Is the LaFleur/Rodgers offense better off without Davante? A look into the data.

What the numbers say about the Packers offense with and without Davante Adams.

A 38-10 loss usually prompts a lot of questions. What went right? What went wrong? What can be fixed and how to fix it.

Editors Note: This is a guest post submitted by one of our readers that I thought would be an interesting read for everyone.

Packer fans began drawing connections between last years’ losses and the loss to Tampa Bay. For many fans, it was hard not to be frustrated watching an offense that moved the ball, seemingly at will, in the first four games, suddenly sputter and stall. It was reminiscent of the whiplash Packers’ fans felt when Davante Adams returned from a four game absence in 2019. Packers’ fans were excited. The offense had been roaring in his absence under new coach Matt LaFleur, and surely the addition of Adams would be like dumping nitrous in the tank. Instead, the Packers were thumped by the Chargers, limped past a mediocre Panthers team, and then tripped out of the Bye and got stomped by San Francisco. Why, fans wondered, did adding a top five receiver somehow make the offense worse? The same questions are being asked again. Is the offense actually better off without Davante?

The search for answers began with some simple box score scouting. Simply, in the six games without Davante Adams between 2019 and 2020, the Packers’ offense scored 30+ points in five games. They only eclipsed that mark twice in the 15 games Adams’ started and finished in the LaFleur era. Modern statistics lets us do much better than box score scouting though.

To evaluate the passing offense, and Rodgers and Adams’ impact on it, we can use advanced data and analytics to give us a much more in depth view of what is happening game to game.

Ben Baldwin’s website, www.rbsdm.com/stats gives us Rodgers and Adams Expected Points Added in each game both throwing and receiving. For those not familiar, expected points added (EPA) is a statistic that tries to value each play in terms of its effect on the expected points it helps a team gain by comparing down, distance, and field position at the start and end of a play. The analytics site Inside the Pylon helpfully explains EPA by referring to Super Bowl XLIII. In that game the Arizona Cardinals had first and goal from the Steelers’ 1 yard line. At the start of the play, their EPA was +6.97, reflecting the near certainty of a touchdown from teams in this position. James Harrison famously intercepted the ball and returned it 100 yards for a score. The expected points on the end of the play was -7.00. The difference between those two numbers is -13.97 reflecting the loss of a TD for Arizona and the gain of a TD for the Steelers. EPA attempts to capture the effect of every play in generally the same fashion. It’s a helpful tool that recognizes that a 5 yard pass on 3rd and 10 is a lot different than a five yard pass on 4th and 4.

For the purposes of evaluating Adams’ and Rodgers’ play in games under Matt LaFleur, we can look at their total EPA in passing situations. We can also look at the number of targets Adams gets when he plays and what percentage of Rodgers’ total targets he makes up. Courtesy again to Ben Baldwin’s website for this data. It also makes sense to look at how long Rodgers takes to throw from snap to release. It’s become well understood that, generally speaking, throwing the ball faster is better. The NFL’s own “Next Gen Stats” can give us this data.

All that data is represented in the chart below. Left to right the columns note the year, Week and opponent; Rodgers total passing EPA; Adams’ receiving EPA; Adams’ total targets; how many targets the whole offense had; the percent of Rodgers total targets thrown to the most heavily targeted receiver; and Rodgers average time to release. Finally, we note whether Adams played, he and Rodgers raw production, the final score and a few relevant notes.

CLICK ON IMAGE TO EXPAND TO FULL SIZE

 

When looking at a big mash of data like this, it can be helpful to look for trends that jump off the page. The most eye-catching trend has to be that in games Adams started and finished (15 total), he was the most targeted receiver in 13 of those contests and by a whopping margin. In games where Adams was the most targeted receiver he was thrown to 35% of the time. In games where he wasn’t the most targeted receiver, the most heavily targeted player averages only 24%. Clearly, Rodgers likes to throw to his number one, but that doesn’t tell us much on its own. Adams is a top 5 or 10 wideout in the league and has been playing with Rodgers for a long time. Getting the ball to him not only makes sense, it’s to be expected. He’s probably open quite a bit.

The only other stat that immediately leaps off the page is Rodgers total EPA in games without Adams. As expected, he posted some really exceptional numbers in those six contests; he averages 17.35 total EPA on dropbacks per game. In games Adams played and finished, that number averages only 2.29 total EPA for Rodgers. That’s certainly an eye-popping difference, but six games is a small sample in the NFL, and we’re not accounting for quality of defense, so we’ll set that aside for now.

