The Costello thread and some quick research I did to show how disappointing our 2018 offense was got me to thinking about the broader sweep of evidence over time suggesting the extremely strong correlation between great QB play and great offenses. The way I thought I'd do this is look at all the years in which both Total QBR and FEI stats are available and indicate the FEI rankings for every offense that's had a top 10 QB in that span:
#1 2007 Florida
#1 2008 Oklahoma
#1 2009 Cincinnati
#1 2010 Stanford
#1 2011 Wisconsin
#1 2012 Texas A&M
#1 2013 Florida State
#1 2014 Oregon
#1 2015 Western Kentucky
#1 2016 Oklahoma
#1 2017 Oklahoma
#1 2018 Oklahoma
#1 2019 LSU
#2 2008 Texas
#2 2009 Houston
#2 2010 Boise State
#2 2011 Baylor
#2 2012 Baylor
#2 2013 Texas A&M
#2 2014 Ohio State
#2 2017 Oklahoma State
#2 2018 Alabama
#2 2019 Alabama
#3 2007 Texas Tech
#3 2010 Auburn
#3 2011 Stanford
#3 2012 Oregon
#3 2014 Georgia Tech
#3 2015 Arkansas
#3 2016 Western Michigan
#3 2018 Georgia
#3 2019 Ohio State
#4 2007 Missouri
#4 2008 Florida
#4 2009 Stanford
#4 2010 Nevada
#4 2011 Houston
#4 2013 Oregon
#4 2014 Alabama
#4 2015 Stanford
#4 2017 Ohio State
#4 2019 Oklahoma
#5 2007 West Virginia
#5 2009 Nevada
#5 2012 Alabama
#5 2013 Baylor
#5 2014 Auburn
#5 2016 Texas Tech
#5 2017 Penn State
#5 2018 Ohio State
#5 2019 Clemson
#6 2007 Oklahoma
#6 2008 Rice
#6 2009 BYU
#6 2010 Wisconsin
#6 2011 Boise State
#6 2012 Clemson
#6 2014 Georgia
#6 2015 Baylor
#6 2016 Clemson
#6 2017 Georgia
#6 2019 Wisconsin
#7 2007 Oklahoma State
#7 2008 Oklahoma State
#7 2009 Florida
#7 2011 Oklahoma State
#7 2012 Georgia
#7 2013 LSU
#7 2017 Central Florida
#7 2018 Clemson
#7 2019 Minnesota
#8 2008 Penn State
#8 2010 Michigan
#8 2011 USC
#8 2012 Oklahoma
#8 2016 Washington
#9 2008 Missouri
#9 2010 TCU
#9 2012 West Virginia
#9 2013 South Carolina
#9 2017 Louisville
#10 2008 USC
#10 2010 Ohio State
#10 2011 Alabama
#10 2012 Kansas State
#10 2014 Michigan State
#10 2015 Ole Miss
#10 2017 Arizona
#10 2019 Navy
#11 2007 Louisville
#11 2010 Virginia Tech
#11 2015 Oregon
#11 2018 Missouri
#12 2007 Oregon
#12 2010 Northern Illinois
#12 2013 Georgia
#12 2014 Texas A&M
#12 2015 Cal
#12 2016 South Florida
#12 2017 Alabama
#13 2009 Pitt
#13 2015 Clemson
#13 2018 West Virginia
#14 2008 Georgia
#14 2016 Louisville
#15 2009 Virginia Tech
#15 2014 USC
#16 2008 Boise State
#17 2016 USC
#17 2018 Michigan
#18 2011 Northwestern
#18 2013 UCLA
#19 2019 Florida
#20 2012 Michigan
#21 2007 Tennessee
#23 2016 Ole Miss
#23 2019 Louisville
#25 2015 Michigan
#26 2017 LSU
#26 2018 Houston
#27 2007 Fresno State
#27 2015 TCU
#29 2009 NC State
#35 2009 Michigan State
#36 2013 Oklahoma State
#36 2014 Boston College
#39 2018 Stanford
#40 2016 North Carolina
#41 2013 Mississippi State
#45 2011 South Carolina
Remarkable observations flow from this:
* When a team has a great QB, the median season is that they have the #7 offense in the country. Typically when Stanford has had great QB play we've also had a great offense, but the latest data point is super concerning. Only Mitch Trubisky, Dak Prescott, and Connor Shaw have been squandered like KJ Costello was in 2018 (and South Carolina did better by Shaw in 2013).
