What advanced stats show about New England's Schedule + Patriots Defense (UPDATED with playoff #s)

[Soul_Survivor88]

Saw an interesting piece in the Globe by Alex Spier. I don't subscribe to Dan Shanuessy's "Tomato Can" nonsense, nor do I feel having a harder schedule necessarily makes you a Super Bowl winner as Felger has tried to tell us....but I think advanced stats might give us an interesting picture.

Alex Speier: The Patriots’ defense looks good, but just how good is it? - The Boston Globe

• As measured by Football Outsiders, the Patriots faced by far the worst group of offenses in the NFL this year. Indeed, according to Football Outsiders, New England’s defense had the easiest schedule (with their opponents collectively 7.1 percent below average) of any team in the NFL over the last six years. None of the Patriots opponents this year finished in the top seven in points scored.

• In terms of DVOA, the Patriots had just the 16th-best defense in the NFL this year – the worst mark in the era of the 12-team playoff format (1990 to present) for any team that led the league in points allowed. Football Outsiders calculates Defense-adjusted Value Over Average (DVOA) by comparing a team’s defensive performance on every play to a league-average outcome while adjusting for the context of the game situation and quality of opponent.

• There have been six Super Bowl winners since the 1990 season that finished 16th or worse in the Football Outsiders metric. Most recently the 2012 Ravens, who were decimated by injuries during the regular season, finished 19th in defensive DVOA but saw enough parts come together in January to roll through four playoff wins.

• That said, of those six Super Bowl winners with Patriots-like DVOA ranks, five of those featured a top-10 defense (in terms of DVOA) the year before winning the Super Bowl – including those 2012 Ravens, who led the NFL in DVOA in 2011. That suggests a team with a solid defensive core that may have been affected by various factors, including injuries. Four of those six finished in the top five in the year before they hoisted the Lombardi. The only exception was the 2009 Saints, a prolific turnover-generating group that ranked 17th in DVOA (one year after finishing 26th) en route to a championship … and infamy in the form of the Bountygate scandal.

Updated with stats courtesy of Scott Kacsmar via Football Outsiders and Ryan Hannable of WEEI

Patriots defense looks good on paper, but with lack of competition real answer will come Sunday

After the Seattle game, the Patriots have faced only one team that finished the year not in the bottom third in points per game, and that was the Dolphins in Week 17, who were without their starting quarterback and had backup Matt Moore start. They also faced the worst offense in the league (Rams) and the third-worst offense (Texans) twice over the course of the season.​

FOOTBALL OUTSIDERS: Innovative Statistics, Intelligent Analysis | AFC Championship Preview 2017

Out of the 108 teams to reach the Conference Championship Game since 1990, the 2016 Patriots faced the sixth-easiest slate of quarterbacks based on average DVOA, and were the only defense to face quarterbacks with negative total DYAR. The 2016 Patriots and the 2002 Buccaneers are the only teams to not face a single quarterback who finished in the top 12 in DVOA that regular season.​

 

[The Scrizz]

DVOA is literally pulled out of their asses. I can understand that they haven't played great offenses. The way I see it they have 2 big things going for them. They tackle well (including their cb's) and don't give up big plays.

 

[RayClay]

Who cares? Do the defenses perform well at the end of the season? I don't see many problems. Secondary seems good. Experienced safeties, good cornerbacks, with a stud, good tall CB and a good ballhawk type. Behind them are experienced backup and some young corners,

Compare that to other defenses.

Linebacker? One great one, two guys who are playing fine complementary football. Possible role player Mingo in reserve, who has great speed. Roberts is back to being a tough early down player. They're getting it done IMO, regardless of rating.

Good three man rotation at DT, good 4 man rotation at DE.

Whatever number crunchers say, who has a defense with less weaknesses? It's not a pro bowl group, but we can leave that to the teams for whom the pro bowl is their only prize.

 

[RayClay]

By the way, it's interesting that shank uses the term "tomato can" so much, since his head literally looks like a tomato can.

 

[patchick]

I have no problem with people who point out that the #1 scoring defense was "wind-aided." There's no denying that the Patriots played an uncommonly lousy lineup of quarterbacks this year.

My optimism about the defense doesn't come from the stats but from watching them over the final stretch of the season. It's been a work in progress but it seems to be coming together well with new key cogs like Rowe, Flowers and Van Noy hitting their strides. For once, there isn't a single defensive position that stands out as a weak link.