A theory we can quickly dismiss is the notion that Rodgers’ time to throw is much worse when Adams plays. Most quarterbacks are much better when they throw quickly and in rhythm. In games Adams started and finished, Rodgers average time to throw was 2.92 seconds. In games where Adams didn’t play, his average time to throw is 2.80 seconds. That’s certainly faster, but probably just evidence in general of the offense being more productive as a whole. It’s certainly not strong evidence that Rodgers is just patting the ball waiting for Adams to come open.

Rodgers does target Adams a lot though, so maybe it’s a question of “as Adams goes, so goes the offense.” Rodgers two best games where Adams started and finished were Week 13 of 2019 at the Giants and Week 1 of 2020 at Minnesota. Rodgers posted total EPA of 25.4 and 18.3 respectively. Adams posted EPA of 12.4 and 7.2. Rodgers targeted Adams in those games at a rate of 35% and 41%. There seems to be something to go on here.

The issue though is that because Rodgers targets Adams so much, there’s a strong correlation between Rodgers EPA and Adams EPA. When Rodgers does well, so does Adams and when Rodgers does poorly so does Adams. There’s nothing here that says whether or not either player drives the other’s success, or the success of the offense as a whole.

So let’s move back to the strongest trend we initially spotted, the most targeted receiver, and take a look at games where Rodgers' most targeted receiver got around 1/3 of the total targets or more (rounding up). That happened 10 times. In those 10 games, Adams was the most targeted receiver nine times (the one outlier was MVS in Week 3 of 2019 vs. Denver when he got 37% of the targets). In those nine games, Rodgers average total EPA on dropbacks was 1.62. That's not great. In the nine games where Adams received around a third of the total targets or more his own average total EPA was 3.94. Also not great, but pours a bit more water on the notion of Adams “driving the whole offense.”

We’re starting to see some glimmers of a conclusion, so let’s flip the script and take a look at all the games where Rodgers' favorite receiver got no more than 25% of his targets. That has happened seven times. Five of those times were games Adams did not play, one game was in 2020 Week 2 when Adams got hurt after collecting three catches for 36 yards, and the remaining game was Week 17 of 2019.

We see a huge difference here. In the seven games where no target gets more than 25% of the looks, Rodgers average total dropback EPA in those games was 15.44. It certainly seems like Rodgers is simply a lot better in games where he spreads it around as opposed to games where he focuses on just one guy. The obvious criticism here is that most of this EPA bump is coming from those six games where Adams didn’t play and we’re trying to figure out whether or not Adams not playing is the reason for that.

So let’s expand the sample to include all 13 games where one player was targeted less than a third of the time. This includes the six games Adams missed, but also seven others where he played. In those 13 games, Rodgers posted an average total dropback EPA of 11.72. That’s still a massive increase in performance and gives us a much more reliable sample size. Isolating just the seven games where Adams plays and no one gets a third or more of the targets, Rodgers average total EPA per game on dropbacks is 6.89, still a significant increase over games where he targets one receiver a third of the time or more.

Where does this leave us then? Is the offense better in games Adams didn’t play? It was, but not because Adams was missing. There seems to be solid evidence that Rodgers is a lot better when he spreads the ball around. Rodgers throws to Adams a ton when he plays. Without controlling for defense it’s hard to say if this is because of choice or necessity (that is, is Adams the only one open, or is Rodgers forcing him the ball). This is where film study meets analytics and both can work hand in hand to raise useful questions. But we can see that no one steps into that “favored target” role when Adams doesn’t play. When the offense is at its best, and Adams doesn’t play, Rodgers isn’t throwing a huge portion of his targets at someone else. Simply, when Rodgers has had the opportunity to heavily feature someone else, he hasn’t. It has generally been Adams or no one getting targeted over and over. 

To be clear, there’s a lot of caveats to all of this. It’s a set of small samples. There’s no control for quality of opponent. There’s no control for win probability (garbage time stat padding). The benchmarks picked (a third of the targets) were arbitrary. Setting the line elsewhere produces a different result.

But we wanted to look at some advanced stats and some deeper numbers to see if the counterintuitive idea that the offense might somehow be better without Davante Adams had any merit. It certainly doesn’t seem that way. The offense seems to be better whether he plays or he doesn’t when Rodgers spreads the ball around. For whatever reason, that has happened a lot when Adams didn’t play and happens less frequently when he does. Obviously, how Rodgers and the offense performs when Adams plays bears watching, and the data we have suggests that passing production and how much Rodgers spreads the ball around have a strong correlation. Keeping an eye on offensive production and to what extent Rodgers spreads the ball around seems like a good next step.

