The NFL’s wild-card round may have been a brutal showcase of grind-it-out defense and uninspired offensive play-calling. But at least it threw another intriguing wrench into what was already shaping up to be a chaotic postseason. Three of the four favorites (according to FiveThirtyEight’s pregame Elo ratings) lost on opening weekend: the home Texans, Ravens and Bears. Now we’re left with an eight-team field in which every member has at least a 6 percent probability of winning the Super Bowl, and no one is above 22 percent. Over the previous four postseasons, the favorite had an average championship probability of 31 percent at this stage of the playoffs. So with things still looking as wide-open as ever, let’s zoom in on some numbers — including classic Elo and a version with our experimental quarterback adjustments — for each divisional matchup.
|Standard Elo||QB-Adjusted Elo|
|Team||Rating||Win Prob.||Base Rtg||Starting QB||QB Adj.||Win Prob.|
The most up-for-grabs game of the divisional round might be the New England Patriots against the Los Angeles Chargers on Sunday afternoon. This year’s Patriots are not the unassailable juggernaut they usually have been, though they literally have been unbeatable at home (where they’ll be this week). In fact, come kickoff, it will have been 470 days since the Pats last lost a home game of any sort and 2,185 days since they fell at home in the playoffs.
But the Chargers may actually be the superior team. Not only did L.A. have the better record (12 wins vs. 11 for New England), it ranks higher than New England in ESPN’s Football Power Index, Football Outsiders’ Defense-adjusted Value Over Average and Jeff Sagarin’s power ratings, among other rankings.
|Football Power Index||4th||5th|
|Defense-adjusted Value Over Average||3rd||7th|
|Simple Rating System||6th||8th|
|Starting quarterback’s Total QBR||7th||6th|
Talent-wise, the Chargers also have more Pro Bowlers than the Pats (7 to 2) and just as many All-Pros (4 apiece). And although Elo does gives New England a 58 percent chance to reach the AFC championship game for the 10th time in 13 years, that’s the second-lowest pregame probability the Pats have had in a divisional playoff since 2006-07 — when they were given a mere 35 percent chance of beating, you guessed it, the Chargers.1 Then again, the Pats did in fact end up winning that one, in San Diego, under crazy circumstances. As always, it’s tough to count out Tom Brady in the AFC playoffs,2 where he is 22-7 all-time as a starter and hasn’t lost since January 2016.
Previewing the NFL divisional playoff round
On paper, the biggest mismatch of the second round features the New Orleans Saints at home against the Philadelphia Eagles. Las Vegas’s sportsbooks opened with the Saints as 8-point favorites, which would seem to make sense in a clash between a 13-3 top conference seed and a 9-7 team that needed an unlikely confluence of events just to make the playoffs at all. But, of course, the Eagles aren’t just any team — they’re the defending champs, with a script that seems all too familiar.
Once again, backup QB Nick Foles has relieved the injured Carson Wentz (probably for the rest of the season?), and once again Foles is leading Philly on an underdog playoff run. When this happened with last year’s Eagles, coach Doug Pederson changed his offensive approach to better accommodate Foles’s strengths, focusing on higher-percentage short passes with more yards picked up after the catch (mixing in the odd deep bomb), and calling more play-action passes with an emphasis on getting the ball out quickly and avoiding mistakes. So how much of that stylistic shift has played out this time around?
|2018 Season (incl. playoffs)|
|2017 Season (incl. playoffs)|
This year, the differences are more subtle. The Eagles have once again thrown downfield less with Foles, but that hasn’t been paired with more yardage after the catch, nor has it resulted in fewer picks. They’re also calling a lower share of play-action passes for Foles than they did for Wentz, a big reversal from last year. But one constant has been Foles’s ability to get rid of the ball under pressure and avoid sacks, which was a key factor on Sunday against Khalil Mack and Chicago’s ferocious pass rush. The Saints were sixth in sacks this season, so Foles’s quick release should come in handy again this weekend as the Eagles try to avoid a repeat of their 48-7 thrashing at the hands of New Orleans in November. But for all the Foles mania, it remains to be seen if Philadelphia’s defense can slow down Drew Brees and the Saints’ offense after allowing 546 total yards in that earlier matchup.
As my ESPN colleague Adam Teicher recently wrote about, Saturday’s game between the Kansas City Chiefs and Indianapolis Colts is a historically great quarterback matchup. The two QBs — Patrick Mahomes and Andrew Luck — spent the regular season tossing 89 combined touchdown passes, an NFL record for QBs facing off in any postseason contest. (Even considering the NFL’s spike in passing numbers, that’s a lot of touchdowns!)
