≡ Menu

Additional Thoughts On The 2016 Falcons Offense

Atlanta had “only” 175 drives on offense this year, tied with Washington for the fifth fewest drives in the NFL. That’s partially because the offense was so good — Atlanta had the 2nd fewest number of drives end in a 3-and-out, behind only New Orleans — but also because the defense was below average, keeping the offense off the field.

Despite that, Atlanta scored a whopping 540 points. But here’s another way to think about it: the Falcons had just 11 turnovers and 48 punts, meaning just 59 drives ended in a punt or a turnover. That’s the fewest in modern history, one fewer than the 2007 Patriots (who had just 167 drives). And Atlanta scored 58 offensive touchdowns, meaning the offense had nearly as many drives end in a touchdown as a punt or turnover. And that’s just, well, crazy:

RkTeamPass TDRush TDOff TDPuntFum LostIntTOs + PuntOff TD:(TO+P)

How does that compare historically? I looked at all offenses since realignment in 2002. Just 19 of them had even two-thirds as many offensive drives end in touchdowns as they did in punts:

RkYearTeamPass TDRush TDOff TDPuntFum LostIntTOs + PuntOff TD/(TO+P)

The Falcons also scored a bunch on field goals. Over half of all Atlanta drives ended in scores — 52.6%, to be exact — making them just the third team since 2002 to reach that mark. The 2007 Patriots were at 52.7%, and the 2011 Saints at 51.4%. The 2016 season was tilted towards offense more than the ’07 or ’11 seasons were, but there’s no question that this is a historically great offense. Atlanta’s offense ranked 1st in points, turnovers, and net yards per attempt, 2nd in yards, passing touchdowns, and first downs, 3rd in passing yards and rushing touchdowns, and 5th in rushing yards and yards per attempt.

  • Arin Franz

    For the most fleeting of moments, I was comforted that the 49ers actually managed 36 offensive touchdowns in 2016. Then my skepticism kicked in, so I looked at their splits by game win probability on PFR. 47% (16/36) of the touchdowns came when the Niners had a win probability less than 20%.

    Their offensive stats are as hollow as my soul after years of Jed York ownership…

    • Do you know what percentage of 49ers offensive drives came with a win probability of less than 20%? I’m sure it’s lower than 47%, but I am also sure it’s higher than the average team.

      • Arin Franz

        I’d need to do some data collection to get the drive data, but the table below lists the number of plays. 57% (569/998) of 49ers plays were run with a WP < 20%. There was a lot of Red Zone watching this season. A lot.

  • Adam

    This is a neat way to measure offense, and really puts the Falcons’ dominance in perspective.

    I would suggest removing fumbles lost from the denominator and replacing it with total fumbles / 2. The randomness of fumble luck can have a big impact on the numbers without being a reflection of the team’s skill.

    • Richie

      This feels like more of a descriptive chart than a predictive one to me. In that case, I would think actual fumbles lost is more relevant than expected fumbles lost. If the Falcons had been especially lucky at recovering their own fumbles this year, it would help explain why their offense was so good, and reflected in this chart.

      • Tom

        I was thinking the same thing…according to PFR, the Falcons have recovered 10 fumbles, which is one more than the average, 9. So nothing out of the ordinary there…

        • Richie

          That must include defense.

          I have their offense at 8 fumbles (fewest in the league), with 4 lost fumbles (second to Pittsburgh).

          The outlier teams are Carolina (lost 10 of 13 fumbles) 77%. Chargers (lost 14 of 22) 64%.

          On the other side, Steelers lost 3 of 16 (19%); Eagles lost 6 of 24 (25%).

          The 49ers had the most fumbles (29) and lost fumbles (15).

          League average is 20.4 fumbles, 8.9 lost fumbles (44%).

          • Tom

            Right. I was thinking that recovering *any* fumble might help the offense’s numbers due to better field position, etc.

  • Richie

    The Falcons didn’t “feel” like a great offense to me. Am I alone in that assessment?

    I’m not sure if that has to do with the way their offense works, or because their W-L record was comparatively disappointing, or because their RB’s were splitting the workload. Or, just because Atlanta wasn’t in many prime time games.

    The 07 Patriots, 98 Vikings and 99 Rams felt like unstoppable offenses to me at the time.

    • Tom

      Yeah, the ’07 Pats and ’98 Vikings are in their own league, but actually the ’99 Rams are pretty far down the list when you remove the special teams/defensive scores, #93. Not saying they’re not good or that we shouldn’t count those scores – is a punt return TD an “offensive score”? – just adding some context to the discussion. The ’99 Rams had 79 of those types of scores, which is the third highest since the merger.

      • Adam

        What was the 99 Rams’ offensive SoS? Has to be among the weakest ever.

