Seemingly everyone has a Super Bowl take or prediction these days — the football media is afforded 14 days to scrutinize two teams and dissect every facet of their matchup. But that doesn’t mean every useful bit of information has been presented on the New England Patriots and the Philadelphia Eagles. We dug into the databases of ESPN’s Stats & Information Group and Football Outsiders to find a few obscure statistical areas to highlight that could help predict Sunday’s showdown.
Stat No. 1: Weighted DVOA
The key efficiency metric used at Football Outsiders is Defense-adjusted Value Over Average, or DVOA for short (explained here). For the season, the Eagles ranked fifth in DVOA (23.5 percent) and actually edged out the sixth-ranked Patriots (22.6 percent). However, when looking at weighted DVOA, which places less importance on early-season games — the ones when the Patriots couldn’t get off the field and Carson Wentz was launching his breakout campaign — things become much different. The Patriots finished No. 1 in weighted DVOA (33.6 percent) and were 10 percentage points higher than the No. 7 Eagles (23.6 percent).
It makes sense for the Patriots to rank No. 1 in weighted DVOA. Four of their five lowest games by DVOA came in the first five weeks of the season. Their only loss in their last 14 outings was the Week 14 game in Miami. It also makes sense that the Eagles held steady between the two stats: Philadelphia had five games with negative DVOA in 2017, and four of them were the last four games of the regular season.
It has taken some time to adjust from Wentz to Nick Foles at quarterback, but Foles and the Eagles were stellar in the NFC Championship Game against Minnesota. That game is not reflected in the 23.6 percent weighted DVOA for the Eagles, but as Neil Paine recently wrote, a dominant performance in the conference title round has not been predictive of success in the Super Bowl.
Since 1986 (as far back as the data currently goes), teams with at least a 5 percentage point edge in weighted DVOA are 15-8 in the Super Bowl. That sounds good for New England, but consider that the two biggest upsets by weighted DVOA were the Patriots’ losses to the Giants in Super Bowls XLII (the Pats had a 39.9-point advantage) and XLVI (a 19.3-point edge).
The biggest mismatches heading into a Super Bowl
Super Bowls with the biggest discrepancy in weighted DVOA since 1986
DVOA | DVOA | ||||||
---|---|---|---|---|---|---|---|
Year | Winner | WEI | Rank | Loser | WEI | Rank | DIFF. |
2008 | PIT | 30.1% | 2 | ARI | -11.5% | 22 | +41.7 |
2007 | NYG | 2.6 | 17 | NE | 42.5 | 1 | -39.9 |
1994 | SF | 39.5 | 1 | SD | 7.7 | 8 | +31.8 |
1991 | WAS | 52.4 | 1 | BUF | 21.1 | 3 | +31.3 |
1993 | DAL | 28.0 | 2 | BUF | -2.9 | 19 | +30.9 |
2003 | NE | 31.2 | 3 | CAR | 1.2 | 16 | +30.0 |
1992 | DAL | 40.6 | 1 | BUF | 10.6 | 9 | +30.0 |
1989 | SF | 42.9 | 1 | DEN | 18.8 | 4 | +24.1 |
1996 | GB | 34.3 | 1 | NE | 13.4 | 9 | +20.9 |
2011 | NYG | 6.0 | 14 | NE | 25.3 | 3 | -19.3 |
2013 | SEA | 43.7 | 1 | DEN | 27.0 | 2 | +16.7 |
2012 | BAL | 8.3 | 11 | SF | 23.8 | 5 | -15.5 |
2015 | DEN | 16.3 | 6 | CAR | 30.9 | 3 | -14.6 |
2010 | GB | 25.1 | 4 | PIT | 38.4 | 2 | -13.3 |
1997 | DEN | 23.4 | 6 | GB | 36.6 | 1 | -13.2 |
2016 | NE | 33.0 | 1 | ATL | 19.9 | 4 | +13.1 |
2004 | NE | 32.8 | 3 | PHI | 20.4 | 7 | +12.5 |
1986 | NYG | 23.8 | 4 | DEN | 11.3 | 10 | +12.5 |
1987 | WAS | 5.3 | 11 | DEN | 17.5 | 4 | -12.2 |
2000 | BAL | 23.3 | 5 | NYG | 14.1 | 10 | +9.2 |
1999 | STL | 28.2 | 1 | TEN | 19.4 | 5 | +8.8 |
2001 | NE | 15.4 | 9 | STL | 21.9 | 4 | -6.5 |
1995 | DAL | 30.8 | 3 | PIT | 24.7 | 5 | +6.1 |
Stat No. 2: Defensive DVOA
For the season, the Patriots rank 31st in defensive DVOA, and that could ultimately be their undoing this week. Our study of Super Bowl winners has shown that no team since at least 1986 has won a Super Bowl with a defense ranked lower than 25th in this metric. Atlanta came really close a year ago, but the 26th-ranked defense eventually wore down against the Patriots in the Falcons’ 28-3 collapse in Super Bowl LI. The 2011 Patriots also came close with the 30th-ranked defense, but Eli Manning led the Giants down the field for another game-winning drive in Super Bowl XLVI. Teams don’t require great balance to win a Super Bowl, but having the No. 1 offense and No. 31 defense like the 2017 Patriots makes them the most unbalanced Super Bowl team in more than 30 years.
