May the least-bad team win

This season has become rather non-bueno, with the Saints sitting at 2–4 having already had their bye week. What’s scary: they’ve played the 31st-easiest schedule in the league through Week 6, according to Football Outsiders. Going forward, they have the 19th-hardest remaining schedule, so things are likely to get worse if the team doesn’t improve.

So it’s time to start looking for things to be hopeful about. The brilliant Mills Baker outlined a few reasons why the Lions loss maybe wasn’t so bad, but I’m more interested in the macro-view. At least as macro as it gets 38% of the way through a 16-game season. I’m mainly interested in two questions:

  1. Can the Saints still win the division?
  2. Are there any signs that might indicate a turnaround?

I don’t have time for a deep analysis of these questions, but I poked around a bit and here’s what I found.

The relative awfulness of the NFC South

Let’s start with the terrible division. The 2010 Seahawks famously won the NFC West with a 7–9 record, though I can’t recall how they did in the playoffs that year. There’s a reasonable chance that the NFC South winner will have a 0.500 record or worse, though Carolina’s recent tie with the Bengals might screw with that. The tie concerns me for other reasons: recent ties have been just as good as wins. The 2012 49ers, for example, finished 11-4-1, taking the division over a 11-5 Seahawks team. In 2013, the Packers finished 8-7-1 and won the NFC North over the 8-8 Bears. It’s possible that the Panthers’ tie will seal the division for them.

Ties aside, the NFC South is really crappy this year. How crappy? Well, let’s start by comparing the 2014 NFC South to the aforementioned 2010 NFC West. I’ll use two methods: Simple Rating System and DVOA.

Simple Rating System

The Simple Rating System (SRS) is a quick-and-dirty method of calculating a team’s ranking that I first saw at Pro Football Reference. Essentially, a team’s SRS is their average margin of victory (or defeat) plus their opponents’ average SRS rating. For more details, see this post at Pro Football Reference. The SRS is pretty easy to understand: average teams have an SRS of 0, above-average teams have an SRS greater than 0, and below-average teams have an SRS less than 0.

The average SRS of the NFC South in 2014 is -9.075. The average SRS of the 2010 NFC West was -8.65. In other words, by SRS, the 2014 NFC South is worse than the 2010 NFC West. Yikes.

DVOA

Defense-Adjusted Value Over Average (DVOA) is a more nuanced way to rate teams. It’s a proprietary measure developed and updated by Football Outsiders that scores each play based on number of yards gained. They then make all sorts of adjustments for yards needed to get a first down (a 3-yard run on 1st and 10 is more successful than a 3-yard run on 3rd and 5, for example), field position, opponent quality, etc. You can read a more in-depth explanation at the Football Outsiders website, but the long and short of it is an average DVOA is 0, positive DVOAs are good, and negative DVOAs are bad.

The average DVOA of the NFC South in 2014 is -13.7. The average DVOA of the 2010 NFC West was -22.65, which is quite a bit worse than the NFC South. Still, by DVOA, the NFC South is well below average, and the worst in the NFL.

All of this is to say that the NFC South is quite winnable, assuming the Saints can stop tripping over their shoelaces. But that takes us to the second part of the analysis:

Are there signs that the Saints might turn it around?

You can roughly estimate a team’s quality by comparing the number of points that they’ve scored to the number of points they’ve allowed, using something that’s called the Pythagorean Wins formula. I hate that name for a few reasons, but whatev. Grant me the serenity to accept the things I cannot change, etc.

The basic formula is this:

Expected winning percentage = points scored^2.37/(points scored^2.37 + points allowed^2.37)

There has been lots of debate about the right exponent to use in place of 2.37, but the differences with different exponents aren’t that big. To bastardize Carveth Read, it’s better to be roughly right than exactly wrong, so I’ll stick with the 2.37.

Anyway, one interpretation of Pythagorean Wins is that teams that have won more than their Pythagorean expectation have been a bit lucky and teams that have won less than their expectation have been a bit unlucky. So I did the math and made a graph. The grey dots represent a team’s actual wins. The orange dots represent a team’s expected wins according to the Pythagorean Wins formula. With the important caveat that 6 or 7 games is a vanishingly small sample, if the orange dot is to the right of the grey dot, the team has been unlucky so far. If the orange dot is to the left of the grey dot, the team has been lucky so far. Check it out:

pythagWins

So the Panthers have outperformed their expectations by a bit and the rest of the division has underperformed by a bit. This bodes well for the Saints, again with the small sample caveat.

It bodes even better for the Saints when you look at winning percentage, not just total wins, since Atlanta has played an extra game:

pythagPercent

All of this is to say that, if the Saints can get a little bit lucky over the rest of the season, then they have a chance.

But there’s more! The Saints are the best team in the NFC South, according to DVOA, SRS, Nate Silver’s fancy new Elo thingy, and Advanced Football Analytics’ team efficiency ratings. They’ve created very few turnovers so far. Given that turnovers are largely random (which is why I disagreed with Gregg Williams’ defensive philosophy), the Saints should create more turnovers going forward. Of course, let’s not succumb to the Gambler’s Fallacy: they might not, too. Or, in the words of someone who knows: coulda, woulda, shoulda.

So: the Saints are probably the best team in the NFC South, but they still aren’t very good, and they have to make up a couple of games on Carolina to overcome the Panthers’ as-good-as-a-win tie. But the division is so bad that an 8-8 or 7-9 division champion wouldn’t be surprising. What does all of this mean? I don’t know.

The [Football Outsiders playoff odds][30] say that the Saints have about a 31% chance of winning the NFC South. That sounds about right to me. But, after doing this analysis, I can see a path from here to a division championship, and then (absurdly) a home game in the first round in the playoffs. In all, that’d be pretty sweet.

[30]: http://www.whodatreport.com/2014/10/division-odds-week-7/