Evaluating the Play of Daniel Jones in His First Season

Rowais Hanna
5 min readMar 10, 2020

--

For Giants fans, 2019 will be remembered as the year that the New York Giants have moved on from the biggest star in franchise history: Eli Manning. Taking his place is the relatively unproven, and highly-debated 6th overall pick in the 2019 NFL draft: Daniel Jones (DJ), also known as “Danny Dimes”.

Now that the NFL season is over, we have a decent sample size of games to evaluate Danny’s performance.

For this evaluation, I relied on the NFL 2019 regular season play-by-play data-set scraped and curated by @nflscrapR, which captures every potential measurable of any given play in a football game, for every game in the NFL regular season. Let’s see if this data can be used to substantiate or disprove a fan’s “eye test” evaluation of DJ’s performance — which can be best summarized in 2 words: “meh, rookie” — with meaningful statistical analysis.

2019 Giants Offense Before and After Danny Started

To begin forming an understanding of the team’s offensive performance, let’s take a look at the types of plays the Giants had on offense with vs. without Daniel Jones.

Playtype count of 2019 Giants Offense: with Daniel Jones and without him.

The chart above may seem very telling, but it’s telling a deceptive story! Since Daniel Jones started 12 out of 16 games (he would have started 14, but was injured for 2 games) this season, the counts of each type of play while he was on the field are naturally going to be larger. A more telling statistic would be the count of each play type on a per-game basis. Let’s take a look at that.

While this tells us a few things:

  • Giants were more comfortable throwing with Eli
  • Giants leaned more heavily on their run game with Danny in the game
  • Offensive efficiency, if measured by how frequently the Giants kicked field goals and punted per game, was relatively unchanged regardless of who was at the helm,

it’s not exactly insightful. Let’s examine the entire offense’s scoring rate next.

Giants Offensive Scoring Rate

In 2019, did the Giants score more with DJ than with Eli?

The Giants scoring offense was clearly better (as measured by TDs/game) on a per-game basis with Danny leading them. This, in part, validates the move to name him as a starter.

While scoring rate, as calculated above, is one way to look at the team’s offensive performance, it doesn’t tell us the whole story. To get a better understanding of the team’s offensive performance with Danny Dimes, let’s take a look at his specific performance.

Daniel Jones’ Touchdowns and Interceptions

Let’s take a closer look Danny’s specific performance, by examining his touchdown to interception ratio over time.

Now we’re onto something! The TD/INT ratio is a clear sign of offensive performance. Danny Dimes either had good games: multiple touchdowns and no interceptions, or bad games: multiple-interception games with only 1 touchdown.
However, football fans know this: it’s a team sport. There are multiple circumstances and conditions where even the best of quarterbacks would struggle. So while TD/INT ratio gives us a little more understanding of Danny’s performance he clearly either played well, or he didn’t — it doesn’t give us a full picture of his performance. Let’s see if we can dig up something more insightful from this data!

Evaluating Daniel Jones Performance using EPA

Expected Points Added (EPA) is a novel statistic that has started to become popular in NFL analytics circles. Created by Carnegie Mellon University Statistics & Data Science gurus Ron Yurko, Sam Ventura, and Max Horowitz, EPA is a first of its kind metric as it addresses an age-old question in American Football: given how much variability there is on a play-to-play basis, how do we evaluate offensive performance on any given play both objectively AND reproducibly? Sound familiar? That’s because this is exactly what we’re trying to evaluate for Daniel Jones here.
In a nutshell, EPA solves for this challenge by taking into account the probability of each scoring event to calculate the expected points of a given play, based on historical data of teams in the same situation for a given play, and the associated game conditions. Learn more about EPA here and here.

So what does Danny Dimes’ EPA in 2019 look like, and how does it compare with the rest of the league?

DJ’s Expected Points Added per dropback over time took a very similar path to his TD/INT per game: up and down. Of note, he had a couple of “average” performance game, and a couple where he performed above the league average, but quite a few more (6) where his play cost the team the game.

Summary and Final Thoughts

The exploration of Danny Dimes’ statistical performance confirms the hypothesis, and a football fan’s eye test: Daniel Jones had an up and down year, flashing moments of brilliance and moments of, well, a rookie in the NFL.

That’s to be expected for a rookie. The history of quarterbacks setting the league on fire in their first year is not a long one. More evaluation over time will tell if the Giants found a diamond in the rough who’s learning his way, or if we need to go back to the drawing board.

--

--

No responses yet