With No Runaway Favorites, The NFL Playoffs Should Be Wild

With the NFL’s playoff bracket finally set, it’s time to survey the field and handicap the race for the Super Bowl. What’s interesting about this season is that there are plenty of very good teams but few that could be classified as truly dominant. Nine teams have an Elo rating4 of at least 1600, but none of them has cracked 1700 on the eve of the playoffs. In only one other season since 1990 — when the NFL expanded its postseason to the current format — have this many teams been squeezed into the 1600-to-1700 range on the Elo scale, and even that season (2015) had one team above 1700:

Because of this logjam of good-not-great teams, nobody heads into the playoffs with better odds than the New Orleans Saints’ 21 percent chance to win it all, according to Elo. That’s the third-lowest pre-playoff championship probability for a Super Bowl favorite since 1990, trailing only the 2015 Carolina Panthers and 2009 San Diego Chargers at 20 percent apiece. It’s also much lower than the 30 percent average for the typical pre-playoff favorite before this year.

Overall, this year’s favorites are less likely to win the Super Bowl than usual — meaning the Saints and Kansas City Chiefs have a lower probability than the typical top two going into the playoffs — while most of the lesser teams have a better chance than you’d expect to see in an average year.

This year’s playoffs are more wide-open than usual

Probability of winning the Super Bowl by rank (among playoff field) for the 2018 season and the average of the 1990-2017 seasons, according to FiveThirtyEight Elo ratings

rank 2018 Team 1990-17 Avg.
1 Saints 21%
30%
2 Chiefs 20
21
3 Patriots 14
14
4 Rams 13
10
5 Bears 7
7
6 Ravens 6
5
7 Chargers 4
4
8 Eagles 4
3
9 Seahawks 4
2
10 Cowboys 3
2
11 Texans 3
1
12 Colts 3
1

All numbers are as of the final regular-season game of a given year. 2018 probabilities may not add up exactly to 100 percent because of rounding.

Source: Pro-Football-Reference.com

Given all of this, the wild-card round could take on more significance than usual, since it’s not a stretch to imagine one of the teams playing this weekend taking home the Lombardi Trophy when all is said and done.

If Elo had to pick a favorite from that category, it would be the Chicago Bears, who are currently tied for third in the league in Elo and will host the defending Super Bowl champion Philadelphia Eagles (tied for No. 7) on Sunday at 4:40 p.m. ET. Chicago finished the regular season having allowed the league’s fewest points, so this is a classic Monsters-of-the-Midway Bears team in that sense. But quarterback Mitchell Trubisky is also playing much better than the typical Chicago QB from playoffs past — he’s no Jim Miller or Rex Grossman, for instance. According to ESPN’s Total Quarterback Rating,5 Trubisky was the NFL’s third most effective quarterback on a per-play basis this season. While he had some lows (such as a dreadful 29.5 QBR in an opening-week loss to the Packers) to go with the highs (like a 98.9 QBR vs. Tampa Bay in Week 4, one of the highest single-game marks on record), Trubisky’s strides as a second-year passer helped Chicago’s offense — which ranked a respectable 13th in expected points added — be more in line with its dominating defense.

In fact, according to our experimental quarterback-adjusted Elo ratings, Trubisky enters Sunday’s game with the best QB adjustment of any Bears postseason signal-caller since the 1986 Super Bowl, when Jim McMahon was worth approximately 36 more points of Elo than an average quarterback (and promptly shredded the New England Patriots defense for 256 yards and a 104.2 passer rating in a 46-10 rout). Trubisky himself is worth an estimated 18 points of Elo, which is why Chicago stands out if we map out the QB adjustment and base (QB-neutral) Elo rating for each of this year’s wild-card-round combatants:

The remainder of the wild-card field lines up roughly in inverse order between quarterback quality and that of the rest of the team. Some teams — such as Andrew Luck’s Colts and Deshaun Watson’s Texans — have gotten to where they are largely because of their standout quarterback play. Others, like the Ravens, are doing a lot better recently than we’d expect from their QBs’ performance alone. Baltimore has won six times in the seven games since Lamar Jackson replaced Joe Flacco as the starter in Week 11, though Jackson himself ranked third-worst among qualified quarterbacks in QBR this season, and that includes his stellar rushing statistics.6 (Beyond his own stats, Jackson’s effect on the team’s overall running game shows up under the team’s QB-neutral Elo rating.)

Everyone else is somewhere in between, including the resurgent Eagles with backup Nick Foles, whose QB adjustment is back roughly where it was after Philly beat the Patriots in the Super Bowl, but took many twists and turns to get there; the Chargers with 37-year-old Philip Rivers, whose own adjustment has fallen by 38 Elo points since Week 13 with a string of mediocre outings down the stretch; the Dak Prescott-led Cowboys, whose team QB adjustment has hovered around average all season; and Russell Wilson’s Seahawks, whose own run-heavy attack masked another season of highly efficient passing.

