Bris Vegas NFC Playoff Seeding Projection – Week 16

What a difference the bye week made. Two weeks ago, the Commanders were on pace to clinch the 6^th playoff seed and there was a buzz developing about their chances to not just make the playoffs, but possibly even make some noise when they got there.

Then came the rematch with division rival New York Giants, the Commanders’ primary competition for the remaining playoff spots, with whom they had tied the last time they played. This game was a pivotal moment in the Commanders’ season. If they could take just care of business against the Giants, they could all but assure themselves of a playoff seed. On the other hand, if they were unable to rise to the occasion, they would find themselves with a very difficult path to the playoffs though they still control their own playoff destiny.

The problem with pivotal moments is that you don’t always pivot in the right direction. It was the Giants, not the Commanders, who rose to the occasion and elevated their play on defense after the bye to gain the critical advantage in the game and the playoff race. The Commanders didn’t help themselves by getting away from the running game when it was working and making baffling decisions to punt in field goal territory. The defense also lacked pop and failed to make the big plays that we have become accustomed to seeing since the Week 6 win over the Bears. The Commanders also had to contend with terrible officiating. However, they left enough opportunities on the field, that the game should not have been close enough for two bad calls on their final possession to make the difference.

Two weeks ago, with the playoff buzz growing, I did my first NFC playoff seed projection. My method uses the efficiency metric Expected Points Added (EPA) to project outcomes of the matchups between opposing teams’ offenses and defenses. While the Commanders are still technically in the drivers’ seat, as they will make the playoffs if they win all three remaining games, they will have to beat two heavily favored games to pull that off. More likely, they will need Detroit to lose some games it should win or for the Giants to not win any to make the post-season. As unlikely as it may seem, their playoff hopes are still alive. Therefore, I will continue to run weekly projections until they are eliminated from contention.

Let’s see if the model found a way for the Commanders to still earn their way into the playoffs after blowing the opportunity against New York.

Method Update

After the so-so performance of the Week 15 predictions, which I’ll get to in a minute, I made a few tweaks to the model for Week 16.

The method is still largely the same as last week. Outcomes of the matchups between each team’s offense and their opponents’ defense are determined by adding the offense’s EPA/play (which is positive for good offenses) to their opponents’ EPA/play on defense (which is negative for good defenses). This is essentially like asking how many points the defense will take off the offense’s average performance. The result is treated as the offensive team’s score. The team with the larger score wins the matchup.

In the first attempt, EPA/play on offense was calculated over the entire span this season that each team has played its current starting quarterback. In an attempt to make the model more predictive, in Week 16, EPA/play on offense was calculated from Week 8, unless there was a more recent change at QB, in which case it was calculated from the current QB’s first start, as before. This is intended to emphasize teams’ recent performance, which should lead to better predictions for teams that have got better or worse since the first half of the season, such as the Commanders and the Lions.

Defensive EPA/play is still calculated from Week 8.

QB Substitutions

There was a bit of movement at QB this week, which required some updates to how EPA/play on offense was calculated, and required making a few calls where teams’ starting QB is unclear at the time of this writing:

Arizona – Backup QB, Colt McCoy was injured in last week’s game against Denver. Third-string QB Trace McSorely is set to make his first NFL start. This presents a problem, as there is no precedent for Arizona playing with McSorely at QB. I decided to calculate EPA/play from Week 10 on the assumption that it will be more like having McCoy at QB than Kyler Murray.

Philadelphia – Starter Jalen Hurts sprained his shoulder against Chicago and it is likely that Gardner Minshew will start against Dallas. There is no precedent for the Eagles playing with any QB but Hurts this season, so there was no good alternative to calculating EPA/play from Week 8.

Atlanta – Rookie Desmond Ridder is making his first start for Atlanta in Week 16. I don’t expect he will make that much difference over Marcus Mariota, so EPA/play was calculated from Week 8.

Baltimore – There has been some talk that Lamar Jackson might be hurried back from injury in Week 16, but that seems unlikely. Therefore, no change was made.

---

Vincent Carchietta-USA TODAY Sports

Week 15 Prediction Results

Before we get to the projection of the remaining season, let’s review how the model did in Week 15.

NFC East

Bears L (3-11) – Eagles W (13-1) CORRECT

Jaguars L – Cowboys W (11-3) INCORRECT

Commanders W (8-5-1) – Giants L (7-6-1) INCORRECT

Commanders – Giants Matchup: The model predicted that the Commanders’ defense would give them the decisive edge over the Giants’ slightly-above-average offense, and that Washington’s own offense would be able to exploit the Giants’ weakness on defense. What it failed to predict is how much the Commanders’ defense would soften up during the bye week. The defense, which had ranked 3rd in the league from weeks 6 through 14, with an EPA/play of -0.110, fell flat in Week 15, ranking 22^nd in the league with an EPA/play of 0.052. A major difference from previous weeks was the lack of high-impact plays on defense, such as turnovers and drive-ending sacks.

