Previous post:

Next post:

The Best Scoring Defenses In NFL History

by Chase Stuart on February 3, 2014

in Defense, History, SRS

Head of the LOB

Head of the LOB.

Congratulations to the Seattle Seahawks and their fans on winning Super Bowl XLVIII. With the win, Seattle has confirmed its status as one of the greatest defenses in NFL history. The Seahawks defense produced a game for the ages on Sunday: facing Peyton Manning, Demaryius Thomas, and one of the greatest offenses ever, Seattle’s defense outscored Denver’s offense, 9-8. Led by Malcolm Smith, Cliff Avril, Kam Chancellor, Earl Thomas and Richard Sherman, the Seahawks stamped their claim with the ’85 Bears, ’00 Ravens, and ’02 Bucs as one of the greatest defenses of the last 30 years.

But today, I want to look at which defenses were the best in regular season history, and see where Seattle stacks up. Bill Barnwell had an interesting post during Super Bowl week. He used the statistical measurement known as the Z-Score to show that Seattle was the tenth best defensive scoring team in NFL history. Don’t be too confused by the idea of a Z-score: in English, this just means that Seattle’s defense — and yes, I am going to conflate the concepts of defense and points allowed throughout this post — was 2.2 standard deviations above average in points allowed, one of just ten teams to ever produce such a result.

So how do we get there? Well, Seattle allowed 14.4 points per game during the regular season. The league average was 23.4 points, which means that Seattle’s defense was 9.0 PPG better than average. The standard deviation of points per game among the 32 NFL defenses in 2013 was 4.08 points per game; therefore, Seattle has a Z-score of 2.20, because the Seahawks allowed points at a rate that was 2.20 standard deviations better than the mean.

Today, I wanted to do the same analysis but adjust for strength of schedule, by deriving offensive and defensive SRS grades. Of course, Pro-Football-Reference has published offensive and defensive SRS grades for awhile, but I decided to crunch the numbers on my own and see if they matched up with what Neil and Mike did (they did). For the uninitiated, SRS stands for Simple Rating System, which is simple to understand but a bit complicated to derive. The SRS is simply margin of victory (or, in the case of offenses and defenses, margin of production above league average) adjusted for strength of schedule. The key is using an iterative process, where, in Excel, we adjust the ratings hundreds of times; after all, to adjust for SOS, you have to adjust for the SOS of each opponent, and the SOS of each opponent’s opponent, and so on.

The table below shows the top 200 scoring defenses since 1932. Here’s how to read the 2002 Bucs line. That season, Tampa Bay allowed 9.4 points per game less than league average. The average defense the Bucs faced — using the iterative method to derive SOS grades — was 0.4 points above average. Therefore, Tampa Bay is credited with an adjusted rating of 9.8 PPG better average. The standard deviation of defensive ratings in the NFL in 2002 was 3.45, giving the Buccaneers a Z-score of 2.83, the highest ever. The table below is fully sortable and searchable, and shows the top 200 defenses.

