clock menu more-arrow no yes mobile

Filed under:

The quarterback as Rorschach test

There is broad disagreement about Daniel Jones from fans, analysts, and NFL people alike. Why is that?

NFL: Indianapolis Colts at New York Giants Vincent Carchietta-USA TODAY Sports

The end of the Daniel Jones saga with the last minute agreement between Jones’ agents and New York Giants general manager Joe Schoen on a four year, $160M second contract might have been expected to end the frenzy of discussion about him throughout the NFL community. Instead, it was only beginning, with some people describing the deal as a friendly one for the Giants, others viewing it as highway robbery for Jones, and yet others questioning why the Giants wanted to keep him at any price rather than selecting a quarterback in the upcoming NFL draft and starting over again.

Strong reactions were not uncommon, not just among fans but in the media and even within the NFL, especially with Lamar Jackson being tagged by the Baltimore Ravens, as Ed Valentine documented.

Why is it so difficult to get a read on how good a quarterback Daniel Jones is?

Choose your metric

Before you read further, stop for a moment, think back on the 2022 season, and make a list of what you think Jones’ five best and five worst games were. It should be easy to agree on most of these, right?

It turns out, though, that the NFL can’t even agree on this. There are two standard overall metrics of quarterback performance that are commonly used to rank them. The first is the traditional NFL Passer Rating, which has been around for ages, and the second is the Total Quarterback Rating (QBR) metric that ESPN developed a little over a decade ago. Both are commonly cited in discussions about the best quarterbacks in the NFL.

Below is a graph comparing Jones’ QBR to his Passer Rating for each of his 18 games this season. QBR is on a scale from 0 to 100, while Passer Rating ranges from 0 to 158.3, so a point in QBR corresponds to about 5/8 of a point in Passer Rating. That is represented by the dotted line.

Looking at the scatter below, you’d think I was plotting two different QBs against each other rather than the same QB in the same game.

Data from ESPN

Basically, points that fall above/to the left of the line are games in which QBR had a higher opinion of Jones than Passer Rating did, and points below and to the right games in which Passer Rating was more impressed than QBR. There are some amazingly stark disagreements:

  • According to Passer Rating, Jones had a fine opening game against the Titans while QBR thought he was awful.
  • QBR thought that Jones was stellar against the Bears while Passer Rating viewed him as “mid” at best.
  • QBR says that Jones’ five best games were Indianapolis, Chicago, Jacksonville, Houston, and the Minnesota playoff game. His five worst games were the Philadephia playoff game, Tennessee, Seattle, the Washington home game, and Carolina.
  • Passer Rating says that Jones’ five best games were Houston, Indianapolis, Tennessee, the Minnesota playoff game, and Baltimore. His five worst were the Philadelphia playoff game, the home Dallas game, Seattle, Detroit, and Chicago.

What makes a franchise quarterback?

To understand why QBR and Passer Rating disagree so much about Jones, we need to understand a little about how they are calculated.

Passer Rating is the NFL’s official passing statistic. It combines information about five different aspects of the passing game: Passing attempts, completions, passing yards, touchdowns, and interceptions. These are used as inputs to several equations that ultimately produce the passer rating. Anything else that the QB does or does not do is not taken into account, nor is the team on which he plays or the specifics of a given game.

QBR is the product of an ESPN proprietary algorithm that uses the concept of expected points added (EPA) on each “action” play involving a quarterback and assigns a fraction of them to the quarterback, regardless of whether the play was a pass, a quarterback run, a sack, a fumble, or a penalty. It is adjusted for game situations so that garbage time drives are devalued and the same yardage gained is valued more highly when it gets a first down or makes it more likely that will happen on the next play. It also corrects for the quality of the defense faced, and also adjusts for home field advantage.

This makes it easier to see how people can have very different opinions of Jones. Here are the top 16 QBs in 2022 ranked by Passer Rating, per Pro Football Reference:

Data from Pro Football Reference

And the same, but ranked by QBR:

Data from Pro Football Reference

Take your pick. There are things to like and things to hate about either metric. Personally, I question any metric that does not have Patrick Mahomes at No. 1, though Tua Tagovailoa was having a great season before it was derailed by the concussions he suffered. On the other hand, Tua was throwing to Tyreek Hill and Jaylen Waddle, so do they get some of the credit?

