Do ballhawks matter?

After waxing eloquent about Evan Williams’ preseason debut in the most recent episode of Blue 58, a nagging question proceeded to bother me throughout the remainder of the day: is making a lot of plays on the ball actually indicative of good defensive performance?

I’ve taken this as a matter of course for as long as we’ve been tracking this stat, but it never occurred to me to track if it’s actually true or not. Are we just wasting our time tracking how many ballhawks a guy gets? Is our whole Ballhawk Index project worthless?

The answer to the second question gets a firm no from me, even without any research. Making plays on the ball is always going to be valuable, and tracking how many plays a guy makes is therefore, I think, self-evidently worth doing. You might as well keep track of something that’s always going to have some level of importance.

But can we use ballhawks as shorthand for overall good defensive performance? That took a little more doing, and the answer was satisfyingly surprising.

Using Pro Football Reference’s season finder tool, I searched for every player in the NFL last year that had at least five passes defensed. I started there because ballhawks are primarily a defensive back-driven stat, though some edge rushers and linebackers filtered into the data set as well. That search yielded me 203 names, and I looked up Pro Football Focus’s defensive grade and coverage grade for every player on the list, compiling them all into a spreadsheet.

Here’s what I found:

This chart shows the relationship between every player’s total ballhawks and their defensive grade from last season. In mathematical language, the data set indicates a correlation coefficient of about .447 between ballhawks and defensive grade, indicating a moderate correlation between the two. In plain language, the numbers show that there’s a moderately good chance that the more plays on the ball you make, the higher your PFF defensive grade will be.

Breaking down the numbers a bit more, of the 15 players who had more than 20 ballhawks last year, only four had PFF defensive grades under 80 and only two had PFF grades under 70. From there, it gets a lot messier, but I feel confident saying that a high volume of ballhawks pretty firmly establishes you as a very good defender. And a shoutout to Khalil Mack: at the vintage age of 32, he managed 17 sacks, 10 passes defensed (!), and five forced fumbles. He’s a true game wrecker.

The 20+ ballhawks club also was pretty uniformly good in coverage, too. Only three of the 15 players who broke 20 ballhawks last year had a coverage grade under 80, and one of them was Mack, who is never going to be counted upon for his elite coverage skills. 

But below 20 ballhawks, things get messy. Looking at the players who had between 15 and 18.5 ballhawks last year, you see a much wider range of defensive grades. I suspect some of this is partly due to injury; there are no doubt several players who would have had more plays on the ball had they played a few season, but topped out in this range because they missed a few games. Nevertheless, there are still fewer elite grades in this group.

And what about the inverse? Can you be a good defender without recording a bunch of plays on the ball? Absolutely yes. The data pretty clearly seems to show this.

Among players who had a season-long defensive grade of 80 or better, some players recorded as few as five plays on the ball. Again, we see some injury limitations here. Jaycee Horn’s five ballhawks came in just six games, for instance. But not everyone was limited by injuries. Jordan Battle had a defensive grade of 82.4 for last season, but recorded just eight ballhawks in 17 games. By the grades, at least, it seems he was doing his job, but that job did not result in a lot of appearances in the box score.

That trend holds a little less true for coverage grades, though, at least on the high end. Among players with a coverage grade of 80 or better last year, only Horn and fellow Carolina Panther Xavier Woods had fewer than 10 ballhawks. It appears that being good at coverage does have some correlation with producing ballhawks.

There are two interesting outliers worth discussing here as well. Tyrique Stevenson and Benjamin St-Juste have the unique distinction of producing a ton of counting stats while grading out pretty poorly; both had 20+ ballhawks, but they received defensive grades of 60.2 and 59.9, respectively. Why? Here are my theories.

Stevenson was a rookie in 2023, but played a ton of snaps for the Bears, lining up for more than 800 plays in 16 games. That’s a pretty big role for anybody, much less a rookie, and I think his statistical profile portrays exactly what he appeared to be: a talented player who makes a lot of plays (a high ballhawks total) but also gets exposed fairly regularly (a low defensive grade overall). 

St-Juste, meanwhile, was in a similar situation. He wasn’t a rookie in 2023, but he did see a big increase in his snap totals. After playing just 318 snaps on defense in 2021, he played 655 in 2022 before jumping all the way to 1,064 in 2023. I think he’s probably like Stevenson in that he’s a little over-exposed. He was good enough to merit a bigger role, but his role may have gotten too big.

I’m going to work to expand this investigation our in the future, probably to make it a little more position-specific and sprinkle in some other changes as well (if you have suggestions, you know what to do). But overall, I think this shows exactly what I’ve hypothesized: if you’re a good defender, chances are you’ll make a lot of plays on the ball, but you can be a good defender without making a lot of plays on the ball.