Rk
Team
Year
G
PPG OvAvg
SOS
Adj PPG
LgStDv
Z-Score
1TAM2002169.40.49.83.452.83
2KAN1995166.40.46.92.562.68
3BAL2006168.1-0.37.82.992.62
4MIN19701490.19.23.772.43
5CHI1985169.20.39.44.132.28
6DEN197816616.93.122.22
7MIN1971149.40.59.94.482.21
8DEN1977146.62.89.44.242.21
9SEA201316908.94.082.19
10CHI19631411.7-0.810.95.112.13
11KAN1969148.5-0.18.43.972.12
12CAR2013168.30.38.64.082.11
13KAN1997166.20.36.63.122.1
14MIA1998164.71.86.53.122.1
15SFO2011167.9-1.36.63.142.1
16NOR1992166.117.13.372.09
17CHI1988166.8-0.26.63.172.09
18SFO1995165.4-0.15.32.562.07
19PIT1975149-0.28.84.282.06
20GNB20101670.87.93.832.05
21BAL19681410.2-0.39.94.862.04
22RAM19751410.9-2.38.74.282.03
23SEA2012167.40.37.73.822.02
24PIT2010167.50.27.73.832.02
25BUF1988165.40.96.43.172.02
26PIT2008168.10.28.24.092.01
27CLE1994167.5-1.46.13.052.01
28CHI1986168.8-1.17.73.872
29NYJ2009166.70.77.53.741.99
30NWE2006165.80.15.92.991.98
31PHI19491211.3-0.3115.691.93
32KAN196814100.110.15.241.93
33PIT2011168-1.963.141.93
34BAL20001610.4-2.384.181.92
35PHI1980166.60.26.83.611.89
36GNB19621411.7-0.411.35.981.89
37BUF1999166.50.97.43.911.89
38BAL1971149.4-0.98.54.481.89
39GNB1965147.13.410.55.551.89
40MIN19691411.4-0.810.65.661.88
41NYG1993165.9-0.55.42.91.87
42NYJ1998164.71.25.83.121.87
43PIT1976149.30.39.65.171.86
44CHI2005168-1.46.63.591.84
45NYG1990166.90.77.64.151.84
46DET1962149.71.310.95.981.83
47TEN2008167.407.54.091.82
48CHI2001167.5-0.57.13.921.8
49DAL2009165.80.96.73.741.8
50NOR1991165.80.56.33.51.79
51PIT1978166.1-0.65.63.121.78
52DET1970144.81.86.63.771.76
53IND1987155.7-0.15.63.181.76
54SFO2012165.716.73.821.76
55PHI2001167.2-0.46.83.921.74
56CLE1948149.2-0.58.75.061.72
57TAM1999166.10.66.73.911.72
58DEN1984166.10.56.73.891.71
59DET1956124.726.73.931.71
60CLE1953128-174.131.71
61MIA2000166.50.57.14.181.7
62CLE1987155.7-0.35.43.181.7
63PHI1981166.9-0.26.73.941.69
64NYG1938116.3-0.85.53.281.68
65SDG1979164.715.73.421.68
66DEN1979163.725.73.421.66
67BAL19641460.56.43.861.66
68CHI1984165.70.76.43.891.66
69BAL2004164.726.84.081.66
70DAL1978165.3-0.25.23.121.65
71MIA198295.6-0.15.53.341.65
72GNB19661410.1-0.1106.061.65
73TAM1997164.30.85.13.121.65
74GNB1996167.3-0.76.64.021.63
75DAL1993164.40.34.72.91.63
76PIT1994165.6-0.653.051.63
77TEN2000168.7-1.96.84.181.63
78JAX2006163.51.34.92.991.62
79NWE2003166-1.14.92.991.62
80DEN1989166.5-0.85.73.491.62
81WAS1974144.20.44.52.811.62
82SFO2013166.40.26.64.081.62
83DEN2005164.51.35.83.591.61
84PIT1974144.7-0.24.52.811.6
85MIA1983166.2-15.23.231.6
86KAN1979163.71.85.43.421.59
87NWE2004165.21.36.54.081.59
88SDG1961148.8-0.48.45.321.58
89RAM1945104.90.55.43.41.58
90JAX20041642.56.44.081.58
91NYG19441010.5-1.68.95.681.57
92HOU1967148.5-0.18.35.311.57
93CHI2012165.40.563.821.57
94NYG1939117.7-1.36.44.091.55
95ATL1977148-1.46.64.241.55
96GNB1961145.60.96.54.191.55
97WAS198295.9-0.85.23.341.55
98DTX1960146.1-0.163.891.53
99BUF1980164.21.35.53.611.53
100PIT1972147.8-1.26.64.361.51
101PIT1946118.3-0.285.321.51
102NYG1986165.80.15.83.871.51
103MIA1973148.7-0.58.25.461.51
104TAM2003164.30.24.52.991.51
105GNB1964144.51.35.83.861.51
106WAS1943105.8-1.14.63.091.5
107GNB1941115.60.25.73.831.49
108CHI1993164.304.32.91.49
109PIT1979163.71.45.13.421.49
110PIT2001167-1.15.83.921.49
111MIA1972148-1.66.54.361.48
112IND2007165.30.15.43.641.48
113NYJ2004165.20.964.081.48
114CHI19481210.70.210.97.371.48
115DAL1994164.8-0.34.53.051.46
116IND2005165.205.23.591.45
117CLE1951129.3-1.57.85.411.44
118BAL2010165.20.35.53.831.44
119PHI1991163.71.353.51.43
120BUF1964146-0.75.33.711.43
121RAM1974145.3-1.242.811.43
122PHI19501211.21.612.88.941.43
123DET1964143.525.53.861.43
124CLE1946149.9-1.38.66.061.42
125NYG1958127.3-0.474.911.42
126PIT19571250.35.33.731.42
127WAS199116504.93.51.41
128MIA1975144.71.364.281.41
129NYG1981164.60.95.53.941.41
130PHI2007162.92.25.13.641.41
131CAR1996166.8-1.25.64.021.4
132SFO1989164.80.14.93.491.4
133RAM1970144.80.45.33.771.39
134DET1969147.50.47.95.661.39
135DET1938113.70.94.63.281.39
136CLE1947148-175.021.39
137RAM1968146.20.56.74.861.39
138SFO1984167-1.65.43.891.39
139NWE1988162.51.94.43.171.38
140MIA2003164.5-0.44.12.991.37
141CHI19591251.26.24.511.37
142MIN1988165.7-1.44.33.171.37
143BAL2008166.8-1.25.64.091.36
144DAL1995163.30.23.52.561.36
145PIT2004165.8-0.25.64.081.36
146RAM1967147.8-0.975.111.36
147CHI1932145.10.25.23.881.35
148NYG1951128.5-1.27.35.411.35
149WAS1945106.4-1.84.63.41.35
150MIA20061631.142.991.34
151BAL1967147.7-0.96.85.111.34
152BUF2003163.40.642.991.34
153CHI1942118.2-2.85.44.021.34
154SFO1987154.7-0.54.23.181.34
155NYG1959127.2-1.264.511.33
156DTX1962146.5-0.55.94.481.32
157SFO19921640.54.53.371.32
158CLE1989164.7-0.14.63.491.32
159BAL2009165.2-0.24.93.741.31
160DAL1996164.80.55.34.021.31
161CHI1937113.90.34.13.151.31
162DET1952126.30.775.341.31
163NYG1989164.9-0.34.63.491.3
164PHI19481210.2-0.79.67.371.3
165KAN1991163.21.34.53.51.3
166CHI1967146.30.46.65.111.3
167CLE1949127.1-16.14.71.3
168JAX1999167.3-2.25.13.911.29
169WAS1971145.805.84.481.29
170GNB1947124.51.96.34.911.29
171DET1937113.40.643.151.27
172PIT1996164.40.85.14.021.27
173MIA2002162.91.54.43.451.27
174MIN1973147.5-0.575.461.27
175CLE1988162.31.843.171.27
176MIN1975147.7-2.35.44.281.26
177SFO1981165-0.14.93.941.26
178BUF1966144.30.34.63.671.25
179NYJ19851650.15.24.131.25
180OAK2002162.71.64.33.451.25
181NYG19501210.40.711.28.941.25
182GNB1967146.9-0.66.35.111.24
183NOR2013164.40.654.081.23
184OAK1973147-0.36.75.461.23
185SDG2007163.90.54.53.641.22
186RAI1984163.80.94.83.891.22
187SFO19911640.24.33.51.22
188SFO1951124.91.86.65.411.22
189SDG1977142.52.65.24.241.22
190CHI2010164.20.54.73.831.22
191NYG19401130.43.32.741.22
192HOU1975144.40.85.24.281.22
193CAR2009162.22.34.53.741.21
194PHI1992163.40.74.13.371.21
195DET1961143.125.14.191.21
196PIT1990165.1-0.154.151.21
197HOU1980164.8-0.44.43.611.21
198NWE2009163.70.84.53.741.2
199CLE1957125.5-14.53.731.19
200PIT1992164.7-0.743.371.19

