

- KC and Tampa both had a great passing offense, and a pretty crappy pass defense. Many of their games were complete shoot-outs.

- Buffalo, Jacksonville, Tennessee and to some extent the Cardinals had a terrible offense, but also didn't allow a lot of yards. These were the snooze-fests.

- One clear outlier is Minnesota, which was actually quite tight in pass defense, but had a moderately effective offense. On the other hand, Cincinatti allowed oceans of passes, but didn't advance very effectively.

So did the shoot-out teams rush more or less than the snooze-fest teams? Nope, there is no such correlation.

Rush defense and offense

So let's do the same for Rushing defense vs Offense. First we zoom in:

Again, we see this werid but interesting correlation between passing TDs and attempts allowed, which I will ignore. I prefer to look at rush attempts vs rushes allowed, and 1st downs vs 1st downs allowed:

Now the teams are generally ordered from top-left to bottom right, meaning that teams that rush more, allow fewer attempts, and vice versa.

- Seattle, Baltimore and (somewhat surprisingly) the Saints are the top rushing teams

- Arizona didn't just suck at passing, but also at rushing

- outliers are Philly and the Steelers, who had a good rushing defense but also didn't rush a lot (the Steelers?? Yes, the Steelers). Weirdly enough there are no outliers to the top-right.

So, did the teams with a strong rushing attack pass less? ...nope.

Older years

I am thoroughly confused. Teams that pass well defend less well against the passing game; teams that rush well defend better against the rush? None of this makes sense at all, so let's see if we can find the same correlation in 2017 and 2016:

...Nope.

By this time I was too desillusioned to mkae more beautiful graphs, but let's just say: no. If you look at larger groups of data, for multiple years, all the above correlations disappear.

Conclusion

In the end, there simply is no strong correlation between defense skills, and pass offense. Garbage time may be a well known phenomenon in Fantasy, but it's not something you can safely predict or chart or build your team on. It happens, but not consistently.