 

[UGAPatsfan]

I'm going to go on record with an unpopular opinion which I hope is proven 100% wrong:

I don't believe this team will win the Super Bowl, and defense will be the reason. This team has given up plays to HORRIBLE offenses. There have been long stretches with zero pass rush and I believe that will end up biting us. We may me able to get through the AFC but I can't see this defense holding up against many of the NFC offenses. Again, I would love to be wrong, it's just this gut feeling I had that this team just doesn't have enough horses on D.

 

[stinkypete]

I'm going to go on record with an unpopular opinion which I hope is proven 100% wrong:

I don't believe this team will win the Super Bowl, and defense will be the reason. This team has given up plays to HORRIBLE offenses. There have been long stretches with zero pass rush and I believe that will end up biting us. We may me able to get through the AFC but I can't see this defense holding up against many of the NFC offenses. Again, I would love to be wrong, it's just this gut feeling I had that this team just doesn't have enough horses on D.

What defense does not give up big plays now and then? You don't have to be the 2000 Ravens to be a championship caliber D.

 

[deroc5050]

I'm going to go on record with an unpopular opinion which I hope is proven 100% wrong:

I don't believe this team will win the Super Bowl, and defense will be the reason. This team has given up plays to HORRIBLE offenses. There have been long stretches with zero pass rush and I believe that will end up biting us. We may me able to get through the AFC but I can't see this defense holding up against many of the NFC offenses. Again, I would love to be wrong, it's just this gut feeling I had that this team just doesn't have enough horses on D.

I am actually more worried about the offense not punching it into the endzone and settling for FG's is the reason that could derail the Patriots. But I still think they are at the very least going to the Super Bowl.

 

[hovis]

I love FO, but I'm not sure if their DVOA stat properly accounts for "bend but don't break" style defenses. Their ratings can be insightful, but like any metric, can't be viewed in a vacuum.

 

[SammyBlueCat]

Most of us watched the games and can testify regarding the vaunted "eye test".
Forget the stats - the defense looked mediocre early in the season and improved steadily.

 

[Vindicate]

It is a point to be addressed but as a lot have noted, this defense has performed to great standards against these offenses. They've done what is expected and then some. Hopefully this translates to being serviceable and effective against great offenses. We're not going to shutout everyone, or hold them to one touchdown but there it is as if people think that a good offense will score 30+ points on us is making big assumptions.

Seahawks game was fresh out of a weird funk for the defense, a lot of big changes, still acclimating our LB core. And that game was also a result of terrible communication. That wasn't DB's getting beat every play, it was DB's literally not covering Baldwin for 20+ yard gains. While that sounds worse, it is possible to address communication errors. When you drop 9 into coverage and someone is wide open, that is communication to a T.

Giants have a historically bad offense (1.6 TD's a game). And then the teams with potent offenses have defenses that are ripe for the picking against Brady and co.

We can't keep looking at the Patriots in a vacuum. Every team, besides the Falcons, Patriots and Dallas, are pretty damn weird this year. It's a down year. Remember at the beginning of the year we were dreading the schedule? It was supposed to be tough.

That being said, the path to the SB is through Foxboro. Our Oline is performing VERY well against some of the best front sevens. I'm tellin' ya, as long as Brady can play against stingy defenses that get in his face during the regular season, that's battle tested enough.

 

[QuantumMechanic]

It would be interesting to see what the weighted DVOA figure is for the NE defense. FO has two DVOA numbers -- the full season one and the one which weights recent games more heavily. I believe that by the end of the season the weighted DVOA pretty much ignores the first 8ish games.

Also, ranking is meaningless (in anything, not just DVOA). You want to look at the actual numbers. If for some stat, any stat you have:
1) Team A - 16.0
2) Team B - 15.9
3) Team C - 15.8
4) Team D - 10.0

then while C might be #3, it's actually in essentially a 3-way tie for #1, especially given the gap between #3 and #4.

 

[QuantumMechanic]

Ok - here's the weighted defensive numbers (the more negative a number is for the defense, the better the defense is):
1) DEN -19.8
2) NYG -18.5
3) ARI -12.4
4) TB -9.1
5) CAR -8.5
6) PHI -8.2
7) PIT -7.6
8) HOU -7.0
9) SEA -7.0
10) SD -6.7
11) NE -6.0
12) BAL -5.1

NE's full-season defensive DVOA is -1.5 and (as noted) their weighted DVOA is -6.0. So DVOA does have them being better more recently.