Seattle ranks 9th in this metric, showing that they deserve to be remembered as one of the great defenses in league history. Richard Sherman, Earl Thomas, Kam Chancellor, Bobby Wagner, and an incredibly deep defensive line combined to create one of the best defenses since the merger. One thing to note: without the second (and third, and fourth, and so on) level adjustments inherent in the SRS process, Seattle’s strength of schedule would look below average. But the offenses Seattle faced played a number of games against Seattle, San Francisco, St. Louis, and Arizona — and all four NFC West defenses ranked in the top 12 in defensive SRS — which makes those offenses look worse than they were. If you want to discuss strength of schedule, you should look at the ’00 Ravens, who fall to 34th on the list after adjusting for the weak offenses that team faced. I should also note that from 2013, Carolina (#12), San Francisco (#82), and New Orleans (#183) make the list of top 200 scoring defenses.

Let’s walk through some of the top teams:

  • It’s hard to find much fault with the ’02 Bucs. Between Warren Sapp, Simeon Rice, Derrick Brooks, Ronde Barber, and John Lynch, the team had five borderline HOF players. Tampa Bay had some dominant defensive seasons — the ’97, ’99, and ’03 teams make the list — but the 2002 team was special. In the playoffs, they were even better.
  • The 1995 Chiefs are a bit of an oddball. That season, offensive numbers exploded, and Kansas City is one of just six teams to allow 240+ points and lead the league in scoring defense. The standard deviation among defensive ratings was super tiny that season, which inflates the Chiefs’ Z-score (Kansas City was only 6.9 PPG above average). Neil Smith, Dan Saleaumua, Derrick Thomas, and Dale Carter were Pro Bowl players, but this was not an all-time great defense.
  • The ’06 Ravens come up pretty often, as Baltimore ranked 2nd in both YPC allowed and NY/A allowed, while ranking 1st in yards allowed, 1st in points allowed and 2nd in turnovers. Coached by Rex Ryan, this was a dominant unit with a rookie Haloti Ngata, in-their-primes Terrell Suggs, Adalius Thomas, Ray Lewis, Bart Scott, Chris McAlister, and Ed Reed, and good supporting players like linemen Trevor Pryce and Kelly Gregg. Had these Ravens won a Super Bowl, they’d be remembered with their ’00 brethren.
  • The Purple People Eaters are well-represented. The 1970 and 1971 teams rank in the top ten, and the 1969 team was arguably better than both! That year, Minnesota made the Super Bowl on the back of a dominant defense that allowed 11.4 points per game fewer than average, the third best rate since 1932. That was the result of a favorable schedule, but even the adjusted result of +10.6 points per game is unbelievable. The only reason the Vikings don’t look better is because in 1969, the NFL was a smaller league and the standard deviation was a wide 5.66 PPG.
  • A few Bears teams also make the list, with the ’85 team coming in 5th place. The 1986 team actually allowed fewer points, but note that (1) the league average was 1 point per game lower in 1986, making that result less impressive, and (2) the ’86 Bears faced a much easier schedule. If you want to argue that the ’85 Bears wasn’t the best Chicago defense ever, you’re better off backing the ’63 version that was an incredible 10.9 PPG better than average after adjusting for strength of schedule (one could make the same claim about the less-heralded ’48 team).
  • I like the idea of using the Z-score, but it’s far from clear that it’s the best way to analyze the results. Simply looking at points per game above average is just as (and, some would argue, more) legitimate. In that light, the 1950 Eagles rank #1. In fact, that Philadelphia team deserves more discussion.

The 1950 Eagles were coached by Greasy Neale, and began the season in ignominious fashion. The prior season, with yet another absurdly dominant defense, Philadelphia won the NFL championship by shutting out the Rams. But in the season opener, the Eagles were embarassed by Otto Graham and the upstart Cleveland Browns, playing their first NFL game after the AAFC “merged” with the senior league. Not quite prepared for the high-powered Cleveland offense, Graham went 21/38 for 346 yards and 3 touchdowns, numbers that look pretty good even today.

