Divisional Round Sunday: Advanced Reviews
Divisional Round Sunday: Advanced Reviews
The Bills fall flat at home to Joe and Bengals, while the Cowboys extend a historic streak of playoff failures
The adjusted scores quantify team play quality, with emphasis on stable metrics (success rate) and downplaying higher variance events (turnovers, special team, penalties, fumble luck, etc). Adjusted expected points added (EPA), in conjunction with opportunity-based metrics like total plays and drives, projects adjusted points. Adjusted scores have been tested against actual scores and offer slightly better predictive ability, though their primary benefit is explanatory.
All 2022 and historical Adjusted Scores and other site metrics are available in a downloadable format to paid subscribers via Google Sheet.
BILLS vs BENGALS
The Cincinnati Bengals came to snowy Buffalo and stunned the Bills. There was certainly tension during the game, but the Bengals were in control almost the entire way. Because the Bills offense is normally so explosive, you couldn’t be sure they were dead until well into the fourth quarter.
It’s a convert-or-not league, and the Bills couldn’t keep drives alive. They were four of 12 on third downs, including failures on 3rd & 2, 4, 5 and 7. The Bengals, on the other hand, made hay on the highest leverage plays.
Joe Burrow found Hayden Hurst for a 15-yard touchdown on 3rd & 7 (+3.2 EPA). Tee Higgins drew a 26-yard pass interference down to the Bills’ 4 yard line on 3rd & 2. And Burrow converted multiple 3rd & 10’s with passes to unsung options Trenton Irwin and the aforementioned Hurst.
It’s wasn’t only the late downs that the Bengals dominated, as their elite 54.7% success rate was nearly 13% better than the Bills offense. The Bengals were effective running the ball with Joe Mixon (+3 EPA), but were smart to still lean into the passing game (+11.3% pass rate over expectation), despite weather concerns.
Craziest thing about that game is that I assumed the Bengals likely path to victory was via a couple huge Allen turnovers. Yes, Allen threw an INT, but the game was over already at that point. The interception by Cam Taylor-Britt moved the Bills win probability from roughly 0% to roughly 0%.
The numbers of the game were for “Pass EPA”, or EPA on pass attempts that didn’t result in an interception, for the two quarterbacks. Joe Burrow had a 13.8 EPA advantage over Allen, whose number (1.6 EPA) was his third worst of the season. His lowest two games were facing the New York Jets.
Burrow’s Achilles heel of sacks didn’t have an effect in this game, taking only one for a small 1.7 expected points lost. Burrow also virtually matched Allen in running efficiency, the fifth lowest EPA total for Allen on the ground this season.
All credit goes to Lou Anarumo and the Bengals defense for shutting down what the Bills fundamentally want to do on offense, not just hoping Allen has a big turnover game.
No matter what the pundits will say about the Bengals matching up well against the Chiefs having beaten them in three straight, I think the Chiefs should be happy to face them at home rather than the Bills at a neutral site. The Chiefs open as 1-point favorites, and probably would have been underdogs against the Bills in Atlanta.
WHAT’S NEXT FOR THE BILLS?
Despite also losing the Divisional Round last year, this offseason will come with a lot more soul searching for the Bills. Offensively, there wasn’t anything more the Bills could have done against the Chiefs’ fast-scoring machine then, but it ground to a halt at home this year.
I expect the Bills to prioritize receiver and offensive line depth in the offseason, and hope Von Miller can still play at an elite level entering his age-34 season. The Bills are currently projected to be $8.4 million over the cap in 2023, with more than $150 million allocated to 10 players.
One huge difference will be the expiration of Allen’s low-cap window following his 2021 extension. Allen’s cap hit goes from $10 million and $16 million the last two years to $40 million in 2023. Stefon Diggs’ cap hit will jump from $11.7 million to $20.3 million.
49ERS vs COWBOYS
It’s now 12 straight postseasons for the Dallas Cowboys without making the Conference Championship, a record they hold alone that stretches back to Troy Aikman, Emmitt Smith, Michael Irvin and Barry Switzer as head coach. Dak Prescott now matches Tony Romo with four postseason appearances, each with one Wild-Card loss and three in the Divisional Round.
Neither offense shined in this game, thought the San Francisco 49ers posted an above average success rate and EPA per play against the second best defense in the NFL according to my schedule adjusted power rankings.
On the other sideline was the best defense in the NFL, and the 49ers proved again they deserve that title by holding the Cowboys to 25th percentile efficiency, with a dropped pick-6 that didn’t make it into the stat sheet.
The Cowboys did provide some unintentional comic relief with the final play.
You have to assume laterals and whatnot were in the plan, but it never go there. Listen, I’m an analytics guy searching for every edge. At the same time, they’d have to be a really strong case for an increased win probability on a play that going to make you look like a total fool if it doesn’t work. Especially, when the opponent has timeouts and can sit back and think about how to defend the formation.
The analytics say 0.01% chance to win and not memeing yourself is worth more than 0.02% chance to win with that type of play.
I couldn’t help myself to having a little fun with the Dak supporters after the game, but it almost seemed like a Russian misinformation campaign the way my twitter feed was flooded with their pre-rebuttals to the Dak haters.
I can’t say I always follow this advice, but I’d suggest everyone starts first with their own independent analysis on events, not only positioning to answer critics. Otherwise, you’re letting the other side, people you presumably don’t even respect, dictate your actions. Did Dak get great support? Not really. Was he bad, also yes.
In fact, Dak’s most positive plays were an underthrown, 46-yard DPI to CeeDee Lamb (+3.1 EPA) and a 4th & 4 scramble that was a whisker away from a sack (+2.4 EPA). The Cowboys biggest “offensive” play of the game came on special teams, recovering the Ray-Ray McCloud muffed punt at the 49ers’ 21 yard line.
Dak’s second interception, which was much uglier than the first, was a real killer and the most negative play of the game (-5.1 EPA), mostly because it came in a tremendously advantageous context (2nd & 2 at the 49ers’ 18 yard line).
Brock Purdy continued to defy the odds. I pointed out earlier this week that my adjusted quarterback efficiency (AQE) metric had him as roughly a league-average quarterback after adjustments, partially because of an easy schedule. The Cowboys didn’t prove to be the test I thought they could be, as Purdy largely avoided mistakes, although with the benefit of 50% designed-run rate.
The Eagles open as 2.5-point favorites hosting the 49ers next week, which is a little more in their favor than my initial numbers. I’ll post more later this week on the team’s updated power rankings.
FOURTH DOWN DECISIONS
There wasn’t much to see in the Bills-Bengals matchup, but the 49ers and Cowboys both provided fodder for nerds to lament.
I get that Kyle Shanahan - a boomer at heart - wouldn’t want to take risks with the league’s best defense and a 7th round rookie quarterback. But if you’re Mike McCarthy, and your quarterback has a $50 million cap hit next season, you can’t pass up the chance to take that win probability, no matter how good the defense you’re facing might be.
The Cowboys were the underdogs on the road, and needed to embrace all the edges. You can’t punt twice around midfield with fewer than five yards to go in the most important game of the season.