NFL Categories: 
4 points
 

Comments (38)

Fan-Friendly This filter will hide comments which have ratio of 5 to 1 down-vote to up-vote.
arthurl's picture

October 22, 2020 at 06:20 pm

Well the reason they got beat last week had nothing to do with Adams. Rodgers got rushed, rattled, and knocked down early and often. Even when the game started he was taking hits and the Bucs defense dominated the line of scrimmage. Rodgers got unsettled it appeared as I watched the game. If he doesn’t have the protection, it doesn’t matter much about the passing game. They also got dominated in the run game. I hope it was a one-off, but appears the Bucs will have their number just like the Niners did last season. They are too physical for the Pack.

We as fans better hope if they do play them again, it’s in GB where team has a shot.

7 points
7
0
Oppy's picture

October 22, 2020 at 10:09 pm

Rodgers got rattled because he threw a pick-6 and another interception over the course of three plays.

The Buc's defense clearly took advantage of Rodgers' being rattled, and compounded it; but their defense didn't cause it.

Rodgers' composure fell apart of his own accord. It was easily seen on his face before the defense started battering him.

-1 points
2
3
edp1959's picture

October 23, 2020 at 04:58 am

Ignorance at it's best.

0 points
2
2
jannes bjornson's picture

October 23, 2020 at 02:50 am

Hard to pass the ball and block for yourself at the same time.

2 points
3
1
Packers2020's picture

October 23, 2020 at 07:20 am

Sorry, arthur that is not completely accurate.

If you go back and look at the pick 6, Malik Taylor was wide open over the middle. AR threw off his back foot with no pressure.

There were 3 other plays in the series after that pick 6 where WRs were open but AR did not throw on time. Those drives end different if completes those.

Not all ARs fault but alot of it was.

4 points
4
0
Thegreatreynoldo's picture

October 23, 2020 at 09:19 pm

Somewhere one should note that on both interceptions AR's target was Adams. And that Adams wasn't open on the first INT and was in traffic on the second, though the pass hit his hands. Sterling Sharpe caught those.

0 points
0
0
Coldworld's picture

October 22, 2020 at 06:40 pm

If this is even an issue it’s not about the Packers being better without him: they are not. It’s about Rodgers not getting over reliant on one receiver. Adams is merely the latest.

6 points
9
3
edp1959's picture

October 23, 2020 at 04:59 am

When you only have 2 seconds to throw the ball where are you going to look first?

2 points
2
0
Coldworld's picture

October 23, 2020 at 07:35 am

If you actually look at the film you will find that Rodgers looked very hard and for a long time.

Later, your point has some validity, but that was after the interceptions.

1 points
2
1
Leatherhead's picture

October 23, 2020 at 10:21 am

Two seconds isn’t much time to you or me, but to a pro who has been coached and been practicing for years it’s all they need. Some QBs look off their target before coming back to it.

0 points
0
0
Oppy's picture

October 23, 2020 at 04:57 pm

NFL game speed, you don't understand what it is

0 points
0
0
NickPerry's picture

October 23, 2020 at 05:22 am

THIS by Coldworld is right on the money. Myself and others were/are concerned because Rodgers DOES tend to lock onto Adams and forget about other receivers when Adams is on the field. Rodgers HAS to find that balance. He HAS to look for others without the number 17 on his jersey. The offense just works better when EVERYBODY is involved.

8 points
8
0
WestCoastPackerBacker's picture

October 23, 2020 at 07:14 am

Maybe it's about the fact they haven't had much for a reliable receiver past Adams in some time. Since Cobby got old and injured and Jordy slowed down some and moved on, there's been no one waiting in the wings as the long progression of great Packer WRs slowed after Ted retired. Rodgers knows how to spread the ball around just fine like when he had Jennings, Driver, James Jones, Jordy and then Cobb (throw in some Jermichael Finley and even Richard Rodgers and Andrew Quarless). I mean GB won a Title with Andrew Quarless as the starting TE! There's just been no one after Adams to develop into the reliable WR; and he used to have 3 or 4 of them. In a league where turnovers signal losses overall; he's been smart to focus on Adams.

-1 points
0
1
Leatherhead's picture

October 22, 2020 at 06:43 pm

I’m a guy who thinks that WRs are overvalued in general, but a good pass-and-catch combo puts a lot of pressure on the defense.