        • Tom

          It actually *was* the weakest ever (since 1970 anyway), -3.0! They averaged 8.9 points per game over league average, -3.0 SOS, leaves them with 5.9 Off SRS. The difference between OSRS (all scores) and Off SRS can usually be ignored because the average team puts up about 2 “non-offense” scores per game, so it’s not that big of a deal when we compare OSRS. But there are some instances, like this one, where I think we do need to look at those scores if we’re trying to use points to rank offenses (and defenses).

          • Adam

            Thanks. When you combine their historically weak SoS, the 11 non-offensive touchdowns, and playing in a dome, the 1999 Rams have to be the most overrated offense since the merger, right?

            • Tom

              That’s kind of what I’m thinking…but then again, they were #1 in total yards, #1 in yards per play, #1 in ANY/A, #2 in yards per carry, #1 in points per drive, etc. So it’s kind of weird…almost like maybe they weren’t overrated, but should have scored more than they actually did? I sense that I’m about to waste more time on this…

              • Richie

                Watching them, they definitely seemed unstoppable. Of course, we didn’t know they had a weak schedule at the time (or even know as much about SOS).

                Then the next year they had a better OSRS while playing a stronger schedule.

                • Tom

                  I agree…as I noted above, they were ranked #1 in a bunch of offensive categories…DVOA has them at #4 for the year. I’m thinking they were awesome but I guess not “526 points” awesome…

                  • sacramento gold miners

                    Have seen very little of the Falcon’s strong offense this season, but did the ’99 Rams ever throw away a game which meant something like Atlanta did versus Kansas City earlier this season? The best regular season win for the ’16 Falcons appears to be the home victory over Green Bay, but at this point, the ’99 Rams road win at the Titans seems more impressive.

                    The ’99 Rams attack seemed to be more aggressive than other strong offenses of recent years, that’s my impression.

                    • The 1999 Rams lost the game in Tennessee. Warner fumbled on two consecutive drives, directly leading to two Steve McNair touchdowns. Marshall Faulk lost another fumble, and Jeff Wilkins missed a short field goal that would have sent the game into overtime. That’s essentially four turnovers (three by the offense) that cost them a close game against a playoff team from another conference. In that regard, it is quite similar to the Falcons against the Chiefs.

                      Of course, the Rams later avenged the loss in the Super Bowl, which I doubt will happen for Atlanta.

                    • Richie

                      Every team makes mistakes. Of course, it was impossible to lose in the way that the Falcons did until 2 years ago when the returned PAT rule was implemented.

              • Adam

                Hmm, I think it’s fair to say that all those return TD’s took away opportunities for the offense to score (probably 20-30 points).

                Also, I have a theory that SoS has less effect on units at the extreme ends of the performance spectrum. If an offense is vastly superior to every defense in the league, it doesn’t much matter how good one defense is relative to another. Silly analogy…If a meteor crashes in your town, it doesn’t matter whether your shelter is built out of tree branches or concrete – the meteor wins regardless.

                • Tom

                  Adam – I agree with both of your thoughts. That’s what leads me to think that perhaps points per drive – that sweet spot between points per play (if anyone even uses that) and points per game – might be the preferred (simple) metric for evaluating offenses and defenses. Because you’re right…those return TD’s do take away possible drives…so we can look at the Rams points per drive at 2.37 and really see how good they were.

                  And yes, I agree with the meteor analogy. It doesn’t really matter that the Texans’ house is made of reinforced concrete and that the previous homes the Pats have crushed have been made of tree branches…that Patriots meteor is going to crush the Texans’ house.

                  • Adam

                    Here are the top offenses in Adjusted Pts / Drive over the past 20 years. The 1999 Rams rank ninth, but behind the other two GSOT seasons. This year’s Falcons team is “only” 17th, mostly from being dragged down by era adjustment (2016 yielded the highest PPD in history).

                    I can’t figure out how to copy excel tables into these comments, so I just copied a tweet instead 😉

                    • Tom

                      Pretty cool…so this makes sense…the ’99 GSOT Rams are much better than my #93 ranking suggests. On a points per game basis they don’t rate so hot, but perhaps it’s because they were being “robbed” of possible scoring drives. And yeah, I’ve got the other two versions of the GSOT being better as well, in the same order as you.

                    • Adam

                      In 1999 the average offense had 191 drives, but the Rams only had 182. That pretty much explains the difference in ranking between our methods. Although keep in mind that my data only goes back to 1997, so if we could go back to 1970, the `99 Rams would probably fall into the 20’s.

                    • Tom

                      Adam – just to wrap up this discussion (or continue it) I looked up a post Chase did in June of 2015 that ranks offenses by “estimated” points per drive. The ’99 Rams rank #39; Bryan Frye then posted an era-adjusted list in the comments in which the Rams are at #30. So…are they overrated? If so, I wouldn’t say “grossly” overrated…maybe slightly…


                    • Adam

                      Ah yes, one of my favorite posts. You’re right, #30 all time is only slightly overrated in an analytical sense. Calling them the most overrated offense ever was certainly an exaggeration. That said, I still believe the `99 Rams are highly overrated from a perception standpoint; many people place them among the handful of greatest offenses of all time, which they are not. And the vast majority seem to think the `99 offense was the best of the GSOT teams, when in actually it was the worst.

