Possibly the closest Super Bowl matchup in history also poses the question: how much does it mean when certain aspects of an NFL team improve dramatically in the second half of the season?
27 Nov 2013
by Tom Gower and Mike Kurtz
Tom: Say, Mike. Have you noticed this year's AFC South is one of the worst divisions in DVOA history? The four teams currently have an aggregate DVOA of -85.1%.
Mike: It is amazing how bad the division is. I'd have to look at the numbers more closely to call it one of the worst in history, however.
Tom: As far as I can tell, the 2004 NFC West is the worst at -95.0%, followed by the 2010 NFCW West at -90.6% and the 2008 NFC West at -88.8%.
Mike: ... how did you get those numbers so fast?
Tom: This is a topic we have discussed on the site before and I started poking around with the AFC South data. Take a look, for instance, at this column, though note back in 2007 we were on a different version of DVOA, so if you go and re-check, say, the 1998 NFC North, you will find different numbers. See also Doug Drinen's look at the same issue using P-F-R’s Simple Rating System.
Mike: This feels like an ambush topic. What is interesting is that a lot of these divisions have one decent team. The 1999 NFC West even has the best team in the league according to SRS. What is also interesting is that both of these columns were written by a person named Doug. Is there perhaps something intrinsic to Doug-ness that instills a desire to order things?
Tom: One of those divisions Mr. Farrar wrote about back in 2007 was the 2005 NFC West, which had the very good Seahawks and an aggregate divisional DVOA of -64.6%.
Mike: Everyone else was just horrendous.
Tom: Yes, the AFC South is a bit unusual in having two teams that are not particularly bad. The Jaguars are just so bad they offset that.
Mike: So bad that through 11 games, they are the fourth-worst team of the DVOA era.
Tom: Even those terrible Lions teams weren't enough to drag down the NFC North to anywhere close to the levels we're seeing from the AFC South.
Mike: I'm assuming that AFC South this year is fourth on your list.
Tom: Yes, at least thus far. Back in 2007, I also ranked divisions by Pythagorean wins, though I would want to re-do that project using 2.54 as the post-1978 Pythagorean exponent. Perhaps an offseason project, as well as confronting the issue that going back before 2002, of course, you run into divisions of non-uniform sizes. How should we think of a five-team AFC East compared to the four, five, or six-team AFC Central?
Mike: Apparently ranking things is a Tom thing, also. Maybe a better way of judging these divisions would be average DVOA rather than aggregate. Although it is worth noting that the four worst divisions by DVOA are all in the four-team-per-division era.
Tom: Maybe there's something to the schedule that produces that. It is a curious phenomenon. Even divisions with a couple really bad teams, like the 1991 AFC East, end up with a non-historically bad -65.8% aggregate DVOA (-13.2% average per team).
Mike: I think part of that is that the Bills were so good that year. Going back to SRS, that same 1991 AFC East ranks out as the fifth-worst division of all time by average of the teams, whereas by average of the teams, it only clocks in at -13.16% DVOA.
Tom: Maybe it's just that choosing four teams, and also eight divisions per year, makes it more likely to not have one of the good teams. Then again, while that might be true for random lotteries, is NFL team quality similarly distributed?
Mike: I don't think you can say quality NFL teams are randomly distributed.
Tom: I don't think they are either, but what's the best way to prove that? This ties back into the conference strength argument. Is there some reason in particular the AFC has been better head-to-head most years, or that the NFC had that stretch of domination from 1989 to 1995?
Mike: Theoretically, if the league were randomly distributed then each division would be perfectly average over an infinite sample.
Tom: Over an infinite sample. That the NFC West was pretty lousy for about seven years in a row is not seven years of randomness.
Mike: We could sum up these SRS averages. That's a pretty big sample, although realignment is a problem.
Tom: We should note the 2013 AFC South is pretty bad by SRS, with a per-team average of -5.2, but is far from the worst division, even of the DVOA era. SRS also looks more favorably on the 2010 NFC West, which I think would rate as the consensus worst division in recent years if only because the Seahawks won it going 7-9.
Mike: Perhaps our statistics aren't really at the point where we can answer the questions we've been asking.
Tom: What?! How dare you suggest that we don't know and see all!?
Mike: It's merely odd to see such jarring differences between two well-respected statistics. Obviously, we prefer DVOA.