0 points
1
1
OldPacker's picture

October 22, 2020 at 08:49 pm

Very interesting .. if you just casually watch the game you see AR looking at Adams during the play ..doesn't seem to look off... This past week opened my eyes .. AR was forcing the ball to Adams ... amazing this article addresses the issue..MM / AR did the same thing with Finley .. One game I think the 1st 7 plays every ball went to Finley over the top/... Glad you found some data to support the issue,, Hopefully when they watch the tape they see the problem. Some games you need to use Adams as a decoy .. considering most teams game plans focus on Adams .. For it to work he has to find a play that works like .. ( Jordy & AR ) those sideline ( back shoulder) passes could not be stopped..
Cheers

5 points
7
2
edp1959's picture

October 23, 2020 at 05:01 am

When you only have 2 seconds to throw the ball where are you going to look first?

-1 points
1
2
OldPacker's picture

October 24, 2020 at 05:57 pm

true

0 points
0
0
Fubared's picture

October 22, 2020 at 10:08 pm

Good article. My two cents. First Adams runs good routes, has good hands and proven himself to where he is Rodgers fav. He is Jordy now. He is Rodgers not just go to guy but savior. His comfort zone, his trust zone.
Everyone one else is just, I'll pass you the ball depending on the play, pass rush etc but you won't be a feature creature I have that in Adams.
The bad if Adams is open good if he's not, not so good.
4odgers makes it easier on D's by concentrating more on Adams instead of being worried by multiple options. Or teams ain't covering a Sternberger if films show he rarely gets the ball. It puts more pressure on those few who do get the ball.

2 points
2
0
Oppy's picture

October 22, 2020 at 10:11 pm

It's almost as though the offense runs best when the offense is run as intended, which is simply to get the ball out on time to the open receiver in progression.

Man, I think I've heard that somewhere before.

5 points
8
3
edp1959's picture

October 23, 2020 at 05:04 am

Show us your proof offense wasn't run as intended. How much progression can happen in 2 seconds. Wow, what ignorance.

-5 points
1
6
Coldworld's picture

October 23, 2020 at 07:37 am

If you think that offense ran as planned after the interceptions then you are right to acknowledge your ignorance.

0 points
2
2
Oppy's picture

October 23, 2020 at 05:07 pm

edp, one time when I was a young adult working as a roofer, a 19 year old laborer cut his finger with a razor knife. I went over and asked him how bad it was, and he told me it was real deep.

Seeing the amount of blood, I asked him if he cut clean down to the bone. He looked at me and laughed and told me I was hilarious, I should stop being a clown and go get him some help. I asked again, is it to the bone? He looked another co-worker and said, incredulously, "Hey, Tommy, listen to this Einstein over here, he just asked me if I cut my finger down to the bone!" and started laughing.

I asked him what was so funny. He looked me dead in the eye and said, "Are you serious? Fingers ain't got no bones in 'em. If they did, you wouldn't be able to bend them." He wandered off over to tommy, told him what I said, and then said, "Next thing this guy is going to try to tell me the sun causes cancer."

He was so ignorant that he was sure I was a fool. There was literally no way I was going to be able to explain to him.

This feels a lot like that, edp.

1 points
1
0
Oppy's picture

October 25, 2020 at 10:54 am

EDP is ignorant of the fact that nfl passing plays are generally designed for the QB to deliver the ball in less than 3 seconds from the snap.

0 points
0
0
Gman1976's picture

October 22, 2020 at 10:26 pm

I was unable to watch the game due to a family commitment, but immediately after I read how poorly the offense did I thought of last year. Correct me if I'm wrong, but wasn't the offense running well until Adams returned from an injury? Afterwards, the offense sputtered again and again.

0 points
1
1
Coldworld's picture

October 22, 2020 at 11:38 pm

Offense was running well until Rodgers threw an interception and they went all out After him as their primary focus. Rodgers went ice cold.

-1 points
1
2
Lare's picture

October 23, 2020 at 05:40 am

NFL teams spend millions & millions of dollars every year to pay people to do nothing other than analyze film & data every single day looking for tendencies, trends and weaknesses. If any of us can do it for free looking at data that's available on the internet just think of what they're identifying.

For us fans, it's easy to just chalk something up to "he had a bad day" or "the scheme didn't work" or "it was the wrong play". It would be interesting to know how many of those happen due to data or film analysis.

2 points
3
1
Leatherhead's picture

October 23, 2020 at 09:47 am

I agree completely. These scouts and coaches and personnel people know so much more than us it’s ridiculous.

Take Wagner, for example. A team of professionals analyzed every single snap this guy took last year. They looked at his punch, and his footwork, and his ability to mirror. They looked at him in short yardage and pass protection. And THEN they decided to offer him a contract, which got a whole different bunch of professionals determining his market value.