Over the team’s final 11 games, Philadelphia allowed just 9.6 points per game. Now, it’s easy to think “hey, it’s 1950, how impressive is that?” Well, in 2013, teams averaged 23.4 points per game. In 1950, teams averaged…. 22.9 points per game. In other words, 9.6 points per game was absurdly dominant even in 1950. The strange thing about this team was that Philadelphia had a habit of blowing out teams or losing in close games. The Eagles lost by 2 points to the Steelers, two to the Giants at home, four to the Giants on the road, four to the Cardinals, and six in the rematch in Cleveland. Meanwhile, every game Philadelphia won was by at least a touchdown. As a result, the Eagles went 6-6, despite the team’s average win coming by a score of 35-9. If they’re not the greatest .500 team ever, I’m not sure who is.

We can use the same method to analyze the worst defenses ever:

Rk
Team
Year
G
PPG OvAvg
SOS
Adj PPG
LgStDv
Z-Score
1BAL198116-12.60.6-123.94-3.06
2TAM198616-9-1.8-10.83.87-2.79
3NOR198016-100.6-9.43.61-2.6
4WAS195412-14.1-1.9-16.16.33-2.54
5NWE197214-11.60.6-11.14.36-2.53
6IND200116-10.20.3-9.83.92-2.51
7SEA197614-11.5-1.4-12.95.17-2.5
8DET200916-9.40.2-9.23.74-2.46
9CIN19348-19.61.6-187.34-2.45
10HOU198316-6.9-0.9-7.93.23-2.44
11DAL196012-9.20.5-8.73.69-2.35
12RAM19829-7.6-0.1-7.83.34-2.32
13STL200016-8.8-0.7-9.54.18-2.27
14NYG196614-14.10.3-13.76.06-2.27
15HOU197314-12.50.1-12.35.46-2.26
16DTX195212-13.31.3-125.34-2.25
17NOR197714-6.8-2.6-9.44.24-2.22
18DET200816-10.31.2-9.14.09-2.22
19ATL199616-8.4-0.5-8.94.02-2.21
20GNB198316-5.6-1.5-7.13.23-2.21
21CRD194410-14.82.4-12.45.68-2.18
22HOU19829-7.10-7.13.34-2.12
23WAS194712-8.6-1.7-10.44.91-2.11
24CIN199916-7.9-0.3-8.23.91-2.1
25SFO200416-6.8-1.8-8.64.08-2.1
26MIN198416-90.9-8.23.89-2.1
27WAS199416-5.5-0.9-6.43.05-2.1
28TAM201116-8.72.1-6.63.14-2.09
29ARI200316-7.41.2-6.22.99-2.08
30CIN199816-70.5-6.53.12-2.08
31PIT193311-9.2-0.9-10.14.89-2.06
32SDG200316-6.70.6-6.22.99-2.06
33PIT194111-8.60.8-7.93.83-2.05
34DEN201016-7.4-0.4-7.83.83-2.03
35SEA197714-9.50.9-8.64.24-2.02
36CIN200216-6.8-0.1-6.93.45-2.01
37BUF197114-8.8-0.1-8.94.48-1.99
38DEN196414-80.7-7.33.71-1.97
39CIN199116-8.21.3-6.93.5-1.97
40ARI200016-7-1.1-8.14.18-1.93
41DET200216-6.5-0.1-6.73.45-1.93
42TEN201216-6.7-0.7-7.43.82-1.93
43WAS199816-5-0.9-63.12-1.92
44BAL197816-82-63.12-1.92
45DEN196314-10.60-10.65.59-1.9
46NYJ199616-7.90.3-7.64.02-1.9
47ATL199216-7.10.7-6.43.37-1.9
48HOU196614-5.8-1.1-6.93.67-1.88
49CLE199016-8.81-7.84.15-1.88
50NYG196414-6.5-0.8-7.23.86-1.87
51ATL198715-7.51.5-5.93.18-1.87
52ARI199816-2.3-3.5-5.83.12-1.86
53ARI201016-5.1-2-7.13.83-1.86
54PIT193512-6.5-1.2-7.74.17-1.85
55ATL200316-5.50.1-5.42.99-1.82
56STL199516-4.60-4.62.56-1.81
57ATL199316-5.40.1-5.22.9-1.8
58BOS197014-6.5-0.2-6.83.77-1.8
59CHR194714-9.20.2-95.02-1.79
60STL200716-5.7-0.8-6.53.64-1.79
61SFO197916-5.9-0.2-6.13.42-1.79
62STL200516-6.2-0.2-6.43.59-1.78
63SDG198316-71.3-5.73.23-1.78
64MIN200216-6-0.1-6.13.45-1.76
65CIN199416-5.1-0.3-5.43.05-1.76
66NOR199916-6.3-0.6-6.93.91-1.76
67SFO200616-5.1-0.2-5.22.99-1.75
68SDG198516-5.7-1.6-7.24.13-1.75
69IND199316-4.9-0.1-5.12.9-1.74
70RAM193811-6.10.3-5.73.28-1.74
71CHI201316-6.5-0.6-7.14.08-1.74
72SFO199916-7.50.7-6.83.91-1.73
73BUF197014-4.8-1.7-6.53.77-1.73
74CRD195712-5.1-1.3-6.43.73-1.73
75HOU196514-9.2-0.2-9.45.43-1.72
76PIT198816-6.10.6-5.53.17-1.72
77BUF198416-7.20.5-6.73.89-1.72
78STL200816-70-74.09-1.71
79DET198715-4-1.4-5.43.18-1.71
80CHI199716-5.60.2-5.33.12-1.71
81TAM197614-10.31.4-8.85.17-1.71
82NWE199016-7.80.7-7.14.15-1.7
83CIN199216-4-1.7-5.73.37-1.7
84IND201116-4.7-0.6-5.33.14-1.7
85ATL198916-6.70.8-5.93.49-1.69
86CLE197414-6.41.6-4.72.81-1.69
87BAL199616-7.10.3-6.84.02-1.69
