On Monday night we saw history. Drew Brees eclipsed Dan Marino's record of 5,084 total passing yards on a late-game touchdown pass to Darren Sproles. Regular readers know that I'm no fan of most ever or least ever records because they're usually just trivia that end up giving the word 'stats' a bad name. Readers also know that total passing yards is not a particularly meaningful way of measuring a quarterback's skill. But it's hard to let the occasion pass without taking note.
The Monday Night Football crew did a good job of reminding viewers of the context of the record, even going as far as providing analysis showing how far above average both Marino's and Brees' were for their respective seasons. The NFL's passing numbers have steadily inflated over the years, largely due to rule changes that favor offenses. But like many other sports, it's possible that the athletes have simply improved over time. Defenders can improve too, but who's to say that athletic improvement on both sides of the ball doesn't disproportionately favor the offense. The fact is we simply have no way to tell.
We can use some statistical tools to get a feel for how outstanding each season was. Drawing the line at the top 30 passers in both seasons, we can calculate the number of standard deviations Marino and Brees stand above the season average. Marino's 1984 was 2.4 standard deviations above average, while Brees' 2011 (so far) is 1.9 standard deviations above average. Marino achieved his numbers on 564 attempts while Brees has 622 attempts, and counting. Brees has 13 interceptions compared to Marino's 17. According to PFR's Adjusted Net Yards Per Attempt, which factors in yards, attempts, interceptions, and touchdown passes, Marino beats Brees 8.9 to 8.0.
One of the unspoken assumptions when discussing Marino's 1984 record is that his record is a 'pure' or 'true' record, and the record set in 2011 is asterisked by the liberal passing rules of today's NFL. But do you know who's record he broke and when that was set? It was Dan Fouts in 1981. Before that the record belonged to...Fouts in 1980. And before that...Fouts in 1979. But prior that, the pro record was an AFL mark set by Joe Namath in 1967.
It's no coincidence that the record was set and reset in three consecutive years starting in 1979. The 1978 season was the first with modern passing rules. Most notably, contact with receivers was penalized beyond 5 yards and linemen were allowed to block with extended arms and open hands. That season was also the first lengthened from 14 to 16 regular season games. A record like 'total passing yards' is bound to be broken soon after those changes, and sure enough it was--repeatedly. The notion that the 1984 record was somehow pure or true, and the 2011 record is tainted due to rule changes is myopic in the extreme.
Marino was not a great quarterback because of his 1984 record--not at all, given that it was only 6 years after the tectonic changes of 1978. Marino was a great quarterback because he won games. But the fact that his record stood for 27 years despite more and more liberal passing is astonishing.
- Home
- run-pass balance
- Brees versus Marino, 2011 versus 1984
Brees versus Marino, 2011 versus 1984
By
Brian Burke
published on 12/28/2011
in
analysis,
player analysis,
run-pass balance
Subscribe to:
Post Comments (Atom)
>>Drawing the line at the top 30 passers in both seasons (because there were 30 teams in 1984)...
There were only 28 teams in 1984.
@ bytebodger
agreed - and lopping off the worst two (or four) quarterbacks for Brees and not for Marino would seem to unfairly affect Brees' std dev.
maybe it would be better to compare each to the top X qbs of each year (top ten maybe), or just use all of the quarterbacks playing at each time.
Have to concur with anon above. Should just use top 10 QBs or so, although I think Marino will still be far ahead.
Top 28 QBs, it's 2.5 for Marino vs 1.9 for Brees. Top 10 QBs, it's 2.2 for Marino vs 1.5 for Brees.
Standard deviation does account for number of samples...that's kinda the point.
Will, this isn't about sample size, it's about quality of the QBs included in the sample. With top 30 for Marino you get at least 2 split-time/backup QBs included.
I wonder what the standard deviation of Fouts' 1981 season was? He had nearly 900 more yards than the second place finisher.
"Marino was a great quarterback because he won games."
