2020 NFL Draft Discussion

[Jaegybomb]

Just not to that extreme.

[incognito_man]

Here's the source (after googling it): https://www.espn.com/nfl/story/_/id/28403463/barnwell-guide-2019-nfl-playoffs-how-all-12-teams-win-super-bowl

How they're better than you think: One trend that has stuck for the Packers throughout the season: They're a fundamentally different offense with Aaron Jones on the field. I referred to Green Bay's on/off splits on offense with Jones on the field earlier this year, so let's get an update.

The Packers have run 603 offensive snaps with him on the field. Those snaps have added 90.4 expected points to Green Bay's offensive output, or 0.15 expected points per play. Only two offenses in football generated more than 0.15 EPA per play this season -- the Ravens and Chiefs. With Jones on the field, the Packers are the third-most dominant offense in football.

Without Jones, they aren't the same. When Jamaal Williams has replaced Jones in the lineup, the Packers have generated minus-24.33 EPA across 337 snaps, or minus-0.07 EPA per play. The Williams version of the Green Bay offense ranks 29th in the NFL. Aaron Rodgers' QBR drops from 67.7 with Jones on the field to 28.7 with Williams in the lineup instead.

 

So, it's EPA, not DVOA, but the point remains (EPA won't change significantly when a team fails to convert a 3rd and 15, for example - the low chance of success is already baked into it).

Of those 337 snaps, Williams had 75 touches (22%) on 1st down, 49 touches (15%) on 2nd down, and 22 touches (7%) on 3rd and 4th downs.

Of Jones' 603 snaps, he had 164 touches (27%) on 1st down, 103 touches (17%) on 2nd down, and and 18 touches (3%) on 3rd and 4th downs.

The snap use differential doesn't appear large enough to explain the jump from 3rd to 29th between Jones and Williams, either. I think it's pretty evident that Jones was a valuable weapon for the GB offense.

[incognito_man]

9 minutes ago, AlexGreen#20 said:

Would it be normalized in that comparison? I'm not sure how it would be?

yes, because the result is always compared to identical situations for every other team. A high % down is a high % down for everyone. A low % down is a low % down for everyone. DVOA (and now EPA as I just realized it actually was) is just comparing the deltas of actual vs expected results.

[AlexGreen#20]

10 minutes ago, incognito_man said:

Here's the source (after googling it): https://www.espn.com/nfl/story/_/id/28403463/barnwell-guide-2019-nfl-playoffs-how-all-12-teams-win-super-bowl

How they're better than you think: One trend that has stuck for the Packers throughout the season: They're a fundamentally different offense with Aaron Jones on the field. I referred to Green Bay's on/off splits on offense with Jones on the field earlier this year, so let's get an update.

The Packers have run 603 offensive snaps with him on the field. Those snaps have added 90.4 expected points to Green Bay's offensive output, or 0.15 expected points per play. Only two offenses in football generated more than 0.15 EPA per play this season -- the Ravens and Chiefs. With Jones on the field, the Packers are the third-most dominant offense in football.

Without Jones, they aren't the same. When Jamaal Williams has replaced Jones in the lineup, the Packers have generated minus-24.33 EPA across 337 snaps, or minus-0.07 EPA per play. The Williams version of the Green Bay offense ranks 29th in the NFL. Aaron Rodgers' QBR drops from 67.7 with Jones on the field to 28.7 with Williams in the lineup instead.

 

So, it's EPA, not DVOA, but the point remains (EPA won't change significantly when a team fails to convert a 3rd and 15, for example - the low chance of success is already baked into it).

Of those 337 snaps, Williams had 75 touches (22%) on 1st down, 49 touches (15%) on 2nd down, and 22 touches (7%) on 3rd and 4th downs.

Of Jones' 603 snaps, he had 164 touches (27%) on 1st down, 103 touches (17%) on 2nd down, and and 18 touches (3%) on 3rd and 4th downs.

The snap use differential doesn't appear large enough to explain the jump from 3rd to 29th between Jones and Williams, either. I think it's pretty evident that Jones was a valuable weapon for the GB offense.

That's what I'm looking at here. And that's what the data is telling me. I'm just curious to see what the hell the difference would be. These are generally swung massively by big plays.

You may be right, but that difference seems unbelievably large to me. i'm wondering if it's redzone carries. 

[MrBobGray]

1 minute ago, incognito_man said:

Here's the source (after googling it): https://www.espn.com/nfl/story/_/id/28403463/barnwell-guide-2019-nfl-playoffs-how-all-12-teams-win-super-bowl

How they're better than you think: One trend that has stuck for the Packers throughout the season: They're a fundamentally different offense with Aaron Jones on the field. I referred to Green Bay's on/off splits on offense with Jones on the field earlier this year, so let's get an update.

The Packers have run 603 offensive snaps with him on the field. Those snaps have added 90.4 expected points to Green Bay's offensive output, or 0.15 expected points per play. Only two offenses in football generated more than 0.15 EPA per play this season -- the Ravens and Chiefs. With Jones on the field, the Packers are the third-most dominant offense in football.

Without Jones, they aren't the same. When Jamaal Williams has replaced Jones in the lineup, the Packers have generated minus-24.33 EPA across 337 snaps, or minus-0.07 EPA per play. The Williams version of the Green Bay offense ranks 29th in the NFL. Aaron Rodgers' QBR drops from 67.7 with Jones on the field to 28.7 with Williams in the lineup instead.