Jalen Hurts, Geno Smith, and Jared Goff, none of whom were on anyone’s list of great NFL QBs before the season began, were in the top ten in both metrics. Whatever you think about them, they all had great seasons no matter how you slice it. On the other hand, those QBs got to throw to A.J. Brown, Devonta Smith, D.K. Metcalf, Tyler Lockett, and Amon-Ra St. Brown, and Hurts and Goff played behind two of the best offensive lines in the NFL.

But no comparison is more fascinating than Jimmy Garoppolo (No. 3 in Passer Rating, No. 16 in QBR) vs. Daniel Jones (No. 13 in Passer Rating, tied for No. 6 in QBR). The comparison is made more difficult by the fact that Garoppolo played only 11 games due to his foot fracture in December. But in per-game statistics, Garoppolo had slightly higher yards and clearly more TDs per game, and only a slightly higher interception percentage. These seem to account for his lofty Passer Rating. Of course Garoppolo had George Kittle, Deebo Samuel, and Brandon Aiyuk to throw to, which doesn’t hurt.

QBR is a different story, though. What is missing from the Pro Football Reference QB ranking table is QB rushes, which are simply included in the same table as running back (and other position) rushing stats. And there’s the clue. PFR, along with many fans, media members, and even NFL people, are still not far along in the process of incorporating QB rushing in their assessments of QB value.

It’s a bit easier to isolate QB rushing using the Pro Football Focus rushing statistics, which allow one to at least isolate non-RB rushers. Here are their top 16 ranked by rushing yards:

Data from Pro Football Focus

(There are actually 17 entries above, since there is no way to get Deebo Samuel out of there. And Taysom Hill plays...what position, exactly? PFF calls him a tight end, but sometimes he’s a QB.)

Jones was fourth among QBs in rushing yards and rushing TDs, both of them just behind Josh Allen (who had an awful 16 fumbles vs. Jones’ 7). On the far right, we see that Jones tied for second among QBs in explosive (10+ yards) runs. Explosive plays are the bread and butter of modern offensive philosophy, but that is usually translated to mean downfield passes that get big chunks of yardage. Explosive is explosive, though, however you do it.

This helps to explain a decent part of the Passer Rating - QBR discrepancy that we saw in Jones’ individual game assessments. Against Chicago Jones only went 8 for 13 with 71 yards and 0 TDs, so naturally Passer Rating (which after all is about passing) viewed that dimly. On the other hand Jones had 6 rushes for 68 yards and 2 rushing TDs. There’s nothing like scoring TDs from 21 yards and 8 yards out yourself to raise your EPA. On the flip side, Jones completed an efficient 17 of 21 passes with 2 TD passes against Tennessee, which Passer Rating likes a lot. But Jones was intercepted in the end zone, in the fourth quarter of a one-score game, which is a cardinal sin in the EPA religion (and was to Brian Daboll, who reamed Jones out on the sideline), while rushing for only 25 yards on 6 attempts and being sacked 5 times for 32 yards lost (his highest of the season), and this gave him a dismal QBR for that game.

Other passing indicators

There are even simple advanced metrics that cast a more or less favorable light on Jones as a passer. Jones was the fifth-most frequently pressured QB in the NFL in 2022:

Data from Pro Football Reference

Yet he had the fourth-lowest percentage of bad throws:

Data from Pro Football Reference

And the highest percentage of on-target throws in the NFL:

Data from Pro Football Reference

On the other hand, Jones had the second-lowest intended air yards per pass attempt, and it’s easier to be on target with short passes than long passes:

Data from Pro Football Reference

Of course with the Giants’ 2022 offensive line who could blame Mike Kafka for scheming short passes? And look at the company Jones is keeping in that top 10 - Herbert, Stafford, Brady, Burrow, all of them with reputations as downfield bombers.

PFF compiles statistics on “big time throws (BTT),” which are passes thrown downfield or into tight windows with good location and timing, and “turnover worthy plays (TWP),” which are passes likely to be intercepted or fumbles that could have been recovered by the opponent regardless of whether a turnover actually occurred. Here are Jones’ four seasons:

Data from Pro Football Focus

Under Pat Shurmur Jones began his career as a middle-of-the-pack QB in BTTs but fourth highest in TWPs. He also had his career high in TD passes (24) under Shurmur. The TWPs went way down during the Joe Judge-Jason Garrett era, initially with an increase in BTTs, which seemed like progress except that the TD passes cratered. But then came the 2021 season when Jones was put into a shell, with even fewer TWPs but hardly any BTTs. It appeared to be the result of injury: Jones had 6 BTTs through Game 4 (the OT victory in New Orleans), but only one in the seven games after that before his season-ending neck injury. Game 5 at Dallas, when Jones suffered a head injury trying to score, seems to have been the turning point.

