Page 3 of 3
Re: Let's Talk About Regression (and expectations)
Posted: 25 Aug 2020 08:16
by NCF
I understand regression to the mean, but let's consider this for just a moment. If we were a team that performed just average in close games last year and ended the season 8-8 or 9-7, don't you think this team would be a very trendy pick to rise up and exceed the previous year? We have talked about the OL and that is a legitimate concern, but we have also been around the horn a million times on WR and I think we feel better about that than the rest of the world. Then, on defense, again, I think we have good reason to feel a big jump is coming and no one outside of our little world is talking about that. This team is better, in my eyes, and even if we drop a couple close ones, I think this team's record is going to surprise people.
Re: Let's Talk About Regression (and expectations)
Posted: 25 Aug 2020 08:22
by go pak go
Yoop wrote: ↑25 Aug 2020 08:07
go pak go wrote: ↑14 Aug 2020 14:12
Yoop wrote: ↑14 Aug 2020 13:34
I was addressing the reasons we do well, very well at times in 1 quarter as opposed to another, and this goes all the way back to 1919
offenses set up defenses to gain mismatches, however depending on the defenses often the mismatches only last a few plays or a few series, then you have to use something else, it explains some of the lop sided scoring by quarters that BSA's post showed.
Yeah but the hole in this argument is you are lumping all offenses in this bucket saying that offense has an advantage over the defense to start. Which I won't contend. But the stat that BSA brought up means that we are SIGNIFICANTLY better than the rest of the league in Q1 and Q2 and SIGNIFICANTLY worse than the rest of the league in Q2 and Q4.
We are not comparing Packers scoring in Q1/Q3 vs Packers scoring Q2/Q4 as you are suggesting.
We are comparing Packers scoring in Q1/Q3 vs the league Q1/Q3 and Packers scoring Q2/Q4 vs the league.
last year being MLF's first reflects a smaller playbook, hense it's hard to create more mis matches, imo thats why you see lope sided scoring quarters, when you run out of plays to create them obviously defenses will adjust to stop them, which happened to us in the 2nd and 4th quarters, as I said the game has always been this way.
At least this is trying to explain your theory. So with our larger playbook in year two we should be able to have our "battle attacks" last a bit longer.
Re: Let's Talk About Regression (and expectations)
Posted: 25 Aug 2020 08:27
by YoHoChecko
NCF wrote: ↑25 Aug 2020 08:16
I understand regression to the mean, but let's consider this for just a moment. If we were a team that performed just average in close games last year and ended the season 8-8 or 9-7, don't you think this team would be a very trendy pick to rise up and exceed the previous year? We have talked about the OL and that is a legitimate concern, but we have also been around the horn a million times on WR and I think we feel better about that than the rest of the world. Then, on defense, again, I think we have good reason to feel a big jump is coming and no one outside of our little world is talking about that. This team is better, in my eyes, and even if we drop a couple close ones, I think this team's record is going to surprise people.
Yeah, and that's my annoyance.
Like, most analyses say that the 13-3 Packers performed closer to a 10-win team last year (some say 9). So even treading water gets us a 10-win season, and yet some people are taking the under on 9 wins? People are talking about missing the playoffs?
it's just weird. I feel strongly that everything is being painted through a lens of the Love pick. The analysts feel like that was an acknowledgement that they're building for some distant future and giving up on the present. And so they're giving up on the present.
I see it as a single move for the future, and an acknowledgement that we only have 2-4 years left with Rodgers. I see it (and the money we guaranteed to Rodgers last year) as a sign that we're going to try to maximize Rodgers' remaining time by surrounding him with defense and a strong running game to maximize his remaining abilities the same way the Broncos did with late-stage John Elway
Re: Let's Talk About Regression (and expectations)
Posted: 25 Aug 2020 08:37
by NCF
YoHoChecko wrote: ↑25 Aug 2020 08:27
I feel strongly that everything is being painted through a lens of the Love pick.
Yes!! That is a very succinct way of explaining exactly what is happening.
Re: Let's Talk About Regression (and expectations)
Posted: 25 Aug 2020 08:38
by go pak go
Honestly, I think this team is just about as talented (if not more talented) as 2014.
We attack in a completely different way, but this team has tremendous talent. I thought this team had tremendous talent last year too.
My biggest concern is Defensive Line #2, Tackle #2 and depth behind Inside Linebacker if Kirksey goes down. But man I don't see how you can't get excited about this roster.
Like just look at this defense alone.
I would find it reasonable to say we will see drops in production from the Smiths because they were so dominant last year.
