Separating the jobs and giving GM final say over personnel has led to sustained success
The crucial first step of the Green Bay Packers' resurrection in the early 1990s was former team president Bob Harlan's decision to give his general manager final say over all football operations.
Harlan's franchise-defining move was a radical change from the way the Packers had operated since the late 1960s. And as general manager Ted Thompson prepares to select his next draft class for Mike McCarthy to coach in less than two weeks, it's worth remembering how well the Packers' current front-office model has served them.
Now, it bears pointing out that there's no one way to run a successful NFL team. The GM-over-coach system has its potential pitfalls, and a dual coach-GM can be wildly successful. Bill Belichick, anyone?
But the Packers' performance the past 45 or so years argues strongly that they should continue with this structure well into the future, beyond Thompson and McCarthy, no matter how strong the temptation to change at some point.
During the Packers' disastrous years between Vince Lombardi's departure after the 1968 season until the hiring of Ron Wolf as head of football operations near the end of the 1991 season, the Packers had the fourth-worst winning percentage (.421) in the NFL. The only teams worse were the three most-recent expansion franchises of that era: Atlanta (.409), New Orleans (.395) and Tampa Bay (.293).
Phil Bengtson, Dan Devine, Bart Starr and Forrest Gregg all bombed in the dual coach-GM role. The Tom Braatz-Lindy Infante 50-50 power-sharing arrangement from 1988-91 was untenable.
But since '92, the Packers have been the league's second-most-winning franchise (.637) behind only New England (.641). During that time, they've had their current structure for all but three full years, when Mike Sherman was the dual coach-GM from 2002-04. (Wolf retired in '01, but only after conducting the '01 draft).
Sherman's record actually was pretty good in those three seasons (32-16) with Brett Favre as his quarterback, but he was 1-3 in the playoffs. More importantly, Bob Harlan, the team's president at the time, became convinced that doing both jobs detracted from Sherman's coaching. Cornerback Mike McKenzie's contract holdout early in the '04 season was Sherman's breaking point.
"Mike's personality changed," Harlan said this week. "The pressure got to him and he became a different person."
In '05, Harlan separated the GM and coaching jobs and in the 10 seasons since, the Packers have won a Super Bowl and played in two other NFC championship games. The front-office structure isn't the only factor behind the Packers' success the past 23 years. But it's the foundation.
For instance, would a coach-GM have had Ted Thompson's long-haul mind-set to draft quarterback Aaron Rodgers in the first round in 2005 even though the Packers were in the running for Super Bowls at the time with Favre at quarterback? Probably not.
A look around the league shows that all but a handful of teams basically share the Packers' model. There's ambiguity and idiosyncrasy in the division of power with some franchises, but we can say that in a general sense 26 teams have a GM who has final say over personnel. That leaves six who give that power to their head coach, either by title or in practice: New England (Belichick), Philadelphia (Chip Kelly), Seattle (Pete Carroll), New Orleans (Sean Payton), St. Louis (Jeff Fisher) and Tampa Bay (Lovie Smith).
The ambiguity is in places such as Seattle, where GM John Schneider runs the draft and is the engine in all personnel moves, but Carroll retains final say, even if it's a rubber stamp. That system has worked beautifully for the Seahawks because Carroll and Schneider by all accounts have an exceptional relationship and similar mind-sets for building a team — both are aggressive in acquiring players and have a high tolerance for risk.
Belichick's success — four Super Bowl wins and two other Super Bowl appearances since the 2001 season — also argues for the dual-role model. But I'd counter that he's an outlier over the past 25 or so years in sustaining success. And if Belichick the GM didn't have Belichick the coach, the bet here is he'd be no more successful than anyone else. His cut-above coaching makes it worth giving him both jobs, if that's what it takes to get him.
And look at the coaches who have failed in the dual role. Mike Shanahan looked like a genius while winning back-to-back Super Bowls in the 1997 and '98 seasons with Denver. But after John Elway retired, Shanahan was only 115-109, never advanced to a conference championship and was fired from two jobs.
Or Bill Parcells, the man who famously said that if he was going to cook the dinner, he wanted to buy the groceries. He won two Super Bowls when GM George Young bought the groceries with the New York Giants. But in Parcells' 11 seasons as coach-GM with New England, the New York Jets and Dallas, he never won a Super Bowl and played in only one.
Doing both jobs is just too much for one person and eliminates a valuable, if potentially acrimonious, check and balance within an organization.
Former Packers coach Mike Holmgren disagrees — he held the dual role in Seattle from 1999 through 2002. In an interview this week, he said that if the coach-GM tries to do it all, it is a mistake.
"But I think it's possible," he said. "What Jimmy Johnson did in Dallas, the best example. What he did, he allowed his coordinators to kind of coach the football team. Jimmy's emphasis was on personnel. I was the other way (in Seattle from '99-02). I wanted to stay involved with the football coaching part of it and then kick the personnel heavy lifting to the personnel guys."
Essentially Holmgren is describing Seattle's setup with Carroll and Schneider. And it's true that giving the coach final say on personnel avoids the greatest weakness of the GM-over-coach model, which is the potential for friction between the two.
"(Combining the job) defines the final decision," Holmgren said. "I think teams get in trouble — you see it every year, spitting contests, 'Listen, I got you the best players, you just aren't coaching them.' And then the coach says, 'Wait a second, you didn't. I'm coaching my butt off, you didn't give me the best players.' That happens all over. So you eliminate that thing by combining them."
But it's hard not think that the temptation for a coach-GM to try to do more and more is too great to resist. Remember, these are control freaks by nature.
Better to have a GM who hires the coach and lives with the consequences. That hardly guarantees success. But it's the best first step toward it.
"If you don't have a good general manager, I don't think there's any way you can succeed," Harlan said. "I'm convinced of that. The system we've had has worked so well."
— firstname.lastname@example.org and follow him on Twitter @PeteDougherty.