Deep Passing Study - Where is the efficiency located?

Author: Billy Jones

Introduction

Welcome back to my blog series on the passing game in the NFL. In my last post, I used some visuals to understand the distribution of passes thrown in the NFL. In this post, I'll delve deeper into the question of, “is throwing the ball deep a good thing for the offense?” We've all seen the highlight reels of quarterbacks throwing bombs down the field for touchdowns, but how effective is this strategy in actual gameplay? Do long passes result in more yards, or are shorter, more consistent passes the more efficient route?

To answer these questions, I’ll look at the data on passing efficiency by location on the field. I’ll analyze metrics such as completion percentages and yards per attempt for passes thrown deep, intermediate, and short distances. By doing so, I hope to find out if throwing the ball deep is a more efficient way to move the ball down the field. So, let's see what the numbers say about the effectiveness of throwing the deep ball in the NFL.

Ground Rules

Before I jump back into the analytics, I would like to remind the readers of the ground rules we will be playing with. The data used for this analysis was obtained from NFLFASTR for 2018 through 2022.

Visualizations and Analysis

I’ll start by looking at passing efficiency by air yard distance (aka. depth of target). I am going to look at two primary metrics, completion percentage and passing yards per attempt. 

Analysis: As expected, completion percentage decreased as the distance of the pass increased. The highest completion percentage was in the short pass category, followed by intermediate passes, with deep passes having the lowest completion percentage. However, when I looked at passing yards per attempt, I found some surprising results. Short passes had the lowest yards per attempt, and I saw significant increases in the intermediate and deep passes. This is somewhat surprising as I assumed the decrease in completion percentage was going to balance out the increased distance of the pass attempt. This finding may provide some support for the hype surrounding quarterbacks' performances in the deep ball segment of the NFL combine. While it's important to remember that the combine is just one piece of information in evaluating a quarterback's potential, it seems that the ability to throw deep accurately could be a key tool when evaluating an NFL quarterback.

Now let’s see what some of our efficiency metrics look like by where it was on the field. In working with the NFLFastR dataset I found that there was a bit of undefined data in the passing location data.

It was extremely immaterial, but it is important to acknowledge it existed. I reviewed the data and it appears to be related to spikes and interceptions. For the 5 years in review there were 381 unmarked (0.422% of the data) passes.

Analysis: Overall, I found that the middle of the field was the most efficient location for passing, with the highest completion percentage and yards per attempt. However, I also found that the middle of the field was the least targeted section of the field in terms of passing attempts. The majority of passing attempts were focused on the left and right (wide) sections of the field. This mismatch seems odd to me, and I wonder if there is something behind this…

Conclusion

This wraps up part two of my study on football passing efficiency. In the next part I breakdown why I think teams don’t just throw deep and/or down the middle of the field at a high frequency. Can’t wait to share the next piece of the analysis!

*This blog post was enabled by ChatGPT. The text was generated by me, and the content is my own, but some sentences and wording were provided by the model. I take full responsibility for all information produced in this blog. More information about OpenAI and their technology can be found at https://openai.com.*