But Alexander, Savage, Kirksey, Gary are all players we would expect too see play and production increase substantially. And then you hear how Kevin King so far is probably having the best camp of the corners in camp. Like we got a lot of turnovers last year....I don't see why that can't stay the same if not increase this year.
I mean my goodness. I am freaking pumped.
Re: Let's Talk About Regression (and expectations)
Posted: 25 Aug 2020 08:48
by YoHoChecko
go pak go wrote: ↑25 Aug 2020 08:38
My biggest concern is Defensive Line #2, Tackle #2 and depth behind Inside Linebacker if Kirksey goes down. But man I don't see how you can't get excited about this roster.
Like just look at this defense alone.
I would find it reasonable to say we will see drops in production from the Smiths because they were so dominant last year.
But Alexander, Savage, Kirksey, Gary are all players we would expect too see play and production increase substantially. And then you hear how Kevin King so far is probably having the best camp of the corners in camp. Like we got a lot of turnovers last year....I don't see why that can't stay the same if not increase this year.
I mean my goodness. I am freaking pumped.
Agreed
I actually texted a couple friends recently saying "look, maybe I'm just starved for football and overhyping myself based on training camp reports, but I genuinely think the Packers are a Marcel Dareus signing away from being one of the top 4 rosters in football"
Now, that still assumes health and talent doesn't necessarily win the games. But that's how I feel.
Re: Let's Talk About Regression (and expectations)
Posted: 25 Aug 2020 09:44
by go pak go
YoHoChecko wrote: ↑25 Aug 2020 08:48
go pak go wrote: ↑25 Aug 2020 08:38
My biggest concern is Defensive Line #2, Tackle #2 and depth behind Inside Linebacker if Kirksey goes down. But man I don't see how you can't get excited about this roster.
Like just look at this defense alone.
I would find it reasonable to say we will see drops in production from the Smiths because they were so dominant last year.
But Alexander, Savage, Kirksey, Gary are all players we would expect too see play and production increase substantially. And then you hear how Kevin King so far is probably having the best camp of the corners in camp. Like we got a lot of turnovers last year....I don't see why that can't stay the same if not increase this year.
I mean my goodness. I am freaking pumped.
Agreed
I actually texted a couple friends recently saying "look, maybe I'm just starved for football and overhyping myself based on training camp reports, but I genuinely think the Packers are a Marcel Dareus signing away from being one of the top 4 rosters in football"
Now, that still assumes health and talent doesn't necessarily win the games. But that's how I feel.
I remember how much people gushed over that 2018 Bears defense.
I look at the Packers 2020 defense and I'm like, "we have that." Like have exactly the same fearing defense the Bears had in 2018 except our D line is less deep (though just has good if not better at the top). We have way more depth at OLB but same top end talent. Slightly less talent at ILB. Much less depth at ILB. More talent and depth at Corner.
And then you're comparing Savage to Eddie Jackson. And Amos to.....yeah.
Like we are there. We have it. The window is now.
Why I don't say it much is because we had those same ingredients last year. I was convinced last September we had a top 5 defense. And it didn't happen.
I don't know why. But it didn't. And for that, I don't say too much too much how freaking excited I am for this football team.
Re: Let's Talk About Regression (and expectations)
Posted: 25 Aug 2020 21:01
by BSA
go pak go wrote: ↑25 Aug 2020 09:44
I was convinced last September we had a top 5 defense. And it didn't happen.
I don't know why. But it didn't.
there's always a multitude of reasons, but one contributor was how poor the offense was on 3rd down.
GB kept putting that defense out there and asking them to save the day. GB offense was 9th most punts/ game and 19th in 3rd down conversion rate
And GB defense was just as crazy as the offense in points allowed by quarter
Q1 = 6th
Q2= 20th
Q3= 15th
Q4 = 5th
https://www.teamrankings.com/nfl/team-stats/
.
Re: Let's Talk About Regression (and expectations)
Posted: 26 Aug 2020 07:59
by BF004
Will be interested to see how we do starting out on D.
We don’t play what I would consider a below average offense until week 10 against Jacksonville.
Cousins, Stafford, Brees, Ryan, Brady, Watson, Cousins, Garrapollo
All QBs have been to pro bowls, 3 best WRs (Evans, Thomas, Jones) in football (would be 4 if Hopkins still in Houston).
Then it does get easier down the stretch, Minshew, Rivers, Biscky/Foles, Wentz, Stafford, Bridgewater, Tannehill, Biscky/Foles
If we start hot on D, pass rush and corners keeping us top 10, I think we would have a very good shot of ending the year top 5.
I want top 3 in both scoring and yards.
And I do agree, better offense, time control, field possession, only gunna make the job for the D all the easier.