Wow. When did Brian turn to the dark side?
Had it not been for the strike, Fouts in 1982 might have set (or reset) the record a fourth straight time. In fact, his passing-yards-per-game record from that season--strike-shortened as it was--has stood for two years longer than did Marino's total yards record.
For total passing yards, you can just use team passing numbers for all of the other teams in the league (that way you don't have to worry about injuries and so on). By that measure, I get that Brees is 2.4 standard deviations above the average team in total passing yards vs. 2.7 for Marino.
Brees 2011: other 31 teams average 3612 pyds, with sd=616 yds.
Marino 1984: other 27 teams averaged 3596 pyds, with sd=551 yds.
According to the latest AYPA stats on this site, Brees is 4th in the league, at 6.7. Rodgers is 15% (!) higher, at 7.7. Manning had a 7.8 in 2004, which is the highest listed on this site. I wish the pundits on TV would put more emphasis on efficiency.
For those curious, the 30 players in 1984 that I believe Brian used to calculate Marino being 2.43 standard deviations above average:
Dan Marino, Mia
Neil Lomax, Stl
Phil Simms, NYG
Dan Fouts, SD
Dave Krieg, Sea
Joe Montana, SF
Steve DeBerg, TB
Paul McDonald, Cle
Joe Theismann, Was
Warren Moon, Hou
Tony Eason, NE
Lynn Dickey, GB
Gary Danielson, Det
Ron Jaworski, Phi
John Elway, Den
Gary Hogeboom, Dal
Richard Todd, NO
Steve Bartkowski, Atl
Marc Wilson, LAR
Mark Malone, Pit
Ken Anderson, Cin
Bill Kenney, KC
Jeff Kemp, LA
Joe Ferguson, Buf
Pat Ryan, NYJ
Todd Blackledge, KC
Tommy Kramer, Min
Danny White, Dal
Jim Plunkett, LAR
Mike Pagel, Ind
Baseball stat geeks have states like "OPS+" and "ERA+" that compare performances with the contemporary average. So Pedro Martinez having a 2.00 ERA 10 years ago would have a higher ERA+ number than a 2.00 ERA from the deadball era.
It would be interesting to see a similar stat for adjusted net yards per attempt.
By the same measure Cam Newton hasn't yet really matched Peyton's "most yards passing by a rookie QB" performance. Average passing yards per game have risen by near 12% just since 1998. Adjusting for that Cam is still about 250 behind. Not that he isn't having a heck of a rookie year.
All the passing numbers are steadily inflating as they make it more and more illegal to defense receivers. We're reaching the point where they might as well velcro flags to the recievers' hips and tell the defense to forget everything else.
Baseball stat geeks have states like "OPS+" and "ERA+" that compare performances with the contemporary average ...It would be interesting to see a similar stat for adjusted net yards per attempt.
NFL "+" stats are on the player pages at Pro-football-reference.com.
Matt W. -- The problem with using adjusted net yards per attempt is it relies on sack data, which is not available on older games.