88IND199816-6.51.2-5.33.12-1.69
89PHI194011-4.1-0.5-4.62.74-1.69
90MIN201316-6.6-0.3-6.94.08-1.68
91CRD195812-7.1-1.2-8.24.91-1.68
92CIN198516-5.8-1.1-6.94.13-1.67
93NYJ197514-10.33.2-7.14.28-1.66
94PHI193612-5.3-0.7-63.61-1.65
95BAL195012-15.60.8-14.88.94-1.65
96CLE199916-6.50-6.53.91-1.65
97IND199116-4.8-0.9-5.73.5-1.64
98ATL198516-6.7-0.1-6.84.13-1.64
99GNB195612-8.11.7-6.43.93-1.63
100OAK201216-4.9-1.3-6.23.82-1.63
101SDG19829-4.4-1-5.43.34-1.63
102OAK201116-4.9-0.2-5.13.14-1.62
103MIN196114-7.60.8-6.84.19-1.62
104OAK199716-5.40.4-53.12-1.61
105PHI194310-3.6-1.4-53.09-1.61
106ARI200716-3.3-2.6-5.93.64-1.61
107NYY195112-9.91.2-8.75.41-1.6
108WAS195912-7.80.6-7.24.51-1.6
109DET194611-9.30.8-8.55.32-1.59
110KAN200916-5-0.9-5.93.74-1.59
111DEN200816-6-0.5-6.54.09-1.59
112STL200616-3.2-1.6-4.72.99-1.58
113GNB195112-9.30.8-8.55.41-1.58
114RAM199016-5.6-0.9-6.54.15-1.58
115HOU200516-6.30.7-5.73.59-1.57
116CRD195912-5.7-1.4-7.14.51-1.57
117CHI194510-5-0.3-5.33.4-1.57
118WAS196114-6.50-6.54.19-1.55
119ARI199516-4.90.9-42.56-1.55
120HOU198416-6.10.1-63.89-1.55
121SIS193212-6.20.2-63.88-1.55
122STL200916-5.80-5.83.74-1.54
123TEN200516-5.70.2-5.53.59-1.54
124CHI197514-6.5-0.1-6.64.28-1.54
125SFO200516-6.10.6-5.53.59-1.53
126SDG199716-5.81-4.83.12-1.53
127SFO195512-4-0.4-4.42.84-1.53
128DAL196114-5.6-0.8-6.44.19-1.53
129DET200616-4.2-0.4-4.62.99-1.53
130HOU197014-5.90.1-5.83.77-1.53
131SDG197314-8.1-0.2-8.35.46-1.52
132HOU198916-5.1-0.2-5.33.49-1.52
133DEN199416-4.5-0.1-4.63.05-1.52
134GNB195812-9.21.8-7.44.91-1.51
135NOR199416-5.20.6-4.63.05-1.51
136NYG198016-6.10.6-5.53.61-1.51
137NYT196214-7.10.3-6.84.48-1.51
138BUF201216-4.4-1.3-5.83.82-1.51
139KAN198715-4.3-0.5-4.83.18-1.51
140BAL195312-7.71.5-6.24.13-1.5
141MIA199316-3.2-1.1-4.32.9-1.5
142ATL196614-9.50.5-96.06-1.49
143TAM199316-4.80.5-4.32.9-1.49
144NYJ198916-5.1-0.1-5.23.49-1.49
145CHR194814-8.51-7.55.06-1.48
146SDG197114-5-1.6-6.64.48-1.48
147PHI196714-7.4-0.2-7.55.11-1.47
148MIA198616-4.8-0.9-5.73.87-1.47
149PHI19339-7.90.7-7.24.89-1.47
150ARI200816-4.6-1.4-64.09-1.46
151BAL197414-5.31.2-4.12.81-1.46
152SFO197816-3.5-1-4.53.12-1.45
153BUF200116-60.4-5.73.92-1.45
154BUF201116-4.90.4-4.53.14-1.44
155ARI199616-4.4-1.4-5.84.02-1.44
156JAX199516-3.80.1-3.72.56-1.44
157SEA201016-3.4-2.1-5.53.83-1.44
158GNB198616-5.60-5.63.87-1.44
159CLE197816-3.9-0.6-4.53.12-1.43
160ARI200616-3.7-0.6-4.32.99-1.43
161CRD195312-6.60.7-5.94.13-1.43
162DAL196214-6.4-2.1-8.55.98-1.43
163PIT196814-7.91-6.94.86-1.42
164GNB195712-6.10.9-5.33.73-1.41
165DET199016-5.7-0.2-5.94.15-1.41
166PHI197314-8.60.9-7.75.46-1.41
167ATL196714-8.31.1-7.25.11-1.41
168NYG197114-6.50.2-6.34.48-1.4
169PIT196914-80-7.95.66-1.4
170SFO198016-5.50.4-5.13.61-1.4
171MIN196314-5.9-1.3-7.15.11-1.4
172STL196914-6.9-0.9-7.85.66-1.38
173MIA199116-2.8-2-4.83.5-1.38
174OAK196114-8.30.9-7.35.32-1.38
175NOR197514-5.1-0.7-5.94.28-1.37
176KAN197714-7.81.9-5.84.24-1.37
177ATL197916-4.2-0.5-4.73.42-1.37
178DET200716-6.11.1-53.64-1.36
179TAM198916-5.60.8-4.73.49-1.36
180HOU197214-6.91-5.94.36-1.35
181NOR200516-4.3-0.6-4.93.59-1.35
182NYG194812-9.1-0.9-9.97.37-1.35
183STL198516-4.3-1.2-5.64.13-1.35
184PIT196514-5.3-2.2-7.55.55-1.34
185DET200116-6.31-5.23.92-1.34
186NOR201216-5.60.5-5.13.82-1.34
187KAN197614-7.70.8-6.95.17-1.34
188TAM198516-6.51-5.54.13-1.33
189ATL200716-4.2-0.7-4.83.64-1.33
190DET194812-10.70.9-9.87.37-1.33
191BOS196814-6.8-0.1-6.95.24-1.32
192NYY194912-8.20.6-7.55.69-1.32
193MIN201116-5.91.7-4.13.14-1.32
194NWE199216-4-0.5-4.53.37-1.32
195WAS201316-6.51.1-5.44.08-1.32
196SFO200016-5.70.2-5.54.18-1.32
197CLE200016-5.50-5.54.18-1.32
198NYJ197414-3.2-0.4-3.72.81-1.31
199GNB195312-6.71.3-5.44.13-1.31
200NOR196914-7.2-0.2-7.45.66-1.31