And THEN, you have posters on a fan board saying he isn’t worth the money and we should have done something else. Sheeesh.

1 points
1
0
madtowndan's picture

October 23, 2020 at 08:59 am

Sorry, but I'm not buying any of it. Any picture can be painted with statistics. Sometimes common sense needs to overrule statistics, and common sense dictates that a team is better with one of the best receivers in the league than it is without him. Duh.

-1 points
0
1
malta1099's picture

October 23, 2020 at 11:12 am

Re-read the article. Your point is exactly what the author said. The problem is not DA being in the game, the problem is AR not spreading the ball around. Great article and great analysis!

0 points
0
0
TarynsEyes's picture

October 23, 2020 at 11:09 am

Statistics are like paint to an artist. They can make people see many things or just one thing in particular if used in a certain fashion, although, the easily convinced make the painters/statisticians easiest persuaded customer.

Rodgers favorite WR used to be the one that is open. We can clearly see that is no longer the case as both INTs were forced throws to a well cover Adams last week and that changed the dynamic for the offense for both Rodgers and MLF.

Adams possibly not being 100% can be used as excuse for the lack of separation part or it can be that TB did their job. I prefer the latter because percentages as to health are just pregame excuses for whatever bad comes during the game or it's end result. If you walk on the field to play, then you're expected to perform as though you're at normal game health. If you're not, then you're a hindrance to others and the team and that is on the players and HCs ignorance/ego.

Rodgers looked the rookie who just got smacked and wet himself and MLF got smacked by Rodgers because Rodgers made himself a rookie in a veteran needed offense.

I'm sure the offense will score on Houston and rants of cheers will be again echoing here with bluster, but if Watson plays mistake-free, unlikely, he will show us that our defense will allow shootouts and why GB will not beat a team whose defense doesn't and why a team with a top defense and above average QB will beat the elite QB with an average or less defense or the GB defense is a mirage because of weak offensive teams on the schedule, which is near 100% why we have failed to get to the SB since 2010. Remember, Minn was without Cook, Det was without Golliday, NO without Thomas and ATL was just crumbling to the ground,

A point to be aware of here is that Houston can make a defense look a 100 times better than they are and this could have GB, fans for sure thinking such, written all over it.

5 points
5
0
cheesehead1's picture

October 23, 2020 at 11:28 am

Tampa’s defense was lights out. Ours was soft and pathetic as usual. Can’t see our D improving greatly unless Gute pulls off a couple of trades.

0 points
0
0
badaxed's picture

October 25, 2020 at 10:54 am

The first trade should be for a new defensive coordinator.

0 points
0
0
OnWisconsinGoPack's picture

October 23, 2020 at 03:13 pm

It's not that our offense is worse with Adams, but defenses know they can key on him if Rodgers does not spread the targets. I think the Packers are still one good playmaker away even when Lazard comes back from being much more dynamic

1 points
1
0
Tundraboy's picture

October 23, 2020 at 09:37 pm

No but whenever Rodgers zones in only with Adams, the offense bogs down quickly and becomes unnecessarily predictable. Best only when they are way ahead. after they have spread it around beforehand.

1 points
1
0
Thegreatreynoldo's picture

October 23, 2020 at 09:39 pm

Spreading the ball around seems to be the answer. Easier to do if a team has a decent #2 WR. Lazard underwent surgery for a core muscle injury on Oct. 2 and was expected to miss at least a month. He was out there stretching on Oct. 21. He can return for week 9 by the rules. Usually I expect core muscle injures to linger a long time but IDK.

I hope he returns when he is fit and not before.

1 points
1
0
Taraman's picture

October 24, 2020 at 06:02 am

Interesting look on the statistics.
I also don't think it is as easy that the offense plays worse with Adams on the field. But the tendency to force the throw to him jumped on me while watching the game.
Defenses will focus on DA, making it harder to find him open. On the other hand this should lead to other targets getting open. if you then neglect DA too much, the special attention from the defenses will cease.
So in my opinion it is all a matter of balancing these effects over and over again.

And I'm pretty sure the team knows this all much better than we do.
But looking at it from ARs perspective, especially if you are under pressure, it is hard not to focus on your #1 WR and hope he gets open.

1 points
1
0
badaxed's picture

October 25, 2020 at 10:57 am

Football is a team sport. when it becomes a two-man game (Rogers to Adams), The results are predictable.

0 points
0
0
Taraman's picture

October 25, 2020 at 06:35 pm

Well, this week both of them showed us, that it can also work very well with a focus on one receiver...

0 points
0
0