What Can We Expect Jalen Hurts to Be in 2021?

By far, the number one question for the 2021 Eagles is what will Hurts be? Not only is this important to their season, but maybe more importantly long-term, it will dictate whether the Eagles need to use their massive draft capital to trade up or for a quarterback or if they can use 2 or 3 first round picks to dramatically up-skill other positions.

I wrote previously on metrics that matter for different positions, including quarterback, and here will give a view on what we could expect from Hurts. While I spend a ton of time on analytics on this site, I don’t believe in models to predict an individual player’s season as the best models I’ve seen people promote have R²s of 0.60 (and usually much lower) which means they only are able to explain 60% of a player’s performance. But data can intelligently inform an opinion by removing noise and centering on what matters.

Here are five key things to consider when looking at Hurts’ first year and what he can be in 2021:

1. Hurts’ poor rookie rankings don’t matter

Everybody knows how poorly Hurts graded out in his first season:

Graded 40th out of 41 QBs by PFF in total offense with a grade of 56.2 (just behind new Eagle Nick Mullens)
32nd in the league in clean pocket passer rating at 86.2 (between Andy Dalton and Nick Foles)
And last in the league in completion percentage over expected (CPOE) and 29th in expected points added (EPA) below (from the great site RBSDM.com)

The history of QBs starting with his numbers and actually turning into a good QB is not good. Since 2005, twelve QBs have had a CPOE of -6 or worse with only one, Matt Stafford in his rookie season, becoming an above average QB (the rest were the very uninspiring list of Brady Quinn, Jamarcus Russell, Kyle Boller, Derek Anderson twice, Matt Hasselback, Trent Dilfer, Kellen Clemens, Akili Smith, Andrew Walter, and Bruce Gradkowski).

The issue is that Jalen’s sample size is way too small with only 185 dropbacks and 4 starts (3.5 starts really). If you pull the same EPA and CPOE chart using only the first four starts for the starting-quality quarterbacks drafted in the last 10 years, below is what you get. Note that for both EPA and CPOE, a higher number is better, so the top right are QBs with a better completion percentage above expectations and a higher expected points added while the bottom left are QBs with poor completion percentages over expectations and low or negative expected points added per play.

Jalen is now in very good company with Josh Allen, Lamar Jackson, and Andrew Luck who also looked bad in their first four starts. Does it mean he will be Jackson, Allen, or Luck? No. But nobody should look at data on a chart using a QB’s first four starts and make a declaration. Are there issues he needs to clean up? Absolutely. But way too much is made of these and other similar metrics without considering further context on his accuracy and ability to drive points.

2. Hurts had an elite college CPOE which does matter

Josh Hermsmeyer (@friscojosh) at FiveThirtyEight has a great article here that shows college CPOE is a better predictor of success in the pros than other metrics. From his article:

The test of a good metric is that it is stable over time (for example from college to the NFL) and that it correlates with something important or valuable. Completion percentage over expected is slightly more stable than other advanced metrics like QBR. CPOE is also the best predictor of NFL yards per attempt. Since yards per attempt correlates well with NFL wins, and winning is both important and valuable, we’ve found a solid metric.”
Josh Hermsmeyer, “The NFL Is Drafting Quarterbacks All Wrong”

CPOE adjusts completion percentage for the depth of target (to penalize QBs who only dump off and reward deeper throws) and strength of the defense. In Hurts’ senior season at Oklahoma, he had the fourth highest CPOE among all Power 5 QBs since 2011 at +11, behind Russell Wilson, Burrows, and Justin Fields (who I still believe the Falcons, Panthers, and Lions will regret passing on…) Again, does this mean he will be Russell Wilson? No, but it does show that his rookie season stats are probably the anomaly and as he gets more reps, we should expect his CPOE to return to his norm.

3. Hurts’ mobility is more valuable than most realize

Everybody knows Hurts’ primary rushing stats from his limited action last year:

354 rushing yards on the season, with 296 yards in weeks 13-17 behind only Lamar Jackson (430 yards)
11 explosive runs in weeks 13-17, again only behind Lamar Jackson (15)
5.8 yards per rush on 51 total attempts in weeks 13-17, 4th among QBs in the league behind Jackson (7.7 ypa), Russell Wilson (6.1 ypa but on only 7 attempts), and Ryan Tannehill (8.8 ypa, also on only 7 attempts).