The amazing thing about the 1981 Colts isn’t how bad they were, but that they managed to actually sweep the Patriots. Of course, New England was 2-14 that year, but so was Baltimore. The Colts teams of the early ’80s were horrible, particularly on defense, and went 34 straight games without being a favorite.

Also, honorable mention goes to the terrible Washington defense in 1954. That year, the team allowed 36 points per game, including five games of over 40 points. That might be the worst defense since World War II. Finally, three teams from 2013 make the list, including the Bears and Vikings. Those NFC North defenses weren’t just bad, but also faced the AFC North and NFC East, meaning they had easy strengths of schedule, too.

{ 34 comments… read them below or add one }

Shattenjager February 3, 2014 at 12:44 pm

The ’50 Eagles do have their numbers a bit artificially inflated by the second Browns game, though: http://www.pro-football-reference.com/boxscores/195012030cle.htm

Some weird blogger discussed the game on this podcast years ago: http://www.pro-football-reference.com/blog/?p=2720

For anyone who doesn’t want to look or listen, Paul Brown was apparently more interested in making the point that his team could beat the Eagles without passing (literally) than with scoring the way his team was capable of doing. It wouldn’t make a big difference to the numbers for that defense, but it would make a bit of one.

Reply

Chase Stuart February 3, 2014 at 1:02 pm

Heh, I had thought of mentioning that game but figured the post was running on long enough as it was. Good memory!

Reply

Shattenjager February 3, 2014 at 9:40 pm

I have listened to those podcasts far more times than a person should. I still keep all of the non-trivia episodes on my iPod. And probably my favorite segment on the p-f-r podcast ever was either that one or the Keith Lincoln one.

And since you’ve seen my blog, you know that I do not share the world’s concern with blog post length! ;)

Reply

Richie February 3, 2014 at 7:41 pm

That is awesome. A game with zero pass attempts. I don’t remember the podcast discussion about it.

Reply

NWebster February 13, 2014 at 2:00 pm

Interesting game.

The lore is that Paul Brown’s gameplan was to stick it in the nose of Neale who derisively referred to the Browns as a “finesse” team. This is actually a bit of a simple view, the Browns actually dropped back to pass twice in the game, one one occation they were called for an offensive hold and the play negated, on the other occation Graham was sacked . . . it’s not noted in the PFR Boxscore, those are a little sparce in the 1950 season.

Also, it was played on a very muddy field in horrible weather, and from a gamescript perspective the Browns jumped out to a big early lead . . . probably had a gamescript of +8 or +9 for the Brown, the type of game you’d expect a team to be run heavy.

Reply

Chase Stuart February 3, 2014 at 1:13 pm

A Seattle 2013/Baltimore 2000 comparison is interesting. Remember, I’m not looking at anything other than points allowed and SOS, so this is just a basic analysis. That said, what’s interesting is how Seattle closes a pretty big gap.

In 2000, the Ravens allowed just 10.4 points per game. In 2013, Seattle allowed 14.4 points per game. So how does Seattle get rated ahead of Baltimore? Well, in 2000, the average team allowed 20.7 PPG, while in 2013, the average team allowed 23.4 PPG. That’s a pretty big jump in just thirteen years, so Baltimore is graded as 10.4 PPG better than average, and Seattle is 9.0 PPG better than average. So an era adjustment cuts the gap from 4 points to 1.4 points.

