Coming off a disappointing AFC Championship Game loss to Indianapolis, the 2007 New England Patriots overhauled their receiving corps with Randy Moss, Wes Welker and Donte' Stallworth. The result was the greatest offense the NFL has ever seen, a 16-0 regular-season record and a team that came within two minutes of a Super Bowl title.
A decade later, coming off a Super Bowl title, the Patriots have reloaded -- and this time, they did it everywhere. They made Stephon Gilmore one of the highest-paid cornerbacks in the game. They traded for one of the league's best young receivers, Brandin Cooks. They signed two highly efficient free-agent running backs, and they're getting Rob Gronkowski back from injury. The season hasn't even started, and there's already plenty of talk of the Patriots putting up another perfect season, and this time finishing the job in Super Bowl LII.
So if the 2007 Patriots played the 2017 Patriots, who would win?
To answer that question, we can compare our forecast for the 2017 Patriots to the actual performance of the 2007 Patriots. But which forecast? Projecting the upcoming NFL season is about a range of possibilities. If we compare the average expectation for the 2017 Patriots to the reality of the 2007 Patriots, the 2007 team will win every time. That's because the 2007 Patriots had almost everything go right; until the end of the Super Bowl, it was the best-case scenario of what that season could be.
So we've compared the 2007 and 2017 teams here, position group by position group. But we've also compared our DVOA ratings for the 2007 Patriots to our projections for the 2017 Patriots in two ways:
The average projection from the one million season simulations we did for Football Outsiders Almanac 2017.
The best-case scenario, the average of the top 10 percent of projections out of those one million simulations.
The results? At their absolute best, the 2017 Patriots would be better than the 2007 team -- but it's more likely that they will fall short of 2007's regular-season perfection.
Note: You will see references to DVOA (defense-adjusted value over average) throughout. For more on DVOA, go here.
2007 DVOA: 43.5 percent (first in 2007 and first all-time)
2017 average projection: 17.3 percent (second in NFL)
2017 best-case scenario: 32.8 percent (first in NFL)
It's hard to imagine a quarterback having a season better than Tom Brady did in 2007. His 88.2 Total QBR that year is the highest since ESPN's metric began in 2006. But you might be surprised by how close Brady now is to Brady then. He had an 83.0 Total QBR in 2016, which means he's coming off a better year than he was in 2007 (his Total QBR in 2006 was 71.5). The depth chart behind Brady also reduces the advantage for the 2007 Patriots: Jimmy Garoppolo, a second-round pick with two career NFL starts, is a much more reassuring backup option than Matt Cassel, a seventh-round pick who had appeared only in garbage time.
Only two running backs had rushing DVOA above 40 percent with at least 20 carries last season: Mike Gillislee and Rex Burkhead. The Patriots signed them both this offseason. By comparison, Laurence Maroney and Sammy Morris had both put up below-average DVOA ratings in 2006. Pass-catching back Kevin Faulk is a Patriots legend, but Super Bowl LI star James White is his equal. The 2007 Patriots also had no back as agile (or as injury-prone) as Dion Lewis.
Receivers, tight ends
Pick your poison: Moss may be the greatest game-breaking wide receiver in NFL history, and Gronkowski may be the greatest game-breaking tight end in NFL history. But if you pick Moss, there's a much better chance you can still use the poison in January. The 2017 Patriots also have Cooks, one of the top young receivers in the league; the 2007 Pats also had Benjamin Watson, one of the top young tight ends. Julian Edelman was supposed to play the Welker role, but now that will be Danny Amendola with Edelman out for the 2017 season. It's a step down for the 2017 squad because Edelman was so in sync with Brady on option routes.
Three of the starters in 2007 were chosen for the Pro Bowl: left tackle Matt Light, left guard Logan Mankins and center Dan Koppen. Light was first-team All-Pro. Not one offensive lineman on the Patriots' current roster has been selected for the Pro Bowl. The 2007 Patriots finished first in adjusted line yards and fourth in adjusted sack rate. The Patriots were ninth and sixth in those same stats a year ago.
2007 DVOA: -5.8 percent (11th in NFL)
2017 average projection: -6.0 percent (seventh in NFL)
2017 best-case scenario: -14.3 percent (first in NFL)
The 2007 Patriots played a 3-4 scheme, and all three starting linemen were first-round picks: Richard Seymour (2001), Ty Warren (2003) and Vince Wilfork (2004). The current interior linemen are rotated more. Alan Branch is a great run-stopper, but he'll be 33 years old this season. Malcom Brown, a first-round pick in 2015, isn't yet at the level shown by the fantastic trio of 2007 linemen. The 2007 team had an advantage on pass plays, but results were similar against the run. The 2006 Patriots were 10th in adjusted line yards, and dropped to 17th in 2007. The 2016 Patriots were 11th.
