The dust is still clearing from a divisional-round weekend that gave us mostly chalk but also one massive upset. The Baltimore Ravens had been our favorite to win the Super Bowl — and pretty heavy ones at that — but Derrick Henry and Tennessee had different plans. Now, the Titans face our new favorites, the Kansas City Chiefs, in the AFC title game, while the San Francisco 49ers host the Green Bay Packers on the NFC side. Looking ahead, we’ve broken down the matchups 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. This week’s spreads aren’t sizable enough to generate any historic upsets, but the upside is that we should see a more balanced set of matchups here on the cusp of the Super Bowl.
Tale of the tape: Kansas City vs. Tennessee
3:05 p.m. ET Sunday
|Patrick Mahomes||Starting QB||Ryan Tannehill|
|2nd||QB Elo rank||12th|
|4th||QB’s supporting cast||2nd|
|5th||Avg. QB Elo defense||16th|
|Pass offense||Biggest EPA strength||Run offense|
|Run defense||Biggest EPA weakness||Special teams|
After a pair of stunning upsets over the defending champs (New England) and the AFC’s No. 1 seed (Baltimore), what does Tennessee have in store next? Certainly the Titans are underdogs again, on the road at Arrowhead Stadium against the Chiefs. But Elo has also learned its lesson about them — to a degree — and has boosted Tennessee’s rating by a whopping 81 points since the start of the playoffs. Still, the Titans have their work cut out for them against a vastly superior (on paper) passing attack directed by reigning league MVP Patrick Mahomes. In leading Kansas City on a furious comeback from down 24-0 against the Texans, Mahomes posted a sky-high 97.9 Total QB Rating from the second quarter onward, so he may not prove as easy for Tennessee’s defense to stymie as Lamar Jackson was. The good news for the Titans, however, is that their offensive strength — the mighty power running of Henry and Co. — ranks fourth in schedule-adjusted EPA per game this season, corresponding well to Kansas City’s weakness in defending the run (29th in rushing defense). The ingredients are there for Henry to have another big day, perhaps similar to the 188 yards he produced against K.C. in a Week 10 Tennessee victory. The only question is whether it will be enough to match the Chiefs’ own firepower again — an unlikely proposition, though not that unlikely. Elo’s spread: Kansas City -5½
Tale of the tape: San Francisco vs. Green Bay
6:40 p.m. ET Sunday
|San Francisco||Category||Green Bay|
|Jimmy Garoppolo||Starting QB||Aaron Rodgers|
|22nd||QB Elo rank||13th|
|1st||QB’s supporting cast||3rd|
|6th||Avg. QB Elo defense||4th|
|Pass defense||Biggest EPA strength||Run offense|
|Special teams||Biggest EPA weakness||Run defense|
The most iconic playoff battle between these two teams happened back in 1999, when San Francisco’s Steve Young hit Terrell Owens for a 25-yard, game-winning TD to defeat Brett Favre and the Packers. The QBs in this year’s edition — SF’s Jimmy Garropolo (-18 QB Elo vs. average) and GB’s Aaron Rodgers (+12) — aren’t quite at the top of their games the way Young (+122) and Favre (+61) were back then, but they are assisted by two of the best supporting casts in the league, including a pair of strong pass defenses. In theory, that makes these teams pretty evenly matched: Even the biggest disparities on one side of the ball (such as Green Bay’s superior running game)2 are matched on the other side (e.g., San Francisco’s run defense is better).3 One thing the 49ers can point to is their 37-8 demolition of the Packers at home in Week 12, when Garropolo outplayed Rodgers and San Francisco outgained Green Bay 339 to 198. But even in that department, there’s some evidence that the team that was blown out in the regular season gains more information for the rematch than the team that dominated. Will that — and plain old regression to the mean — be enough to keep Green Bay more competitive this time around? Probably, though the 49ers are still clear favorites here. Elo’s spread: San Francisco -4
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:
Elo’s dumbest (and smartest) picks of the divisional round
Average difference between points won by readers and by Elo in Week 19 matchups in FiveThirtyEight’s NFL prediction game
|OUR PREDICTION (ELO)||READERS’ PREDICTION|
|PICK||WIN PROB.||PICK||WIN PROB.||Result||READERS’ NET PTS|
|BAL||87%||BAL||84%||TEN 28, BAL 12||+5.8||
|SF||67||SF||68||SF 27, MIN 10||-6.5||
|KC||83||KC||80||KC 51, HOU 31||-8.3||
|GB||74||GB||57||GB 28, SEA 23||-34.5||
A week ago, Elo made a terrible, horrible, no good, very bad pick when it took the Eagles over the Seahawks with 69 percent confidence. This time around, the model whiffed on the Ravens, giving them an 87 percent win probability … but so did basically everyone else. Elo lost only 5.8 points to the field on that pick, which it made up for by gaining 49.3 points over the average reader in the other three games — including a massive 34.5-point gain for picking Green Bay over Seattle. (Elo had been suspicious of the Seahawks for a long time.) As a result, Elo was back to its winning ways for the week, beating the field by 43.5 points on average.
Despite that, some readers did well. Congratulations to Andrew Lindstrom, who leads all (identified) readers in the postseason with 200.0 points, and Jan Hájek, who pulled into first place in the full-season contest with 1,061.3 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 19.
Check out our latest NFL predictions.