Assuming I haven't fat-fingered something though, here's the average any/a data back to 1970
year | anypa
------+--------------------
1970 | 4.1592039800995025
1971 | 3.9222454010999431
1972 | 4.2391175306687772
1973 | 3.8696736852609520
1974 | 3.9333518724508713
1975 | 4.0571453885000443
1976 | 4.0728126333760137
1977 | 3.5559644556143973
1978 | 4.0332389327768669
1979 | 4.6154479917610711
1980 | 4.8387390907906734
1981 | 5.0191006261463538
1982 | 4.8082319925163704
1983 | 4.9877841445752786
1984 | 5.0216349108789183
1985 | 4.8316777041942605
1986 | 4.9623510625230316
1987 | 5.0651147540983607
1988 | 5.0003796987723073
1989 | 5.2773573200992556
1990 | 5.3104598678615817
1991 | 5.1553404241768040
1992 | 4.8838377250195669
1993 | 5.1555909759686121
1994 | 5.4072800569124970
1995 | 5.3915678807058950
1996 | 5.0838854576080853
1997 | 5.1426560036044154
1998 | 5.3014324031273184
1999 | 5.1692021276595745
2000 | 5.1621237777899164
2001 | 5.1675297487880123
2002 | 5.3558288548987185
2003 | 5.2307231865919356
2004 | 5.6641021443771485
2005 | 5.3433639875931871
2006 | 5.3609960767218832
2007 | 5.5234296581964182
2008 | 5.6927645199956413
2009 | 5.6828701257529325
2010 | 5.7426665971968947
(41 rows)
The Monday Night Football crew did a good job of reminding viewers of the context of the record, even going as far as providing analysis showing how far above average both Marino's and Brees' were for their respective seasons. The NFL's passing numbers have steadily inflated over the years, largely due to rule changes that favor offenses...
But do you know who's record [Marino] broke ...? It was Dan Fouts in 1981 ... Fouts in 1980 ... Fouts in 1979 ... Joe Namath in 1967.
For the record, Namath's 4007 yards was 58.3% above the pro team average of 2531. That much above average in 2010 would have been 5612 (vs the new record of 5087) so we can see one of the things about him that grabbed people's attention back in the day.
By standard deviation he was 2.19 ahead of the other 24 QBs with the most yards (there were 25 teams then). That much ahead of the 32 QBs with the most yardage right now would be 5327, with one game still to go.
Jim Glass,
that might be the first time ever i can´t agree with you:
Quote "All the passing numbers are steadily inflating as they make it more and more illegal to defense receivers. We're reaching the point where they might as well velcro flags to the recievers' hips and tell the defense to forget everything else."
I know many people miss the violence of the 70s and find all the new rules disturbing. But you know what? I think Walter Camp hadn´t in mind that players who survived 4 yrs of pro football should have a 2/3 chance to suffer from permanent pain after the glory days.
So i think there is nothing wrong with "inflating" passing numbers caused by rule changes. Look at soccer & hockey: they turned the opposite. Fewest goals ever are scored in the history of this once exciting games. They are more dull nowadays than watching chess :-)
The youth will get used to scores like 44-38, and won´t have any problems with it.
BTW, there is still much hard hitting and too many crippled players in football...
Cheers, Karl, Germany
Good thread and comments. Want to echo mike's comment above:
I wish the pundits on TV would put more emphasis on efficiency.
Also have to laugh at the people claiming Brees' should win MVP simply because he broke the record... I mean, if he came up 10 yards short you'd be fine with giving it to Rodgers but if he broke the record by 10 yards its "He should get it automatically". Not following that line of reasoning.
You also have to factor in the dome vs outdoors factor somehow, where Marino would be even further ahead. Now don't get me wrong it wasn't like Marino was playing in Buffalo every game, but playing in the AFC East stadiums vs the NFC South domes and stadiums is two different stories.
Brees played 11 out of 16 games in a nice climate controlled perfect environment(and 0 of the 5 outdoor games in bad weather by my count).
Marino played in 1 game indoors, two if you count Cowboys Stadium.
Jim Glass, that might be the first time ever i can´t agree with you
It won't be the last, I'm sure. Even the greatest minds can't think alike all the time. :-)
I know many people miss the violence of the 70s and find all the new rules disturbing. But you know what? I think Walter Camp hadn´t in mind that players who survived 4 yrs of pro football should have a 2/3 chance to suffer from permanent pain
The rule changes of '78 had nothing to do with safety, they were about opening up the passing game to get more scoring. (Though they achieved much more of the former than latter.) Same with the bulk of the passing rule changes that have kept coming ever since.
Banning helmet-to-helmet hits and horse-collar tackling of receivers is about safety. Ever-more-tightly penalizing the slightest contact between a defender and a receiver five yards past the line of scrimmage is about letting recievers roam free and untouched to enable QBs to pile up ever bigger numbers.