 

So, it's EPA, not DVOA, but the point remains (EPA won't change significantly when a team fails to convert a 3rd and 15, for example - the low chance of success is already baked into it).

Of those 337 snaps, Williams had 75 touches (22%) on 1st down, 49 touches (15%) on 2nd down, and 22 touches (7%) on 3rd and 4th downs.

Of Jones' 603 snaps, he had 164 touches (27%) on 1st down, 103 touches (17%) on 2nd down, and and 18 touches (3%) on 3rd and 4th downs.

The snap use differential doesn't appear large enough to explain the jump from 3rd to 29th between Jones and Williams, either. I think it's pretty evident that Jones was a valuable weapon for the GB offense.

But doesn't that actually back up his point?  If Williams is playing more often in 2nd and long, 3rd and long situations, then his overall increase in EPA would be expected to be smaller.  That's saying that when Williams is on the field, they aren't very effective, but if that was predominantly bad situations then that would actually be the expected result.

[AlexGreen#20]

1 minute ago, MrBobGray said:

But doesn't that actually back up his point?  If Williams is playing more often in 2nd and long, 3rd and long situations, then his overall increase in EPA would be expected to be smaller.  That's saying that when Williams is on the field, they aren't very effective, but if that was predominantly bad situations then that would actually be the expected result.

It looks like that's accounted for 2nd and 9 is viewed differently than 2nd and 2.

[MrBobGray]

Plus, the massive difference in QBR should also be something of a clue; Williams is largely used as a pass blocker, so he shouldn't be able to affect Rodgers QBR to that significant of a degree unless he was terrible in pass pro, which he is not.  If Rodgers QBR is that much lower when Williams is in, then it means that his QBR rating in more obvious passing situations is significantly lower than the team is threatening both the pass and run.  Which...again, makes sense given how he's been playing lately.

Edited April 8, 2020 by MrBobGray

[cannondale]

Anyone have the % of 3 and 4 WR sets by MLF vs MM ?  I suck at finding that stuff.

[Jaegybomb]

61% 2019

77% 2018

[cannondale]

5 minutes ago, Jaegybomb said:

61% 2019

77% 2018

Thanks. I assume that's the combined % of the 3 & 4 sets. Rumor has it that the Packers and Jones are talking contract. Those % tell me that MLF might value Jones a heck of a lot more than MM

[incognito_man]

22 minutes ago, MrBobGray said:

But doesn't that actually back up his point?  If Williams is playing more often in 2nd and long, 3rd and long situations, then his overall increase in EPA would be expected to be smaller.  That's saying that when Williams is on the field, they aren't very effective, but if that was predominantly bad situations then that would actually be the expected result.

I don't think that #1, their down and distance splits are wildly different and #2, I'm not sure which down and distances are "more likely" to have higher EPA anyway. Would guess 2nd and short might be better league average EPA than other situations, but, still the EPA is rolling. Converting a 3rd and long might be 10% chance but then it adds 10x EPA vs the EPA loss for NOT converting it.

It should all average out based on its underlying normalization.

[CWood21]

1 hour ago, cannondale said:

Yes Gary is an edge. I wanted a DL. Blacklock is a name that could be there. 6'-3" and 290. Not ideal on paper and some say he's lazy. I'm nowhere near smart enough to grade those guys

I feel like Blacklock is being slept on.  Not sure he's a world beater right now, but I think he's someone that 2-3 years down the road teams are going to kick themselves over.

[AlexGreen#20]

3 minutes ago, incognito_man said:

I don't think that #1, their down and distance splits are wildly different and #2, I'm not sure which down and distances are "more likely" to have higher EPA anyway. Would guess 2nd and short might be better league average EPA than other situations, but, still the EPA is rolling. Converting a 3rd and long might be 10% chance but then it adds 10x EPA vs the EPA loss for NOT converting it.

It should all average out based on its underlying normalization.

I'm just trying to figure this out. PFR has EPD for all plays they each touched the ball. In the passing game:

Jones has 68 targets. Williams had 45 targets.

Jones has 15 plays with an EPD of 1 or more. Williams has 14.

Jones has 17 plays with an EPD between 1 and 0. Williams has 13

Jones has 29 plays with an EDP between 0 and -1. Williams has 15.

Jones has 7 plays with an EDP worse than -1. Williams has 3.

How the hell is Williams so much worse in the passing game? Unless this is some unreal noise happening or there's something going on with the packages that's making this an issue? Williams did better as a receiver if you just believe this?

Jones had 236 rushes. Williams had 107 rushes.

Jones has 27 plays with an EDP of 1 or more. Williams has 11

Jones has 85 plays with an EDP between 1 and 0. Williams has 37.

Jones had 109 plays with an EDP between 0 and -1. Williams had 55.

Jones had 15 plays with an EDP worse than -1.Williams had 4.

This data isn't tracking???

 

 

 

 

[cannondale]

16 minutes ago, CWood21 said:

I feel like Blacklock is being slept on.  Not sure he's a world beater right now, but I think he's someone that 2-3 years down the road teams are going to kick themselves over.

He's listed as a DT, so I have no clue how he would be used in Pettine's system. Would he be DT only and only be on the field if Clark is not ? That's not ideal.

Edited April 8, 2020 by cannondale

[AlexGreen#20]

2 minutes ago, cannondale said:

He's listed as a DT, so I have no clue how he would be used in Pettine's system

He would replace Lancaster