Re: Let's Talk About Regression (and expectations)
Posted: 26 Aug 2020 10:16
by YoHoChecko
I read through The Athletic's "A Football Nerd's Guide to the Green Bay Packers" the other night. It's old and I don't know if it got any play here (Funchess is still discussed, for instance), but it has some relevant aspects to this chat.
First, it's by the same person who made the continuity scores, which BSA already brought to our attention here, showing that Green Bay has the second highest continuity score in the league, which could/should translate into improvement, especially relative improvement in a year such as this with an abbreviated offseason.
But second, that maybe some of our good luck is overstated; the close games mark has been discussed a great deal, as well as the high number of interceptions. But based on Football Outsider's "adjusted games lost" metric of health, we were only middle of the pack, rather than what I would have expected (losing a starting OL for 14 games probably boosts that, even though it turned into an upgrade for us). And we were 28th in fumble luck--the theory there being that fumble recoveries, generally speaking, are fairly random events that aren't about your skill at scooping up footballs but more about the way the ball happens to fall and bounce. A regression toward the mean in a positive direction on fumbles could offset a regression toward the mean in a negative direction with interceptions.
Re: Let's Talk About Regression (and expectations)
Posted: 26 Aug 2020 10:34
by go pak go
YoHoChecko wrote: ↑26 Aug 2020 10:16
I read through The Athletic's "A Football Nerd's Guide to the Green Bay Packers" the other night. It's old and I don't know if it got any play here (Funchess is still discussed, for instance), but it has some relevant aspects to this chat.
First, it's by the same person who made the continuity scores, which BSA already brought to our attention here, showing that Green Bay has the second highest continuity score in the league, which could/should translate into improvement, especially relative improvement in a year such as this with an abbreviated offseason.
But second, that maybe some of our good luck is overstated; the close games mark has been discussed a great deal, as well as the high number of interceptions. But based on Football Outsider's "adjusted games lost" metric of health, we were only middle of the pack, rather than what I would have expected (losing a starting OL for 14 games probably boosts that, even though it turned into an upgrade for us). And we were
28th in fumble luck--the theory there being that fumble recoveries, generally speaking, are fairly random events that aren't about your skill at scooping up footballs but more about the way the ball happens to fall and bounce. A regression toward the mean in a positive direction on fumbles could offset a regression toward the mean in a negative direction with interceptions.
So is the "fumble luck" basically a % of fumble recoveries against forced fumbles? So it was saying we recovered less of our forced fumbles than the rest of the league?
Re: Let's Talk About Regression (and expectations)
Posted: 26 Aug 2020 10:53
by YoHoChecko
go pak go wrote: ↑26 Aug 2020 10:34
So is the "fumble luck" basically a % of fumble recoveries against forced fumbles? So it was saying we recovered less of our forced fumbles than the rest of the league?
yes, precisely; though i think it's % of recoveries of fumbles on offense and defense, not just our own forced fumbles
Re: Let's Talk About Regression (and expectations)
Posted: 28 Aug 2020 14:53
by BSA
this snippet comes from Michael Lombardi at the Athletic talking up the Lions
...Operating for the first time under Darrell Bevell’s West Coast offense, the Lions averaged 25.5 points in the eight games that Stafford played. They were 3-4-1, but three of those losses were by 7 points or fewer."
It really shines a light on how effing stupid all of this commentary is about close wins/losses and regression. If you want to prop up a team, you mention their close losses. If your angle is to tear down a team, you talk crap about their close wins. Its all a bunch of internet masturbation
Dear Media: I'll take the team that found a way to win close games and you can have the team that found a way to lose close games.
We'll meet back at the end of the 2020 season and compare notes. Good luck
Re: Let's Talk About Regression (and expectations)
Posted: 28 Aug 2020 16:05
by BF004
BSA wrote: ↑28 Aug 2020 14:53
this snippet comes from Michael Lombardi at the Athletic talking up the Lions
...Operating for the first time under Darrell Bevell’s West Coast offense, the Lions averaged 25.5 points in the eight games that Stafford played. They were 3-4-1, but three of those losses were by 7 points or fewer."
It really shines a light on how effing stupid all of this commentary is about close wins/losses and regression. If you want to prop up a team, you mention their close losses. If your angle is to tear down a team, you talk crap about their close wins. Its all a bunch of internet masturbation
Dear Media: I'll take the team that found a way to win close games and you can have the team that found a way to lose close games.
We'll meet back at the end of the 2020 season and compare notes. Good luck
So even if you take away those 3 losses, they still only won 3 out of the remaining 5 games. And I'll go out on a limb and say they didn't win all 3 of those games by 8 or more.