Tom: I think it's a more sophisticated and complicated measure that is probably somewhat more accurate. Of course, they don't dispute the basic conclusion -- that the 2013 AFC South is really, really, fantastically bad.
Mike: The interesting thing is that even with Jacksonville's near-historic awfulness, the division as a whole is really dragged down because Houston has also fallen in the gutter.
Tom: Right. It was clear to about everybody that Jacksonville could be very bad this year, while Tennessee and Indianapolis have performed fairly close to reasonable expectations.
Mike: The interesting thing about all those awful NFCW divisions is that they also seem to roughly follow the 'two awful teams' pattern
Tom: The key to the NFC West's awfulness seems to have been not having a second team be even mediocre.
Mike: I suppose that is a fairly uncontroversial statement: when half of your division is extremely bad, your division has an increased chance of being historically bad!
Tom: Also, don't forget that there's a whole second season in the NFL, and it has nothing to do with the first season. Aside from 2003, an NFC West team has won a playoff game every year since realignment. So if the Colts do indeed win the division, don't count them out just because they were an average team playing in a woeful division.
Mike: Remember that even Tim Tebow won a playoff game!
Quarterback: After his 1 this week, Geno Smith has now finished fourth, first, and first among quarterbacks in his three non-bye weeks the second half of the Loser League season.
Running Back: Facing the Redskins was not the cure for anything that ailed the San Francisco run game, as Kendall Hunter was the low back this week with 12 yards and 1 point, while Frank Gore was part of the runner-up crowd with 3 points. Joining Gore were Bernard Pierce and Ray Rice, Chris Ivory, Mark Ingram, and Lamar Miller. Condolences to anyone who picked Ben Tate, who finished with seven carries and drew the penalty for 17 points instead of 2.
Wide Receiver: It seems to be teammate week in Loser League, as Miller's comrade Rishard Matthews was the low wide receiver this week with 0 points. Just a point ahead of him were Danny Amendola, Ted Ginn, Greg Little, and Brian Quick.
Kicker: Billy Cundiff made a field goal and missed a field goal to finish with 1 point and was not involved when the Browns decided to go for the two point conversion after their only touchdown.
Check out the Loser League results page for full scores.
KEEP CHOPPING WOOD: Attention Bruce Carter and Jeff Heath, defensive coordinator Monte Kiffin may have been a collegiate coach last season, but this here is the NFL, and in the NFL a player is not automatically down because they go to the ground. They must be touched down. You may give thanks to Tony Romo for minimizing your goat status by winning the game despite your incompetence.
MIKE MARTZ AWARD: Bill Belichick demonstrated a recognition of the idea that the current, unnecessarily modified overtime format may sometimes require rethinking overtime strategy. Mike McCarthy, by kicking a field goal on the opening possession of overtime on fourth-and-goal from the Vikings 2-yard-line, did not.
Tom: As a reminder, all lines are courtesy of Bovada, and were accurate as of time of writing. Last week, you took the New Orleans Saints, who failed to cover against the Falcons. I was tempted by the Buccaneers, who did indeed not only cover but won at Detroit, before ultimately taking the Cowboys, who also not only covered but won on the road.
Mike: I immediately seek out the Jaguars, because they are so very bad. Unfortunately, they are playing the Browns, who are also very bad. I can't trust Cleveland to do anything, especially win by more than a touchdown, so that is out. In fact, there aren't a lot of great options this week. My next thought is Bears +1 at Vikings, but the Bears' running defense is a horror show. The main concern is that a functioning passing offense will destroy Minnesota, which even without Cutler, Chicago still has. I'm guessing Chicago sells out to stop Peterson and can eke out the win. Chicago Bears +1 at Minnesota Vikings.
Tom: Darn, that was the game I was going to pick. The raw numbers I normally use to guide my pick selections suggest that perhaps the Patriots may be favored by a couple points more than they should be, while my mental model was looking toward New England. I'll instead go out on a limb and indicate that, while you're almost never as bad as your last loss indicates you are, the Redskins are dysfunctional enough on offense and bad enough on defense that I think even the Giants will have success against them. New York Giants -1 at Washington Redskins.
Apparently we have already turned all of your fantasy teams into winners! For those not so fortunate, send questions into scramble-at-footballoutsiders.com
24 comments, Last at 21 Jan 2014, 10:01pm by Cheap Pandora Necklaces