* If you have a top 10 QB, there is a nearly 40 percent chance (39 percent) that you'll have a top 5 offense.
* There is only a 9 percent chance a team with a top 10 QB will have an offense outside the top 25, only a 12 percent chance of being outside the top 20, only an 18 percent chance of being outside the top 15, and a 32 percent chance of being outside the top 10.
* At least in the last 13 years, the best offense in the country is always a team with a great QB.
* Among the 91 teams in the last 13 years that have had a top 7 offense, 71 of them had a top 10 QB. But even that significantly undersells the correlation. Most of the exceptions were Group of Five teams. Some of the other exceptions only missed out because of my arbitrary cutoff at top 10 but tell the exact same story (2016 Pitt having #11 QB Nate Peterman, 2015 Notre Dame having #12 DeShone Kizer, or 2013 Auburn having Nick Marshall for example). It's somewhat debatable whether it's even possible to have a great offense without a great QB. One of the best examples I can think of is 2018 Washington State (Gardner Minshew #22 in QBR), but since Minshew is an NFL QB that seems like a weak refutation of this correlation. The single weakest Power Five QB I can think of who manned a dominant offense was 2010 Oregon's Darron Thomas. By and large, the correlation is just so loud it jumps off the page.
Matt's opinion: While Davis Mills has not yet proven himself to be in the echelon of QBs I am talking about in this post, he is not far off. He was the #16 QB in the nation in 2019 by this metric. My bet is he is in the echelon we're talking about here. Stanford's 2020 roster, with or without Little, should be a top 10 offense in the nation. I do not think it will be. Having only Dalman and Sarell as fourth year linemen is a red alert blaring warning sign given Shaw's history. The Rubik's Cube requires an alignment of stars such that elite QB play also has an experienced OL group. We won't have that in 2020. A majority of the line being second year guys is still very young and push-around-able. As a result, I think a top 20 offense is the reasonable target and a top 30 offense is the safe bet. I wish it weren't so but offensive underachievement seems like part of the deal.
#1 2007 Florida
#1 2008 Oklahoma
#1 2009 Cincinnati
#1 2010 Stanford
#1 2011 Wisconsin
#1 2012 Texas A&M
#1 2013 Florida State
#1 2014 Oregon
#1 2015 Western Kentucky
#1 2016 Oklahoma
#1 2017 Oklahoma
#1 2018 Oklahoma
#1 2019 LSU
#2 2008 Texas
#2 2009 Houston
#2 2010 Boise State
#2 2011 Baylor
#2 2012 Baylor
#2 2013 Texas A&M
#2 2014 Ohio State
#2 2017 Oklahoma State
#2 2018 Alabama
#2 2019 Alabama
#3 2007 Texas Tech
#3 2010 Auburn
#3 2011 Stanford
#3 2012 Oregon
#3 2014 Georgia Tech
#3 2015 Arkansas
#3 2016 Western Michigan
#3 2018 Georgia
#3 2019 Ohio State
#4 2007 Missouri
#4 2008 Florida
#4 2009 Stanford
#4 2010 Nevada
#4 2011 Houston
#4 2013 Oregon
#4 2014 Alabama
#4 2015 Stanford
#4 2017 Ohio State
#4 2019 Oklahoma
#5 2007 West Virginia
#5 2009 Nevada
#5 2012 Alabama
#5 2013 Baylor
#5 2014 Auburn
#5 2016 Texas Tech
#5 2017 Penn State
#5 2018 Ohio State
#5 2019 Clemson
#6 2007 Oklahoma
#6 2008 Rice
#6 2009 BYU
#6 2010 Wisconsin
#6 2011 Boise State
#6 2012 Clemson
#6 2014 Georgia
#6 2015 Baylor
#6 2016 Clemson
#6 2017 Georgia
#6 2019 Wisconsin
#7 2007 Oklahoma State
#7 2008 Oklahoma State
#7 2009 Florida
#7 2011 Oklahoma State
#7 2012 Georgia
#7 2013 LSU
#7 2017 Central Florida
#7 2018 Clemson
#7 2019 Minnesota
#8 2008 Penn State
#8 2010 Michigan
#8 2011 USC
#8 2012 Oklahoma
#8 2016 Washington
#9 2008 Missouri
#9 2010 TCU
#9 2012 West Virginia
