Caleb’s Close-Up: Volume III, Time to Throw

In part one, we saw that Caleb Williams generally struggled as a passer during his rookie season, and part two revealed that most of those struggles came on deeper passes. Today, we’ll continue diving deeper into Williams’ statistical profile to try and figure out where improvement is most needed going forward.

Play Action

Let’s start by looking at how Caleb Williams did on play-action drop backs compared to standard passing plays. Before I present the full data, I want to briefly explain two PFF stats that will be used:

Big Time Throw. These are best described as a pass with excellent ball location and timing, generally thrown further down the field and/or into a tighter window. In other words, these are really good, difficult passes that should result in highly valuable big plays. A higher % here is better.
Turnover Worthy Play. These include fumbles in the pocket, interceptions thrown, and interceptable passes that were not caught. A lower % here is better.

Both of these stats will admittedly have some subjectivity, but they provide a useful glimpse into how frequently a QB makes a really good play vs. a really bad one.

The table below shows how Williams performed in a wide variety of statistics in play action (orange) vs. other dropbacks (blue), and also includes Williams’ rank (in parentheses) out of 34 total NFL QBs who had at least 250 pass attempts. Information on the spread of all NFL QBs is also provided for each stat. Cells highlighted in green indicate Williams was among the top 25% of QBs in this category, while those in red indicate Williams was in the bottom 25% of QBs. All data comes from Pro Football Focus (PFF) unless otherwise noted.

(If you can’t view the full figure, click on it to open in a new tab. Sorry about formatting issues.)

A few thoughts:

First, I want to point out that NFL QBs are generally better in play action. The average accuracy and yards/attempt are both appreciably higher, and the turnover-worthy play and sack rates are lower.
- Caleb Williams mostly follows these trends. His accuracy and completion percentage both improved a bit, and his yards/attempt with play action took a huge leap, when he used play action. To be fair, his accuracy was still sub-par in both samples, but better is still better, and the yards/attempt went from one of the worst without play action to around average when play action was used.
- Williams also had fewer turnover-worthy plays and a steady sack rate on play-action plays, meaning there was basically no downside to using him in a heavy dose of play action.
- Given that, it is infuriating that Chicago’s collection of offensive coordinators called play action only 17% of the time last year, one of the lowest rates in the league.
Smart offensive play callers understand that play action makes QBs better and use play action at a high rate. Chicago’s play callers last year were not smart, but Ben Johnson, Chicago’s new head coach and offensive play caller, definitely is.
- Jared Goff, Johnson’s QB in Detroit, utilized play action on 37% of dropbacks in 2024, the highest rate in the league.
- This was not a new trend, as Detroit QB Jared Goff finished in the top five in play action drop backs in all three seasons with Ben Johnson as his play caller.
- As Windy City Gridiron’s Lester Wiltfong Jr. recently pointed out, we should expect significantly more play action in Chicago’s offense in 2025 and beyond. That will help make Williams a more efficient passer.

Time to Throw

Let’s look now at a different way to parse a QB’s pass attempts, based on how long it took him to throw the ball. PFF splits this into less than 2.5 seconds and more than 2.5 seconds.

There’s going to be some overlap with the play action data here, as the time to throw is generally longer on play action than other plays (average 3.1 vs. 2.7 seconds, as shown in the table above). So, we might expect to see some of the similar trends as above, but this is still a different enough way of parsing the sample that some results might be unique.

The table below shows the same stats as the play action one above. Once again, Williams’ data is shown ranked out of the 34 QB sample, information is provided about the spread of the 33 QBs, and cells for Williams’ stats in green or red represent Williams ranking in the top 25% or bottom 25%, respectively.

A few thoughts:

In general, slower-developing plays have more boom – as evidenced by a significantly higher yards per attempt and big time throw rates – but also more bust – as seen through significantly higher sack and turnover worthy play rates.
- Williams followed both of these trends, but was generally worse than average compared to peers on most stats in both categories. I don’t see a clear story here about him being particularly better or worse in either sample.
One thing that does stand out is that Williams had one of the higher rates of slow-developing plays, which matches what was expected of him coming out of college. He loves to scramble around and try to make something happen when things break down, as Kyle Morris pointed out on DBB last February.
- It’s also noteworthy, then, that Williams had one of the longest times to throw on plays that took over 2.5 seconds. It especially stands out in contrast to him having one of the shortest times to throw on plays under 2.5 seconds.
- Basically, most of Williams’ passes were either a designed screen, went to his first read, or resulted in him holding the ball for a long time trying to make something happen.
- Williams needs to improve at operating in structure to get to his 2nd and 3rd reads in a timely manner, which should yield a higher rate of passes that fall in the middle time and help the offense stay on schedule more frequently.
- With that said, scrambling around trying to make a play downfield is always going to be a part of Williams’ game, and the Bears need to get a lot better at teaching their WRs how to work open on plays where Williams breaks contain and buys them 5-6 seconds to do so. His yards/attempt on slow-developing plays needs to be really high to counteract the high sack rate.

Lessons Learned

In case you got lost in all the numbers above, here are today’s main takeaways:

Like most QBs, Williams was appreciably better when utilizing play action, but Chicago’s coaches barely called for play action in 2024. Expect that to change significantly under Ben Johnson, which should help Williams.
The majority of Williams’ passes were either really quick or really slow to come out, and he needs to get better at learning when to scramble and look for the big play downfield vs. taking a 2nd/3rd read or checkdown to avoid the negative plays (sacks, fumbles) that come when he holds the ball for too long.