The youth will get used to scores like 44-38, and won´t have any problems with it.
Yet scoring hasn't increased anything like passing. In 1965 -- as many would think, the stone age era of everyone following Lombardi running the ball -- the average score was 23.1 points per game. That was with a league average 51.3% pass completion rate, 184 yards passing per game, and a near 1 to 1 TD-pick ratio.
This year scoring is *less* by a point, 22.1 per game, with an average 60% completion rate, 229 passing yards per game, 1.4 to 1 TD-pick ratio, and passes increased to 55% of all plays from 47%.
What's happened with all the rule changes is coaches have replaced former running with newly safe high-percentage short-passing, incidentally causing average yards per completion to *plunge* from former times. (Which I don't find particularly exciting. IMHO that makes offenses much more one-dimensional and less interesting, high percentage chuck-chuck-chuck, with the valuable capable running games that once complemented passing and set up the big passes disappearing -- but that's just me.)
So there is no sign of today's youth seeing future typical scores like 44-38. But your larger point is well taken, they may greatly enjoy the future heros of 6,000-yard passing years, especially after the 18-game season arrives. That's clearly the NFL's calculation: fans love passing and passers.
OK, I know we're starting to veer a bit off-topic here, but I would like to point out that running does NOT equal violence and passing does NOT equal finesse. In fact, if I had to pick a side in that argument, I would say that high-scoring (i.e., big-passing) offenses lead to MORE violent plays - not less.
When you open up the field, you allow players to accelerate toward each other at full speed. Sure, there is a lot of "contact" in your average run, but most of those guys in the pile never even managed to reach half speed before the collision.
I used to play offensive and defensive tackle. Even though I was involved in contact on every single play, I rarely felt at any risk of injury because we only had a few feet to gather momentum before we crashed into each other. Now when a safety can level a WR who is running at full speed across the middle of the field, THAT is violent.
Re: ANY/A -- One could just use regular adjusted yards per attempt instead of adjusted net yards per attempt to account for the lack of old data on sacks.
OK, if we take the leaders for each season from 1970 to 2010 in ANY/A and compare their number to the league average, here are the top 5 (100=average)
1. Staubach, 1971 (199)
2. B. Jones, 1976 (191)
3. Brodie, 1970 (181)
4. Stabler, 1974 (179)
5. Marino, 1984 (178)
If we include only the seasons since sacks became official (1982), here are the top 5:
1. Marino, 1984 (178)
2. Manning, 2004 (173)
3. Young, 1992 (166)
4. Rypien, 1991 (162)
5. Cunningham, 1998 (161)
How is Rodgers doing? When I calculate the league ANY/A, I get 5.47. That would give Rodgers an ANY/A+ figure of 172, which would rank third since 1982 and 7th since 1970. However, PFR lists the league ANY/A as 5.9. That would give Rodgers an ANY/A+ of 159, or 16th since 1970.
Drew Brees has an ANY/A of 8, which, using the lower figure above, results in an ANY/A+ of 147.
Regarding all the reliance on AY/A and variations thereof for comparisons, remember that the value of *that too* to winning has changed over the years.
AY/A is a compound stat, completion % times average yards per completion. AY/A assumes that they are equally valuable, and using it to make comparisons over time assumes that their relative value has been constant. Both assumptions are wrong.
Average yards per completion has a stronger correlation with scoring and winning. In today's game the correlation is only modestly stronger than that of completion %, so AY/A is credible enough and commonly cited as an easy stat. But back in the 1970s the correlation for Y/C was twice as strong. So two QBs with the same AY/A could be playing at quite different levels, and the QB with lower AY/A could be the better one.
Since QBs and coaches back then weren't playing to maximize AY/A as they are today, but were playing to maximize something else, judging them by their AY/A numbers, either absolute or relative, can be misleading.
A while back I put more data about this on the Community site.