    • Adam

      I think it’s the lack of coverage. Hell, the national media didn’t even acknowledge their QB as an MVP candidate until week 17. This year it’s been all about the Cowboys resurgence and Brady’s revenge tour.

      • Tom

        I also feel like there’s a sense of, “Yes, we’ve seen this before, we know Atlanta can put up big numbers and that Matt Ryan and Julio Jones are good, but it doesn’t matter because they fail in the playoffs”, etc. I’m guilty of this as well…

        I mean, who talks about New Orleans? Their offense was good as well…7.8 Off SRS, which is good for #38. Not too shabby…

        • Adam

          I feel like people write off the Saints offense to stat padding, as if the numbers Brees puts up are hollow. Maybe a grain of truth to that, but NO has been one of the best offenses for a decade now.

    • eag97a

      The W-L record IMO. It doesn’t scream offensive juggernaut. Most memorable top offenses had very good w-l records besides the offensive numbers they put up.

    • Anders

      I still cant fathom that the Eagles held them to only 15 points (and the Steelers to 3)

  • Tom

    I agree, they are historic. As some (or none) of you know, I run SRS numbers for offense and defense after removing special teams and defensive scores. Below is a screen shot of the top 25 offenses in “Off SRS”. Sure, they’re still a ways off from the ’07 Pats, but being in same neighborhood as the ’84 Dolphins says something. And if we rank by Z-Score, they move up to #21. https://uploads.disquscdn.com/images/ca045622f421dee9eb79e1fa84b5f888656ff9756bedeb3bf4a565f9140e99ad.png

    • eag97a

      Man, only the 98 Broncos won the SB. Talk about defense more important in winning the big one.

      • Tom

        …and the ’91 Redskins. But yeah, there’s six Super Bowl “losers” in that group of the Top 25 offenses…on the one hand it shows you that having a great offense can get you into the playoffs, and maybe the Super Bowl, but it doesn’t guarantee a win…

        • eag97a

          Yup overlooked the Redskins IMO one of the greatest if not the greatest team of all time.

          • Anders

            Pretty sure 91 Redskins is still the only team that Football outsiders thought should have went 16-0 during the season.

        • Johhny Ohrl

          … as a great Defense does not guarantee it either (sometimes not even the play-offs itself; see PHI 1991 for example).
          Looking at points scored is miss-leading as we all should know by now 😉

          Passing efficiency is the name of the game. As it always was since the 1940s… Unless you have a playoff choker at the helm (absurdly overpaid and overhailed PM himself as the prime example here), then even fluke teams made their path to winning the SB.

          Anyway, I had a quick look at Y/PP stats (the prime stat for descriptive and predictive Football stats) of all SB winners (since the 1970 merger):


          6 finished 1st in Y/PP, a total of 19 SB winners finished Top-3 (eight finished 2nd, five were 3rd), a total of 33 finished Top-10 (= 72%).


          8 finished 1st, a total of 19 finished Top-3 (four finished 2nd, seven finished 3rd), 33 finished Top-10 (= 72%). Pretty much balanced so far (even on such a small sample size).

          On average, SB winning offenses finished 7.6 in Y/PP, while defenses finished 6.6. “Advantage” defense. But only because PM & many fluke teams teams screwed the numbers from the absurd 2000s onwards*. Before that, the offense had the (slight) advantage. Those were the good old days…

          Btw: DVOA numbers confirms it, respectively have the offenses at advantage (from 1989 on, when this stats were made available).

          And I bet if somebody looks at standard deviations in yearly Y/PP stats, the offenses will trump defenses, because there were many truly outstandingly efficient passing offenses that won SBs (Rams, 49ers, Saints, Broncos, Cowboys, Redskins… and more).

          (* All six shitty SB offenses that finished 20th or worse in Y/PP, and thus screwed the numbers, came from 2000 onwards. But only three SB defenses finished worse than 20th, only two from 2000 onwards).

          Fun fact: When the pundits said “defense wins championships”, it was not true. And since they started to call the NFL a “QB/Passing league”, the pendulum swung towards the defenses (or better say luck, thanks to NFL-parity BS). They always get it wrong. Easteregg as prime example must be named.

          After all is said and done: My boycott of the NFL worked pretty well. I did not miss a thing since the Denver headbangers went un-punished and therefore wrongly were awarded the SB last year. In the meantime, I witnessed the greatest AFL season ever.

          By the way: I plan to watch the SB. But only if the Falcons make it there. They can correct the numbers back to normal (at least a little). Seems like they play beautiful football this year.

          Greetings from Germany 🙂

      • Johhny Ohrl

        Not necessarily (see my reply to Tom, just below)…

  • John

    The GSOT was always overrated. Too many turnovers

    • Adam

      Yep, those turnovers hurt the GSOT so much that they went 37-11 with two SB appearances.