The first year of the Daboll-Kafka regime saw Jones maintain his very low number of BTTs, but with an increase in TWPs. The small BTT number was highlighted by Robert Mays and Nate Tice on The Athletic Football Show as a red flag for Jones ever becoming an elite QB. Was that the result of the dire situations in the OL and WR rooms? A big question for 2023 will be whether the Giants can open up the passing offense (as Shurmur did) without having the turnovers skyrocket (as they did in the one Shurmur-Jones year). The best QBs in the NFL have about a 2:1 ratio of BTTs:TWPs.

Joe Schoen alluded to Jones’ rookie success, and by omission his lesser production under Judge and Garrett, at the press conference after the Jones contract signing, perhaps a harbinger of things to come:

Q: You came in here with Dabs (head coach Brian Daboll) and were evaluating everybody and certainly did not have to fall in love with Daniel Jones as a quarterback. You didn’t draft him. By giving him this deal, are you saying that you feel you can win a Super Bowl with him?

A: Yeah, that’s the goal. Everybody’s goal is to win a Super Bowl. I think Daniel, he played well his rookie year. He played well for us this past year. I think the coaching staff has confidence in him. As an organization, we have a confidence in him. We’re going to continue to build the team around him. That’s the ultimate goal.

Here is Jones’ 2022 regular season summary of advanced statistics from the Tucker Boynton Quarterback Cards app:

Data from NFL Quarterback Cards

The green (= above average) areas in this chart correspond pretty well with my personal impressions of how Jones played this past season in specific games. (CPOE = completion percentage over expected, DAKOTA = EPA/play + CPOE). Overall they correspond more closely to QBR than Passer Rating.

A simple suggestion

Moving forward, shouldn’t quarterbacks be evaluated on everything they add to an offense? At the very least, let’s consider both passing and rushing yards and TDs.

If we look at total QB offense and sort by total yards, then we get the following QB rankings for 2022:

Data from Pro Football Reference and Pro Football Focus

Jones ranked 11th in total yards in 2022. Coincidentally, that is about where the annual average salary of his new contract will rank once Joe Burrow, Lamar Jackson, Jalen Hurts, and Justin Herbert sign their second contracts within the next year. That suggests that his new contract is fair value for both sides. If we rank him by total yards on a per game basis, he ranks 17th, and ranked by TDs per game, 20th. That suggests that his contract is an overpay.

Any of these bulk production metrics, though, ignore the role of the receivers and offensive line a quarterback has to work with. No one will argue that Jones’ receivers and OL gave him much of an advantage this past season. That is why advanced metrics such as QBR exist - to try to isolate the QB’s role in the success the offense has or the lack thereof.

Jones isn’t the only QB who suffers from the focus on passing statistics alone. Lamar Jackson is struggling to get his second contract, which deserves to be higher than Jones’, although that situation is muddled by his leg injuries and desire to have a 100 percent guaranteed contract.

And consider poor Justin Fields. He is far and away the most dangerous running QB in only his second season, leading the league by a mile in rushing yards and second in rushing TDs. There are still legitimate questions about him as a passer, but like Jones the cupboard has been pretty bare for him in terms of receiver and offensive line help. Yet there was considerable sentiment among fans and media for the Bears to move on and use the No. 1 draft pick on a QB before they did the smart thing and got a haul of draft a top-flight WR in D.J. Moore - for that pick.

In the end, though, all that matters is winning. Jones led the Giants to 9-7-1 and the final eight in the playoffs, and the green areas in the Quarterback Cards figure above correspond pretty well with games the Giants won. So those metrics pass the eye test for me. If there’s more green over the next few years, Giants fans will be happy and maybe even the media and the NFL will change their tune about Jones.

That’s the QB Rorschach test. Some will perceive a franchise QB, some will see a QB who should have been replaced by a 2023 draft pick. At the moment, it’s in the eye of the beholder.