How Elo sees the wild-card round playing out

Win probabilities for Week 18 games according to two methods — standard Elo and a version that contains an adjustment for starting quarterbacks

Standard Elo QB-Adjusted Elo
Team Rating Win Prob. Base Rtg Starting QB QB Adj. Win Prob.
CHI 1640 61% 1644 Mitchell Trubisky +18 66%
PHI 1624 39 1606 Nick Foles +2 34
BAL 1627 60 1650 Lamar Jackson -42 61
LAC 1624 40 1580 Philip Rivers +12 39
DAL 1572 54 1569 Dak Prescott 0 55
SEA 1605 46 1572 Russell Wilson +26 45
HOU 1551 56 1537 Deshaun Watson +28 58
IND 1578 44 1533 Andrew Luck +38 42

Home teams are in bold.

Elo quarterback adjustments are relative to average, based on a rolling average of defense-adjusted QB stats (including rushing).

Source: Pro-Football-Reference.com

Of those, Elo gives the best chance of advancing to the Bears, followed by the Ravens. And upset-wise, the best odds belong to the Seahawks against the Cowboys, regardless of whether we adjust for recent QB performance. Whichever teams win, they’ll have to contend with road games in the divisional round — but given the overall state of the league, they’ll still have a better chance than usual to knock somebody off and forge their own path to the Super Bowl.

FiveThirtyEight vs. the readers

To keep tabs on each team’s classic Elo as the weekend plays out, be sure to check out FiveThirtyEight’s NFL prediction interactive, which simulates the rest of the season 100,000 times and tracks how likely every team is to advance through the playoffs and win the Super Bowl. And even though the regular season is over, you can still pick against the Elo algorithm in our prediction game and keep climbing up our giant leaderboard.

According to data from the game last week, here are the matchups in which Elo made its best — and worst — predictions against the reader picks for Week 17:

Elo’s dumbest (and smartest) picks of Week 17

Average difference between points won by readers and by Elo in Week 17 matchups in FiveThirtyEight’s NFL prediction game

OUR PREDICTION (ELO) READERS’ PREDICTION
PICK WIN PROB. PICK WIN PROB. Result READERS’ NET PTS
TEN 62% IND 55% IND 33, TEN 17 +15.1
NO 87 NO 77 CAR 33, NO 14 +13.3
MIN 56 MIN 50 CHI 24, MIN 10 +3.5
LAC 64 LAC 70 LAC 23, DEN 9 +1.8
PHI 65 PHI 71 PHI 24, WSH 0 +1.6
HOU 71 HOU 74 HOU 20, JAX 3 -0.1
ATL 54 ATL 56 ATL 34, TB 32 -0.4
PIT 83 PIT 83 PIT 16, CIN 13 -1.7
SEA 88 SEA 86 SEA 27, ARI 24 -2.0
KC 88 KC 86 KC 35, OAK 3 -2.3
LAR 85 LAR 83 LAR 48, SF 32 -2.6
NE 89 NE 86 NE 38, NYJ 3 -2.6
GB 65 GB 67 DET 31, GB 0 -4.4
BUF 59 BUF 55 BUF 42, MIA 17 -5.5
DAL 62 DAL 56 DAL 36, NYG 35 -8.2
BAL 81 BAL 68 BAL 26, CLE 24 -10.5

Home teams are in bold.

The scoring system is nonlinear, so readers’ average points don’t necessarily match the number of points that would be given to the average reader prediction.

Even though the readers knew about various Week 17 roster shenanigans (such as resting starters) and Elo didn’t, the algorithm did what it’s been doing most of the season, beating the field by an average of 5 points per reader. (Elo beat the average reader 16 times in 17 weeks during the regular season.) Readers picked up points for trusting Luck and Indy against the Blaine Gabbert-led Titans in Sunday night’s do-or-die regular-season finale, and they also got credit for fading the Saints, who were resting starters against the Panthers in a meaningless contest. But at times that knowledge came back to haunt them, such as when they dropped the odds of the Ezekiel Elliott-less Cowboys against the Giants, only to see Dallas storm back and win. (And it was odd to see Elo underestimate the Browns, which it’s been doing all season, only to have that work out at season’s end.)

Either way, congrats to Jake Horowitz, who led all identified users in Week 17 with 294.2 points, and to good ol’ Greg Chili Van Hollebeke, who hung on to his No. 1 ranking for the season with 1,168.1 points. Thanks to everyone who has been playing — and the game isn’t over yet! You should keep making picks and trying your luck against Elo throughout the playoffs.

Check out our latest NFL predictions.