Another big advantage for 2007 over 2017. The Patriots had ranked 10th in adjusted sack rate in 2006, and improved that to second in 2007. They also ranked second in pressure rate (2007 was the first year Football Outsiders tracked pressure with game charting.) Last year, the Patriots ranked 27th in adjusted sack rate and 18th in pressure rate, and of the four players with at least four sacks last season, only Trey Flowers remains.
The top five linebackers for the 2007 Patriots were all 30 or over, including inside linebackers Tedy Bruschi (34) and Junior Seau (38). Aging linebackers were a big reason the Patriots ranked fifth in defensive DVOA through Week 9, but 23rd in Weeks 10-17. The 2017 Patriots have 27-year-old Dont'a Hightower coming off a Pro Bowl year. Shea McClellin and Kyle Van Noy are also in-their-prime, high draft picks who have succeeded in the Pats' scheme after disappointing with their original teams.
The Patriots went into 2007 with no cornerback depth behind star Asante Samuel and serviceable Ellis Hobbs. The 2017 Patriots signed Gilmore to form a star tandem with Malcolm Butler, and nickelbacks Eric Rowe and Justin Coleman each allowed less than 6 yards per target last year. Each team had a stud safety, but Devin McCourty is coming off a Pro Bowl season and is five years younger than Rodney Harrison was in 2007.
2007 DVOA: 3.6 percent (seventh in NFL)
2017 average projection: 4.1 percent (third in NFL)
2017 best-case scenario: 6.9 percent (second in NFL)
Stephen Gostkowski has been one of the NFL's best on kickoffs since he arrived in 2006. But while the 2007 Gostkowski was still a youngster with no record of kicking NFL field goals in the clutch, the 2017 version of Gostkowski had three outstanding years on field goals until a 2016 off year. The other strong part of special teams in 2007 was the kickoff returns from Hobbs. Today's Patriots aren't special on kickoff returns, but they have one of the NFL's best directional punters, Ryan Allen.
No coach in the NFL does more to adapt his scheme to his opponents' weaknesses, which is why the Patriots are famously malleable and multiple on both offense and defense. But a couple of Bill Belichick's tendencies remain consistent from 2007 to 2017: The Patriots rush only three players more than any other team, and they run their offense at a fast pace until they establish a big enough lead to run out the clock. One major change: Belichick, once among the most aggressive coaches in the NFL on fourth-and-short, has been closer to average since 2012.
2007 average DVOA of opponent: 2.0 percent (10th in NFL)
2017 average projected DVOA of opponent: -3.5 percent (32nd in NFL)
The secret weapon of the 2017 Patriots in their effort to repeat a 16-0 season: the schedule. By virtue of finishing first, the 2017 Patriots get to face a Houston team they crushed with their third-string quarterback in 2016, and a Pittsburgh team they've beaten repeatedly. By contrast, the 2007 Patriots had to play the Chargers, whom they barely beat in the 2006 divisional round, and the Colts, who had beaten them in the 2006 AFC Championship Game. The 2007 Patriots played the NFC East and AFC North, where all eight teams finished 19th or higher in DVOA. The 2017 Patriots will get the NFC South and AFC West, where most of the teams either combine a star quarterback with a shaky defense or have a defense but no quarterback.
The one similarity? In both 2007 and 2017, the rest of the AFC East was terrible.
Even a decade makes a big difference in the NFL. Quarterbacks today are more efficient with fewer turnovers, and teams score more points despite starting drives further from the end zone. Check out league averages in 2007 compared with how 2017 averages will look if current trends continue:
The percent of plays run from shotgun is a particularly interesting way to illustrate how the game has changed. The 2007 Patriots used shotgun on 50.4 percent of plays, the first team in NFL history to use shotgun more than half the time. The 2016 Patriots used shotgun on 53.5 percent of plays -- and that ranked 27th in the NFL last season!
Final tally, point spread
Based on the position-by-position breakdown, the 2007 Patriots are better at five positions, while the 2017 Patriots are better at four. The coaching is a push, and the schedule advantage goes to the 2017 edition.
In terms of a point spread, the 2007 Patriots would be favored over the average 2017 Patriots projection by 4.5 points. However, the best-case scenario 2017 Patriots would be favored over the 2007 Patriots by 0.5 points.