Projecting these to a full season gets silly as Hurts was on pace for over 1,100 yards and 40 explosive runs. He won’t run that much as those would be all-time great season numbers, but even half of this pace would put him in the top few QBs in the league, with only Jackson (1,105 yards) and Murray (773 yards) ahead.

But mobility is more than running – what isn’t written about as much is Hurts’ ability to extend plays. He had a 3.94 second time to throw (TTT) while under pressure, 3rd best in the league behind Lamar Jackson and Deshaun Watson. Below shows average time to throw while under pressure vs. allowed sack rate with him being better than the league at avoiding sacks and giving himself more time to create a play. Again, Hurts is showing up with the familiar group of other QBs that extend plays.

I mentioned in previous posts the Football Outsiders research on how pressure affects quarterbacks, with DVOA dropping a massive 108 points on average when a QB is under pressure. When you look at the past several years of data on the QBs with the best under-pressure performance, the same names show up – Mahomes, Dak, Rodgers, Lamar, Deshaun, and Brees. Again, the QBs that can extend the play show up here (Brees not so much because he extends plays, but because he is just really good and will destroy a blitz, consistently over a 100 passer rating against pressure throughout his career).

Hurts did not consistently turn his scrambles into positive plays in 2020, however, finishing with a 59.1 passer rating against pressure. His primary issue was a way-too-high turnover worthy play rate (TWP) of 6.7%, 7th highest in the league. But, if he cuts down turnovers, Hurts’ ability to extend the play will benefit the offense greatly.

4. Hurts has shown an ability to generate explosive plays

In his limited play time last year, Hurts showed that he has the willingness and ability to generate explosive plays which is increasingly important for offenses today. In 2020, Hurts showed the following:

The highest average depth of target (aDOT) in the league by a significant margin at 10.1 yards (Drew Lock was next at 9.7).
A passer rating of 77.1 on throws beyond 20 yards with a big time throw percentage of 24% (defined as a ball thrown with good placement and timing down the field).
In the intermediate 10-19 yard range, Hurts had an even better passer rating of 101.4.

Hurts was near league average in explosive passes but was elite on explosive runs. Below shows QB explosive passes (Y axis) and runs (X axis), with Jalen near Lamar Jackson when projecting explosive plays per 600 dropbacks.

Below is a second view, totaling explosive passes and runs together with Jalen second in the league on a full year projection. Looking at Sharp Football Stats metrics on team explosive plays (here), the Eagles were 22nd in the league while Wentz was starting and moved up to 12th in the league in weeks 13-17 when Hurts started.

Again, Hurts may not keep this full-year pace, but even a percentage of what he showed in 2020 is extremely valuable.

5. Hurts must protect the ball better

Hurts’ completion percentage needs to improve, particularly in the short-range as explained above, but the biggest area of concern for me is his turnovers. Unlike his CPOE which is at odds with his college history, turnovers unfortunately cannot be explained away as they have followed him throughout his career.

Below shows his interception rates – except for his sophomore year at Alabama, Hurts has consistently been above a 2% interception rate.

Season	Attempts	Interceptions	Int %
Fr – Ala	382	9	2.4%
So – Ala	255	1	0.4%
Jr – Ala	70	2	2.9%
Sr – Okl	340	8	2.4%
Yr 1 – PHI	148	4	2.0%

For the Eagles last year, interceptions did not just come on deep passes either, they came in all ranges with two beyond 20 yards, one in the 10-19 yard range, and one in the 0-9 yard range. If this rate maintained, it would put him in the bottom third of the league in interceptions. Below is just one example from the week 16 Cowboys game where Jalen just made a really bad decision trying to keep a play alive and throwing into 2-3 defenders (his earlier interception in the Cowboys game intended for Fulgham is the same, really poor decision-making). This is Wentz-like without the same arm strength.

Perhaps worse than his interceptions is his fumble rate which has also followed him from college. Ironically, his very first snap at Alabama was a fumble and he went on to total 11 his freshman year. At Oklahoma, he ended his college career with 9 fumbles and Lincoln Riley specifically calling it out. In 2020 with the Eagles, Hurts again had 9 fumbles in only 185 dropbacks, which is an off the charts fumble rate.