The Chiefs are looking to finally snap a decades-long run of postseason futility that hasn’t seen them advance past the divisional round since 1993-94. But the bad luck of running into, um, Luck is part of their ongoing problem at this stage of the postseason. Among the 21 teams that have played at least five divisional playoff games since the Chiefs’ dry spell began, the Chiefs have faced the third-most difficult slate of opposing quarterbacks, according to our QB-adjusted version of Elo:
|Average Opp. Elo||Win Percentage|
|Team||Games||QB-Neutral||QB Adjustment||Effective Rating||Predicted||Actual|
Thanks to a steady dose of all-time greats such as Brady, Peyton Manning, John Elway, Ben Roethlisberger and now Luck, the Chiefs have seldom caught a break by facing a comparatively weak QB at this point in the playoffs. Since 1994-95, only the Jaguars and 49ers have had worse luck in that regard. (Of course, those teams still managed to overcome it and win more often than not, while the Chiefs keep finding ways to lose.) But it’s also worth pointing out that KC has seldom had the superior QB in the matchup, which it does have this year in Mahomes. According to Elo, Mahomes is not only better than Luck, but he has the second-best pregame rating adjustment of any Chiefs playoff starter in a divisional round game since the 1970 merger, trailing only Trent Green from 2003-04 (when Kansas City gained 408 yards — and didn’t punt the ball once — but still lost because the Colts gained even more yards and also didn’t punt.) With a likely MVP under center, the Chiefs are hoping they finally have the ingredient that was missing in those previous postseason disappointments.
Instead of having to face the Dallas Cowboys in January, the Los Angeles Rams probably would prefer it if the playoffs had been held a month and a half ago, back when they were on pace for 14 wins and sat as Super Bowl favorites in Vegas. According to Elo, the Rams peaked after beating the Detroit Lions in Week 13 — after which they went 2-2 and shed nearly 40 points off their rating. But the good news for L.A. is that similar “peaking too early” teams tend to do pretty well for themselves in the playoffs. I looked for other teams that cracked a 1650 Elo rating (with the QB adjustment) through 12 games of an NFL regular season,3 then lost at least 30 points of Elo over the final four games before going into the playoffs. Of those 20 teams, 14 still made the conference championship, with nine making the Super Bowl and five winning it all (the most recent of which were the 2009 Saints). So any rumors of the Rams’ premature demise may have been greatly exaggerated.
If there is a takeaway from the Rams’ late-season slump, it might be a reminder that one of the most important factors for Los Angeles will be getting back to selling the run and keeping the Cowboys off-balance with play-action passing. During the Rams’ 11-1 start, they led the league with 5.3 expected points added per game off play-action passing, more than 0.8 points per game better than New England, the next-best offense. In Weeks 14 through 17, that number dropped to 3.2 EPA per game, which ranked only seventh-best.
|Category||Weeks 1-13||Weeks 14-17|
|Passing EPA per game||1st||7th|
|Yards per pass||5th||15th|
|Passing yards per game||1st||8th|
Though L.A. running back Todd Gurley was on track for an MVP-caliber season before missing games with a balky knee down the stretch, there isn’t evidence that missing a stud RB necessarily hurts a team’s play-action game. The Cowboys had one of the NFL’s best run defenses this season, so overusing Gurley to set up play-action on Saturday might be a waste anyway. But getting Dallas to commit to defending the run and then shredding the Cowboys through the air seems like the Rams’ best ticket to the NFC title game.
FiveThirtyEight vs. the readers
To keep up with every team’s Elo rating during the playoffs, 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 to the Super Bowl. You can also pick playoff contests against the Elo algorithm in our prediction game and keep climbing up our giant leaderboard. (Or you could be like me, and forget to set picks before the first round of the playoffs…)
According to data from the game, here’s how readers did against the computer on wild-card weekend:
|OUR PREDICTION (ELO)||READERS’ PREDICTION|
|PICK||WIN PROB.||PICK||WIN PROB.||Result||READERS’ NET PTS|
|BAL||60%||LAC||51%||LAC 23, BAL 17||+17.0||
|HOU||56||HOU||53||IND 21, HOU 7||+0.8||
|CHI||61||CHI||65||PHI 16, CHI 15||-16.8||
|DAL||54||SEA||54||DAL 24, SEA 22||-22.5||
Readers won big when the Chargers won on the road over Lamar Jackson and the Ravens, and the Colts’ win over the Texans was basically a wash (both the readers and Elo incorrectly picked Houston). But Elo took the weekend on the strength of the Eagles’ win over the Bears — the algorithm liked Philly slightly more than the readers — and especially the Cowboys’ victory over the Seahawks. Despite the game being in Dallas, the average reader assigned Seattle a 54 percent chance of winning, and that led to a massive loss of points after Michael Dickson’s weird onside kick attempt went awry. Elo has now beaten the average reader 17 times in 18 weeks this season.
Having said that, congrats are in order to Alex McQuillen and Ben Zornes, who currently lead all users in the postseason with 200.0 points apiece, and to Neil Mehta, who moved into first place for the season with 1,128.1 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.