Another interesting bit is the run/pass breakdown (those are only given for full-season). Again, the more negative the number, the better the defense.
Not adjusted for quality of opposition: Pass 2.1, Rush -18.7
Adjusted for quality of opposition: Pass 13.9, Rush -23.7

NE's adjusted pass number is worse than their unadjusted one because NE has played crappy passing teams. But the adjusted rush number is better than their unadjusted one because NE has played decent running teams.

FOOTBALL OUTSIDERS: Innovative Statistics, Intelligent Analysis | 2016 DEFENSE EFFICIENCY RATINGS

 

[Tony2046]

Ok - here's the weighted defensive numbers (the more negative a number is for the defense, the better the defense is):
1) DEN -19.8
2) NYG -18.5
3) ARI -12.4
4) TB -9.1
5) CAR -8.5
6) PHI -8.2
7) PIT -7.6
8) HOU -7.0
9) SEA -7.0
10) SD -6.7
11) NE -6.0
12) BAL -5.1

NE's full-season defensive DVOA is -1.5 and (as noted) their weighted DVOA is -6.0. So DVOA does have them being better more recently.

Another interesting bit is the run/pass breakdown (those are only given for full-season). Again, the more negative the number, the better the defense.
Not adjusted for quality of opposition: Pass 2.1, Rush -18.7
Adjusted for quality of opposition: Pass 13.9, Rush -23.7

NE's adjusted pass number is worse than their unadjusted one because NE has played crappy passing teams. But the adjusted rush number is better than their unadjusted one because NE has played decent running teams.

FOOTBALL OUTSIDERS: Innovative Statistics, Intelligent Analysis | 2016 DEFENSE EFFICIENCY RATINGS

Just as the Ratbirds had to overcome their injuries and improved at the end of that season the Patriots have had to overcome personnel changes which could be considered the equivalent.

This team is primed for some SB action and with both the Oline and Dline playing at elite levels there isn't a team they can't beat.

 

[Doctor DDS]

What team in the AFC has a better defense? Houston gives up 5 more PPG and plays in the weakest division in football so they are out. KC gave up 40 more yards per game and 4 more PPG. They did play some better offenses but I would not consider them better. Pitt's defense is not very good. Denver had a defense that could give a team problems but they are gone. I think the NFC has a couple defenses that could cause the Pats some concerns (Seattle and NY) but even then I'd expect the Pats to score at least mid 20's.

 

[convertedpatsfan]

I don't think it invalidates everything, but the first point seems wrong. The Cardinals finished 6th in points scored this year (it says no opponent in top 7 in scoring). They averaged 26.1 points per game, and we held them to 21. That difference held them back from overtaking the Cowboys for #5 in scoring offense btw.

Meanwhile, the Texans played 3 teams in the top 7 in scoring, losing to all 3. They gave up 27 to the Patriots and our 3rd-string QB, and 27 to Oakland. Their best performance against the 3 was probably against Rodgers and Green Bay, as they held them to 21, including 7 in the first half. But late in the 3rd, they wore down and Rodgers scored TDs on back to back drives in the 3rd/4th quarters to put the game away. Does that make their D sooooooooooooo much more legit because they played 3 teams in the top 7 in scoring, even though they lost them all and didn't really hold them back too much, if at all?

Texans fans will claim the D did well against their divisional rival Colts, who finished 8th in points this year, but somehow the Bills finished 10th in points scored and we held them below their average and nobody gives a ****.

We hold the Jets to 17 points and 3 points, and all I see is that we haven't played anyone. The Ravens D gives up 24 points to the Jets, but they're perhaps the most complete D in the league. We gave up 3 points to the Broncos and it's all about how Denver's offense sucks, but that #1 D in Houston gives up 27 to that same terrible Broncos offense and not a ****ing peep. We gave up 31 to the Seahawks, so our D sucks against good competition. But their great D gave up 24, and was one play away from giving up 7 more but their D is great.

This is not the 2015 Broncos, I know that. But this might be the 2nd-best D left in the playoffs after the Giants, especially after the Earl Thomas injury crippled the Seahawks D (and let that be a reminder how valuable a free safety can be, even if they're not making highlight reel plays...thank you D-Mac!). Some may prefer Houston, which gets more pressure, but any D giving up 27 to a 3rd-string QB probably isn't going to keep Josh up late at night.