Below shows all interception and fumble rates for all 2020 QBs with over 100 dropbacks. You want to be in the bottom left and Hurts is light years away, amazingly at almost 5% of dropbacks with a fumble. CJ Beathard with only 114 dropbacks is next closest at 2.63% and then Lamar at 2.41%

Hurts’ decision-making and ball security is, to me, by far the biggest concern and what will wreck his career.

What are the keys to Hurts’ 2021 season?

Hurts gets criticized a lot for missing receivers on plays (all QBs do), a weak arm (arm strength doesn’t equal performance), and accuracy (I think he improves here as described above). Based on the data, there are three key things to watch with Hurts in 2021:

Does the turnover rate drop? By far my #1 thing to watch with Hurts is the turnover issue. He projects to 25-30 total turnovers over a full season and if he cannot significantly improve it, the Eagles will be looking for another QB next year as he just won’t be able to win.

Does he improve short yardage accuracy? Sirianni’s offense will absolutely utilize more shorter throws. When he came to the Eagles and explained his offense, one of his key statements was “Another thing we like to do is get the ball to our players within 5 yards of the line of scrimmage, get the ball to our players running”. Last year in Indy, Philip Rivers had his second lowest aDOT for his career at 7.6 yards, down from 9.0 yards in 2019. Andrew Luck also showed a marked decrease in aDOT under Sirianni at 8.1 yards in 2018, down from 9.0 in 2017. Short yardage, for some reason, was Hurts’ worst range with only a 72.6 passer rating and a PFF grade of 42.6. As explained above, I expect Hurts to improve his accuracy and CPOE across the board, but it will become even more important in this offense on short throws.

Does he continue to generate above average explosive plays? More shorter throws will not mean explosive plays are not important – last year the Colts were 5th in the league in explosive plays. Hurts’ ability to generate with his legs and with extending plays and taking shots downfield is his differentiator and where he needs to provide value.

I said earlier I don’t try to create predictive models for individual players, but I do think data can show what a player may become. In my post on offensive metrics, I showed how CPOE+EPA and rushing yards per 600 dropbacks is a good model for quarterback value. I think Hurts could fall in the 13-16 Approximate Value (AV) range, which would put him in the top half of the league. My assumptions and reasoning for this are the following:

Hurts’ poor rookie CPOE is an outlier given his history in college and even if he moves from a CPOE of -8.3 to just league average, it lifts him to the middle of the league.
Hurts’ ability to create with his legs alone is elite and even assuming a dramatic reduction in his pace from his rookie season (better OL protection, better WRs), it is not hard to see him running for 700-800 yards.
The Eagles offensive line will be better just because repeating the number of games lost to injury would be a statistical surprise – they were so ridiculously injured last year and the injuries piled on top of each other. As mentioned above, better protection has a giant impact on DVOA.
The receivers will be better – I wrote previously here (“The Analytics of the Eagles Draft Philosophy – Wide Receivers”) on why Reagor showed high-level traits on separation and YAC that translate to high performance receivers and here (“DeVonta Will Hit the Over on Everything”) on why DeVonta will outperform expectations.

Below shows rough placements of what Hurts could be assuming he improves accuracy only to league average and that he does hit 700-800 yards of rushing. Both of these are very reasonable expectations and I expect Hurts to be in the top half of the league.

7/14/21 Update:

If you don’t follow Thomas Petersen (@thomasrp93), you should – he has great video breakdowns of the Eagles. He just did a video on every Hurts throw from 2020 which is absolutely awesome and provides a ton of context on the good and bad of Hurts this past year.

If you have 61 minutes, go watch this. This supports a lot of the above with my key takeaways:

The OL was awful (especially Pryor, Toth in limited time, and even Mailata wasn’t great in the Dallas game). The pressure caused a lot of the issues that showed up on stat sheets.
The number one thing for Hurts to improve on is getting the ball out slightly quicker – you will hear Thomas say this frequently in the video.
In general, I don’t agree with the common concerns on Hurts’ arm strength. The one thing that is related is when he does not step into his throws, there are a bunch of throws where he is just using his arm, usually do to pressure and sometimes due to rollouts, but it is noticeable (as with many QBs) the difference in Hurts when he isn’t setting. Improved OL play will help here but this is something for Hurts to clean up.
And unrelated, I am more optimistic on Reagor after watching these – Reagor’s separation abilities are underappreciated (which I wrote about separately).