Our Guide To The NFL’s Divisional Round

Now that the NFL’s wild-card weekend is complete, the top seeds in each conference — including Lamar Jackson and our Super Bowl favorites, the Baltimore Ravens — will be in action for the divisional round. Ahead of the games, we’ve broken down each matchup using our Elo ratings — which track each team’s form, with adjustments for each starting quarterback — and also identified the phases of the game in which each team was best (and worst) according to ESPN’s expected points added (EPA)¹ during the regular season. Our predictions didn’t see the upsets coming in the opening round of the playoffs, so Elo is looking forward to a better weekend this time around!

The Vikings engineered the top upset of wild-card weekend, upending the New Orleans Saints in a game we gave them just a 23 percent chance of winning. Now they’ll get a chance to pull off another surprise victory, this time against the top-seeded 49ers. Minnesota’s Kirk Cousins silenced his doubters some against New Orleans, and he will be the higher-ranked quarterback in his matchup versus Jimmy Garoppolo, according to our QB Elo metric. But Jimmy G. has the better supporting cast around him, including the superior defense according to EPA (where San Francisco ranked second, compared with Minnesota at No. 6). The Vikings did beat Garoppolo and the Niners in Week 1 of last season, holding SF to 237 net passing yards and forcing four turnovers; they might need to follow a similar formula to win Saturday. But it’s easier to envision the 49er defense slowing down the Vikings’ run game and forcing Cousins to make plays, with San Francisco ultimately coming out on top. Elo’s spread: San Francisco -5

Normally, a team that knocks the Patriots out of the playoffs would get more respect than this. But the New England team that Tennessee beat last Saturday wasn’t exactly a vintage version of the Pats — and the Baltimore team waiting for the Titans looks like the scariest opponent in the league. According to Elo, the Ravens have the best QB in the NFL (Jackson) and the best supporting cast as well, including a pass defense that ranks third in EPA per game and sixth at limiting opposing QBs’ Elo performances. The Titans’ best hope is to use RB Derrick Henry and a bruising rushing attack (No. 6 in EPA) to exploit a Baltimore run D that ranks just 22nd this season. But it’s unclear if the Tennessee defense has any answer for Jackson’s ability to shred teams using both his arm and his legs, with help from the Ravens’ 11 other Pro Bowl players. If Tennessee does manage to win this one, it would be a significantly bigger upset than beating the Patriots was. Elo’s spread: Baltimore -13

Welcome to what might end up being the most dynamic quarterback matchup of the playoffs. In one corner, you have Deshaun Watson — Elo’s fifth-ranked QB — who overcame a shaky first half to lead Houston past Buffalo in a wild opening-round thriller. In the other corner, you have Patrick Mahomes — No. 3 in QB Elo — who’s still the reigning NFL MVP, in the midst of possibly the best back-to-back passing seasons in NFL history.² While K.C. has the edge under center, it isn’t by a huge margin. The larger differentiating factor between these teams is in the surrounding talent, where Mahomes’s pals also rank third — including, perhaps surprisingly, the No. 3 QB Elo defense — while Watson’s support ranks just 14th, with the No. 26 QB Elo defense. The Texans did narrowly beat the Chiefs back in Week 6 with a near-perfect game plan: They ran for 192 yards, picked up 35 first downs, spread completions around to seven different receivers and limited Kansas City to just 47 offensive plays (vs. 83 for Houston!) with barely 20 minutes of possession. A similar result is possible for Houston to orchestrate again on Sunday — but it’s not very likely. Elo’s spread: Kansas City -11

Elo has been down on Russell Wilson and the Seahawks in recent weeks, to its own detriment. In the most recent chapter, Seattle defied the model — with a little help from Carson Wentz’s injury — and beat the Eagles in Philly. Now we list the Seahawks as underdogs again heading to Green Bay … so can Seattle pull off yet another upset? The QB matchup certainly tilts in favor of Wilson, who plainly outperformed Aaron Rodgers during the regular season and played very well against the Eagles last week, too. But the Packers built their 13-3 record as much on a superior pass defense (10th in EPA) and running game (seventh) as on Rodgers’s passing (12th). (They also got pretty lucky in close games — which the Seahawks know a thing or two about as well.) If Seattle’s RB corps weren’t so riddled with injuries, it might be able to take advantage of Green Bay’s 29th-ranked run defense. But instead, this game will probably come down to whether Wilson can keep the score close throughout and deliver a winning performance in the clutch, where he had the best Total QBR of any passer in football this season.³ Elo’s spread: Green Bay -7½

FiveThirtyEight vs. the Readers

As a weekly tradition here at FiveThirtyEight, we look at how our Elo model did against everybody who made picks in our forecasting game. (If you entered, you can find yourself on our leaderboard here.) These are the games in which Elo made its best — and worst — predictions against the field last week:

Let’s be honest: Elo’s overconfident pick of the Eagles against the Seahawks was bad. Real bad. With a net loss of 48 points (!) relative to the average reader, it was the worst pick made by either the algorithm or the human pickers all season long, by some considerable margin. Sure, Elo didn’t know that Wentz would get knocked out in the first quarter of that game, causing Josh McCown (whose QB Elo is 52 points lower than Wentz’s) to play most of the contest. But the readers rightly hammered Elo for perhaps reading too much into Philly’s late-season push (plus Seattle’s late-season lull) and the Seahawks’ weak point differential. The model was lucky to have sided with the Texans over the Bills, but for the weekend Elo lost to the average reader by a margin of 20.0 points, giving the model its worst week of the season (and just its third overall loss to the field in 18 tries).

Amid a week of upsets, congratulations are in order for Christian Jensen, who leads all readers in the postseason with 200.0 points, and for Griffin Colaizzi, who took the lead in the full-season contest with 989.0 points. Thanks to everyone who played — and if you haven’t, be sure to get in on the action! You can make picks now and try your luck against Elo in the playoffs, even if you missed Week 18.

Check out our latest NFL predictions.