The main reason for the Patriots’ mediocre start to the season was the defense. New England actually allowed more than 400 yards of offense in each of the first six games but has done that just once in the past 12 games, including the playoffs, and has conceded more than 20 points just twice.
The improvement in defensive coordinator Matt Patricia’s unit has been real, but the Patriots still rank only 22nd in weighted defensive DVOA because they continue to allow a lot of successful plays. The 2017 Patriots have one of the most statistically unique defenses that we have ever studied. They are 32nd in yards per drive allowed but still rank sixth in points per drive allowed. No other defense since the latest league expansion in 2002 has ranked in the top 10 in points per drive allowed while ranking 26th or lower in yards per drive allowed. The Patriots were able to accomplish this with an extreme bend-but-don’t-break style of play. While part of this reflects New England’s red-zone defense that ranks second in points allowed per red-zone appearance, it’s also a matter of real estate. The Patriots’ opponents have the worst starting field position in the league thanks to New England’s strong special teams and ball security on offense — so there are more yards for offenses to gain against this unit. Still, the defense has managed to keep scoring down and is even doing this with just one takeaway in the past six games.
You also can say that the Patriots have had some good fortune on defense this year. Opposing kickers missed nine field goals against the Patriots for a success rate of 71.0 percent, the second lowest against any team in 2017. New England has also controversially benefited from a few touchdowns overturned by replay — against the Jets (Austin Seferian-Jenkins), Steelers (Jesse James) and Bills (Kelvin Benjamin).
Lest we forget, the Eagles finished 2017 ranked fourth in offensive points per drive. Foles threw four touchdowns against the Giants and just shredded a superior Minnesota defense in the NFC Championship Game for 352 yards and three touchdowns. If head coach Doug Pederson can devise another good offensive game plan, the Eagles could be poised for another big night of scoring against a suspect defense.
Stat No. 3: Under Pressure
These were two of the best offenses at handling pass pressure all season long, though Wentz hid a lot of Philadelphia’s protection problems with his mobility. Foles is not nearly the same threat in that regard, and he’s also not playing as well as Brady has been under pressure, which we highlighted a few weeks back.
Using a scatter plot from ESPN Stats & Info, we looked at how teams fared under pressure by comparing their Total Quarterback Rating (QBR) to their average air yards per pass attempt (how far the ball traveled relative to the line of scrimmage). Not only were the Patriots and Eagles among the top three offenses in QBR, but they were also among the five most vertical offenses when pressured, both averaging just shy of 10 air yards per attempt.
Brady in particular has really taken to throwing deep with defenders bearing down on him. Since 2015, Brady leads all quarterbacks by averaging 10.28 air yards per pass attempt while pressured, according to ESPN Stats & Info. In 2017, Brady has clearly been leaning toward the left portion of the field on passes thrown more than 15 yards beyond the line of scrimmage. He has thrown for 723 yards on 52 deep passes outside the left numbers, as opposed to 407 yards on 25 deep passes outside the right numbers.
New England wide receiver Brandin Cooks leads the NFL (including the playoffs) with 381 receiving yards on passes to the deep left this season. He could be in for a big night against a Philadelphia defense that fared better in DVOA on passes thrown to the deep right (-10.8 percent) than the deep left (6.5 percent).
Stat No. 4: The Predictable Game Script?
In their previous seven Super Bowls, the Patriots have never scored in the first quarter. Every game was decided by 1 to 6 points, and either Brady or Eli Manning led a game-winning drive in the fourth quarter or overtime. This season, the Patriots and Eagles led the league in net points per drive, with 1.04 and 0.79 points respectively, and both are accustomed to playing with the lead, though the Patriots needed late game-winning touchdowns to topple the Steelers (in Week 15) and Jaguars in crucial matchups.
We turned to EdjFootball, part of the EdjSports analytics site, for its Game Winning Chance probability data to see which team had the higher average win probability through each of the first three quarters of their 2017 games.
Both teams are seldom in trouble
The average Game Winning Chance of each Super Bowl team at the end of the first through third quarters
Game winning chance | ||||
---|---|---|---|---|
Split | Team | Q1 | Q2 | Q3 |
Regular season | New England | 71.5% | 78.2% | 78.8% |
Philadelphia | 69.6 | 70.4 | 72.0 | |
Postseason | New England | 58.8 | 69.5 | 58.1 |
Philadelphia | 59.4 | 74.0 | 82.0 | |
Total | New England | 70.1 | 77.2 | 76.5 |
Philadelphia | 68.4 | 70.8 | 73.1 |
Including the playoffs, the Patriots hold a slight edge over the Eagles in each quarter, boasting an average Game Winning Chance of 76.5 percent to start their fourth quarters compared with 73.1 percent for the Eagles. However, the Patriots have had a rougher go of things in the playoffs; they held just a 16.2 percent Game Winning Chance to start the fourth quarter against Jacksonville in the AFC Championship. Of course, it sure felt like the Patriots would come back from that deficit more than 16 times given 100 chances, but that’s just because we’ve seen this play out so often.
No matter how well the Eagles start this game, we know the Patriots have the ability to come back. To make matters worse, Philadelphia’s worst defensive quarter this season was the fourth quarter, in which its DVOA fell to 14th after ranking in the top five in each of the first three quarters. While the Eagles have some experience at holding big leads, they weren’t doing so against an experienced New England team with Brady and Belichick. The Eagles have to be able to exploit the holes in New England’s defense to put a big number on the scoreboard and keep Brady off the field as much as possible.