#9 2013 South Carolina
#9 2017 Louisville
#10 2008 USC
#10 2010 Ohio State
#10 2011 Alabama
#10 2012 Kansas State
#10 2014 Michigan State
#10 2015 Ole Miss
#10 2017 Arizona
#10 2019 Navy
#11 2007 Louisville
#11 2010 Virginia Tech
#11 2015 Oregon
#11 2018 Missouri
#12 2007 Oregon
#12 2010 Northern Illinois
#12 2013 Georgia
#12 2014 Texas A&M
#12 2015 Cal
#12 2016 South Florida
#12 2017 Alabama
#13 2009 Pitt
#13 2015 Clemson
#13 2018 West Virginia
#14 2008 Georgia
#14 2016 Louisville
#15 2009 Virginia Tech
#15 2014 USC
#16 2008 Boise State
#17 2016 USC
#17 2018 Michigan
#18 2011 Northwestern
#18 2013 UCLA
#19 2019 Florida
#20 2012 Michigan
#21 2007 Tennessee
#23 2016 Ole Miss
#23 2019 Louisville
#25 2015 Michigan
#26 2017 LSU
#26 2018 Houston
#27 2007 Fresno State
#27 2015 TCU
#29 2009 NC State
#35 2009 Michigan State
#36 2013 Oklahoma State
#36 2014 Boston College
#39 2018 Stanford
#40 2016 North Carolina
#41 2013 Mississippi State
#45 2011 South Carolina
Remarkable observations flow from this:
* When a team has a great QB, the median season is that they have the #7 offense in the country. Typically when Stanford has had great QB play we've also had a great offense, but the latest data point is super concerning. Only Mitch Trubisky, Dak Prescott, and Connor Shaw have been squandered like KJ Costello was in 2018 (and South Carolina did better by Shaw in 2013).
* If you have a top 10 QB, there is a nearly 40 percent chance (39 percent) that you'll have a top 5 offense.
* There is only a 9 percent chance a team with a top 10 QB will have an offense outside the top 25, only a 12 percent chance of being outside the top 20, only an 18 percent chance of being outside the top 15, and a 32 percent chance of being outside the top 10.
* At least in the last 13 years, the best offense in the country is always a team with a great QB.
* Among the 91 teams in the last 13 years that have had a top 7 offense, 71 of them had a top 10 QB. But even that significantly undersells the correlation. Most of the exceptions were Group of Five teams. Some of the other exceptions only missed out because of my arbitrary cutoff at top 10 but tell the exact same story (2016 Pitt having #11 QB Nate Peterman, 2015 Notre Dame having #12 DeShone Kizer, or 2013 Auburn having Nick Marshall for example). It's somewhat debatable whether it's even possible to have a great offense without a great QB. One of the best examples I can think of is 2018 Washington State (Gardner Minshew #22 in QBR), but since Minshew is an NFL QB that seems like a weak refutation of this correlation. The single weakest Power Five QB I can think of who manned a dominant offense was 2010 Oregon's Darron Thomas. By and large, the correlation is just so loud it jumps off the page.
Matt's opinion: While Davis Mills has not yet proven himself to be in the echelon of QBs I am talking about in this post, he is not far off. He was the #16 QB in the nation in 2019 by this metric. My bet is he is in the echelon we're talking about here. Stanford's 2020 roster, with or without Little, should be a top 10 offense in the nation. I do not think it will be. Having only Dalman and Sarell as fourth year linemen is a red alert blaring warning sign given Shaw's history. The Rubik's Cube requires an alignment of stars such that elite QB play also has an experienced OL group. We won't have that in 2020. A majority of the line being second year guys is still very young and push-around-able. As a result, I think a top 20 offense is the reasonable target and a top 30 offense is the safe bet. I wish it weren't so but offensive underachievement seems like part of the deal.