 

[TBR]

DVOA is an ok metric, but let's forget it for a second. I'd be interested in finding out which teams had the 5 easiest schedules in terms of average opponent's offensive PPG, for each of the last few years. Then finding out what those team's PPG allowed was.

 

[kkmack617]

DSRS is a good stat we can look at to try to balance the #1 scoring defense vs easy schedule issue. DSRS measures how many points your defense gives up relative to offenses it has faced. So if your opponents averaged 20 PPG, and your scoring defense gives up 19 PPG, that is +1 SRS.

DSRS
-------
1) DEN = 6.1
2) NYG = 5.4
3) NE = 5.0
4) SEA = 4.5
5) KC = 4.4

So yes, we've faced the easiest schedule in the league, but we're still #3 in scoring defense when you adjust for the level of the competition.

I'd also note that the quality of your offense affects this somewhat. Field position, turnovers, sustaining drives, etc., all can make things harder or easier on your defense. The Patriots have an elite offense so 3rd best defense is probably slightly misleading.

My personal opinion from watching all year and also looking at all the various stats is that our defense is probably around the 8-10th best in the league. There's no reason we can't win a Super Bowl with this team. The 2014 defense was very similar statistically, and that year there were much more dangerous opponents than this year.

 

[luuked]

I don't think it invalidates everything, but the first point seems wrong. The Cardinals finished 6th in points scored this year (it says no opponent in top 7 in scoring). They averaged 26.1 points per game, and we held them to 21. That difference held them back from overtaking the Cowboys for #5 in scoring offense btw.

Meanwhile, the Texans played 3 teams in the top 7 in scoring, losing to all 3. They gave up 27 to the Patriots and our 3rd-string QB, and 27 to Oakland. Their best performance against the 3 was probably against Rodgers and Green Bay, as they held them to 21, including 7 in the first half. But late in the 3rd, they wore down and Rodgers scored TDs on back to back drives in the 3rd/4th quarters to put the game away. Does that make their D sooooooooooooo much more legit because they played 3 teams in the top 7 in scoring, even though they lost them all and didn't really hold them back too much, if at all?

Texans fans will claim the D did well against their divisional rival Colts, who finished 8th in points this year, but somehow the Bills finished 10th in points scored and we held them below their average and nobody gives a ****.

We hold the Jets to 17 points and 3 points, and all I see is that we haven't played anyone. The Ravens D gives up 24 points to the Jets, but they're perhaps the most complete D in the league. We gave up 3 points to the Broncos and it's all about how Denver's offense sucks, but that #1 D in Houston gives up 27 to that same terrible Broncos offense and not a ****ing peep. We gave up 31 to the Seahawks, so our D sucks against good competition. But their great D gave up 24, and was one play away from giving up 7 more but their D is great.

This is not the 2015 Broncos, I know that. But this might be the 2nd-best D left in the playoffs after the Giants, especially after the Earl Thomas injury crippled the Seahawks D (and let that be a reminder how valuable a free safety can be, even if they're not making highlight reel plays...thank you D-Mac!). Some may prefer Houston, which gets more pressure, but any D giving up 27 to a 3rd-string QB probably isn't going to keep Josh up late at night.

The mistake you make is bringing nuance and logic to a topic about meaningless 'next-gen stats'.

There are so many issues with any metric that tries to explain the complex game of football with a few numbers. It's a game of matchups where the worst team (e.g. the Rams) can beat a highly rated team (e.g. the Seahawks) just because it has favorable matchups in the key positions.

No cumulative stat in the world will be able to make sense of that.

Similarly, I'd say that there is a significant difference between the Patriots D in the first third of the season.. (i.e. the one with Collins and Freeny as LBs) and the one that finished the season.

How does anything that happened in September or October have any effect on where the team is now ?

But people love debating their stats and whatnot because looking at numbers is easier than to sit down and watch the tape and actually think about what you see on a more fundamental level.

 

[Ring 6]

I am actually more worried about the offense not punching it into the endzone and settling for FG's is the reason that could derail the Patriots. But I still think they are at the very least going to the Super Bowl.

Interesting concern about the top red zone offense of all teams that made the playoffs.