But the SOS adjustment is nearly as large. Seattle faced a completely neutral schedule, while Baltimore faced a schedule that was 2.3 PPG easier than average. The average offense the Ravens faced — even after adjusting for the fact that all those offenses faced Baltimore and Tennessee and Pittsburgh twice — was 2.3 points per game below average. That includes Cleveland (9.1 PPG below average) and Cincinnati (8.1) twice, along with the Cardinals (-7.2). Those were the three worst offenses in the NFL, and Baltimore faced them five times (and also San Diego (-3.7) and Washington (-2.9), who also ranked in the bottom eight).

As a result of the era adjustment (2.7 points) and the SOS adjustment (2.3 points), Baltimore loses 5 PPG on Seattle’s defense, moving from them 4.0 points ahead to 0.9 PPG behind. If you ignore the Z-score and look at just the PPG above or below average, the top defenses since 1978 are Tampa Bay 2002 (+9.8), Chicago 1985 (+9.4), Seattle 2013 (+8.9), Carolina 2013 (+8.6), Pittsburgh 2008 (+8.2), and Baltimore 2000 (+8.0). That’s five Super Bowl winners out of six teams, the maximum possible given this combination.

Reply

Chase Stuart February 3, 2014 at 3:12 pm

Another thing to add to the offseason to-do list: look at drives faced.

According to Football Outisders, Seattle faced 179 drives (lgavg = 186), ’02 Bucs faced 194 drives (lgavg = 183), and ’00 Ravens faced 189 drives (lgavg = 185). So that would drop Seattle and bump Baltimore, and really bump up Tampa Bay. The Bucs have the 2nd best scoring defense (Adj PPG) since the merger behind only the ’71 Vikings, and they did it despite facing more nearly one full game worth of extra drives.

Reply

Dan February 3, 2014 at 1:49 pm

Sounds like BCS computer logic to me! Era adjustment? SOS adjustment? The only metrics should be: YPG allowed, Points allowed, rushing yards per attempt, passing yards per attempt. These are hard numbers that cant be disputed statistically. These “adjustments” are purely based on opinion.

Reply

Richie February 3, 2014 at 7:44 pm

You don’t think the era matters? You don’t think it matters that rules for defensive backs changed in 1978, and again in 2005(?)? Or over the past couple years that wide receivers and QB’s are even MORE protected from hits? You don’t think it’s a little easier for a WR to go across the middle now, because he doesn’t have to fear having his head crushed?

Reply

NWebster February 13, 2014 at 2:48 pm

Yeah that 1977 Falcons D was way better than these Seahawks.

Reply

Alex Miller February 3, 2014 at 2:41 pm

These are just regular season numbers right?

It should be pointed out that the 2000 Ravens faced the #2 and #3 scoring offenses in the NFL in the playoffs (Broncos and Raiders)…and completely dismantled them. They gave up a combined 6 points and below 200 yards of total offense in both games.

Reply

Chase Stuart February 3, 2014 at 2:46 pm

Correct. But Seattle’s playoff performance was arguably even better, as dismantling the 600+ point Broncos is hard to top. And the ’85 Bears and ’02 Bucs were ridiculous in the playoffs, too. But you would certainly vault the ’00 Ravens over the Steelers and Panthers teams, and have a nice pantheon of four elite defenses: the ’85 Bears, ’00 Ravens, ’02 Bucs, and ’13 Seahawks. Based purely on points allowed, I’d probably put Baltimore 4th in that group. But the differences are small enough that you could conceivably rank them in any order.

And, of course, there’s no reason to limit yourself to just points allowed when ranking defenses. I just thought this would be an interesting way to do it, and era/SOS-adjusted PA ratings is certainly better than looking just at points allowed, even if it’s far from perfect.

Reply

Rod February 3, 2014 at 11:17 pm

The 2002 Bucs defense faced the #1 offense in scoring and yards in the Super Bowl and outscored them 21-15 (of course that’s IF you give the Raiders the BS Joey Porter TD that clearly was not a TD – as for the other points the Raiders got was on a blocked punt). Anyways they scored against the Eagles and had a score against the 49ers that was called back. So that Bucs defense in 3 games scored 4 TDs and should have had 5. So when this article says best scoring defense, they were not only the best defense in terms of keeping opponents from scoring, they also were the best in terms of producing TDs.

Reply

Brian February 3, 2014 at 4:52 pm

Let’s not leave the fabled Orange Crush defense of the late 70′s Denver Broncos out of the discussion. They posted two scores in the top 10 using this metric. I happened to see a replay of Superbowl XII before this year’s game…A talented Dallas offense (No. 2 PPG in the NFL that year) benefited from seven (SEVEN!) first-half Denver turnovers and a horrible call that took away a Denver interception of Staubach in the end zone (result, Dallas FG), yet Denver only trailed 13-0 at the half. Dallas’s lone touchdown drive that half started at the Denver 25…

Reply

Chase Stuart February 5, 2014 at 3:23 pm

With two top-ten seasons, I think they are obviously in the discussion.

Reply

Nick Bradley February 3, 2014 at 8:42 pm

you should expand the sample before getting z-scores. 3-seasons, one before, one after.

that will let you get a good idea of the era.

Reply

Chase Stuart February 5, 2014 at 3:19 pm

I agree with you — and I get into a discussion of this in tomorrow’s post — but that takes a nontrivial amount of extra work. As always, there’s a tradeoff to be made between publishing something and publishing something perfect. I’m generally a good enougher, but I understand that’s not for everyone.

Reply

Nick Bradley February 5, 2014 at 3:31 pm

Thanks – another approach, which would be a trivial amount of time, is just do a regression and see how good above expected mean the team was. defenses have been getting worse over time — its pretty easy to do.

Reply

Chase Stuart February 5, 2014 at 3:32 pm

Doesn’t the Adj PPG column give you the info you need?

Reply

Nick Bradley February 5, 2014 at 4:08 pm

well no; trend is not the same as the average that year.

for example, last year lacked truly elite defenses.

so, just plot every team’s adjPPG over time, create a regression line, and see how much above the line (above expected) each team is.

Reply

Matt February 3, 2014 at 10:09 pm

5 “borderline” hall of famers on the ’02 Bucs???!!! Come on, Sapp and Brooks are 1st ballot HOFers, how is that borderline? Yes Rice and Barber may be borderline, but even Lynch will eventually get in. BTW it seems that the mainstream media when discussing great defenses of all time in a single year hardly ever mention the ’02 Bucs. It is always the ’85 Bears and the ’00 Ravens. Doesn’t feel like the ’02 Bucs get the respect they deserve.

Reply

Rod February 3, 2014 at 11:19 pm

Well Sapp, Brooks and probably Barber are 1st ballot and Lynch will get in. The fact that Strahan is in should mean that Rice’s name should come up at some point down the line. I think he was saying 5 “borderline” as in you have some sure fire HOFers and some that borderline, as in borderline or greater. At least that’s the way I read it.

Reply

Chase Stuart February 5, 2014 at 3:21 pm

Yes, that is the way it was intended.

Reply

Rod February 3, 2014 at 11:23 pm

I think as the HOF eventually includes Lynch and Barber (and perhaps Rice one day) I think people will finally start to realize how great that 2002 Bucs defense was, not only in yards against and points against, but also in terms of producing points. A very nice article, thanks for putting this together.

Reply

Chase Stuart February 5, 2014 at 3:22 pm

Thanks, Rod. Yes, I agree the 2002 Bucs are — for reasons I can’t quite articulate, but I suspect may have something to do with lacking a dominant MLB — generally underrated. It seems like the Butkus/Singletary/Lewis/Lambert types are almost a prerequisite to being in the discussion.

Reply

Andy Prough February 4, 2014 at 11:34 pm

Looks like Mike Tomlin and Dick Lebeau deserve a little credit – 3 teams in 4 years in the top 50 defenses all time? Wow!

Reply

Chase Stuart February 5, 2014 at 3:21 pm

Yeah, those Steelers defenses were outstanding, although I don’t think LeBeau or Tomlin are hurting for credit ;)

Reply

Neil February 5, 2014 at 2:09 am

You totally whiffed on Atlanta’s famed Grits Blitz defense of 1977. http://en.wikipedia.org/wiki/1977_Atlanta_Falcons_season

“The Falcons’ 129 points allowed not only led the league, but established an all-time NFL record for fewest points allowed in a 14-game NFL season….In fact, it was the stingiest defense since World War II. Atlanta surrendered just 9.2 PPG”

Come on, not even a mention???

Reply

Chase Stuart February 5, 2014 at 10:24 am

The Falcons are on the table, which is searchable. The issue for Atlanta is that 1977 was a low scoring environment and their SOS was weak. They were just 6.6 PPG above average, which is akin to the modern 49ers teams.

Reply

Tim February 5, 2014 at 10:22 am

It is statistically invalid to use standard deviation to judge members of different samples, especially with such small samples. This analysis has zero validity and is just garbage. To site a coupe of reasons why, suppose that one year there were two really good defenses; that would result in a vastly different “Z-score” from a year when when there was only one. In additional, if there were a really bad defense in a particular year, that would also change the “Z-score”. Neither of these factors is valid for comparing with a team from a different year.

Reply

Chase Stuart February 5, 2014 at 3:23 pm

You are free to instead use the Adj PPG column, which is why it was included, and referenced throughout the post.

Reply

Tim February 5, 2014 at 2:17 pm

Chase, just to give you an example of why this analysis is wrong. Let’s look at wins by division. The Colts had a std of 1.67 compared to the rest of their division. The Seahawks had a std of 1.42 compared to theirs. Can I conclude from this that the Colts were better than the Seahawks? No. (I can conclude that they dominated their division more than the Seahawks did theirs, but that is not the same as best.) But that is the exact logic that you are trying to use to “prove” that the Seahawks were one of the best defenses of all time. It’s nonsense.

Reply

Chase Stuart February 5, 2014 at 3:20 pm

Thanks for the comment, Tim.

Reply

Mike February 6, 2014 at 3:22 pm

It would be interesting to look at which players/coaches show up on the list most. Ray Lewis played on 6 different teams on the list, under two different head coaches and four different defensive coordinators. Terrell Suggs and Ed Reed were on 5 of those teams, though.

Reply

Leave a Comment

You can use these HTML tags and attributes: <strong> <em> <pre> <code> <a href="" title="">

{ 16 trackbacks }