The Giants’ offense performed about as expected against the Commanders, ranking 11^th in the league in Week 15, with an EPA/play of 0.052. That was a modest improvement over their 13^th place ranking through Week 14 and EPA/play of 0.029.

Their defense, on the other hand, came alive against the Commanders, making high-impact plays including the sack-touchdown in the 2^nd quarter (EPA -5.64), and the sack-fumble to end the Commanders’ second-to-last offensive drive (EPA -4.16). The Giants’ defense had ranked 30^th in the league in weeks 6 through 14, with a terrible EPA/play of 0.105, which was used to make the prediction. In Week 15, against the Commanders, they shot up the league rankings to 6^th place, with a formidable EPA/play of -0.126.

That was largely, but not entirely based on improvements on the Giants’ defense, most notably the coming out party by Kayvon Thibodeaux. Their EPA/play on defense was also inflated somewhat by the bad calls by the referees wiping a score off the record on the final drive and giving the Giants credit for a goal-line stand, instead. In actual fact, the “goal line stand” was really just the result of the Commanders’ failure to execute on two plays and the referees actively interfering on the other two.

NFC North

Vikings W (10-4) – Colts L CORRECT

Jets W – Lions L (6-8) INCORRECT (note: I had projected Mike White to start, Jets started Zack Wilson)

Packers W (6-8) – Rams L (4-10) CORRECT

NFC South

Bucs L (6-8) – Bengals W CORRECT

Panthers W (6-8) – Steelers L INCORRECT

Saints W (5-9) – Falcons L (5-9) CORRECT

NFC West

Seahawks L (7-7) – 49ers W (10-4) CORRECT

Broncos W – Cardinals L (4-10) CORRECT

Week 15 Summary and Excuses

Overall, the model got seven predictions right and four wrong, for a total of 64% correct. While that is not spectacular performance, it is also not terrible, considering that there were four upsets:

Lions (+1) over Jets

Steelers (+3) over Panthers

Jaguars (+4) over Cowboys

Giants (+4.5) over Commanders

The model most likely got the Lions-Jets prediction wrong, because it was assumed that the Jets would stick with Mike White, who has played much better at QB than Zach Wilson. Unfortunately for them, they had to go back to Wilson to start in Week 15 because White is injured, throwing the model off.

The other game I’ll make excuses for getting wrong is the Commanders – Giants. As discussed above, the model got this one wrong by failing to predict that the Commanders’ defense would regress substantially after the bye and that the Giants’ defense would take off the way it did. I don’t think either of those developments was highly predictable; nor was the appalling officiating.

The Jaguars upset of the Cowboys caught a lot of people by surprise. I have nothing to say about the Steelers – Panthers game, except the model likes the Panthers’ offense since Sam Darnold took over from Baker Mayfield. It may soon gain an appreciation of Mitch Trubisky in Pittsburgh (Week 15 QBR 83.5, EPA 5.7 – 7^th best in the league), unless he regresses to his long term average.

---

Photo by Steph Chambers/Getty Images

Week 16 Season Projection

Week 16 Results

NFC East

Eagles L (13-2) – Cowboys W (11-4)

Giants L (8-6-1) – Vikings W

Commanders L (7-7-1) – 49ers W

Commanders – 49ers Matchup: The outlook for this matchup has not changed substantially since last week, although the 49ers predicted margin of victory has increased. The 49ers continue to play at a high level on both sides of the ball, with no let-up since Brock Purdy took over from Jimmy Garoppolo. Due to the Commanders’ backward step last week, the 49ers have gained an advantage on defense, as well as offense. This game could get ugly.

NFC North

Vikings W (12-3) – Giants L

Lions W (8-7) – Panthers L

Packers L (6-9) – Dolphins W

Bills W – Bears L (3-12)

NFC South

Bucs W (7-8) – Cardinals L

Lions W – Panthers L (5-10)

Saints W (6-9) – Browns L

Falcons L (5-10) – Ravens W

NFC West

49ers W (11-4) – Commanders L

Seahawks L (7-8) – Chiefs W

Bucs W – Cardinals L (4-11)

Broncos W – Rams L (4-11)

Week 17 Results

NFC East

Saints L – Eagles W (14-2)

Cowboys W (12-4) – Titans L

Browns L – Commanders W (8-7-1)

Colts L – Giants W (9-6-1)

Commanders – Browns Matchup: Don’t look now, Commanders’ fans, but DeShaun Watson might be getting better after his rusty Week 13 debut. The Browns’ offense is still one of the worst in the league, ranking 28^th since his return with an EPA/play of -0.122. The Commanders’ offense is ranked 18^th in that span, with an EPA/play of -0.034. Let’s hope that Watson is still shaking off the rust when the teams meet in two weeks.

Even with the backward step in the Giants’ rematch, the Commanders (Defense EPA/play -0.99, rank 3^rd ) still maintain an advantage on defense over the Browns (Defense EPA/play -0.017, rank 10^th).

The model still predicts a Commanders’ win, but Cleveland is closing the gap.

NFC North

Vikings L (12-4) – Packers W (7-9)

Bears L (3-13) – Lions W (9-7)

NFC South

Bucs L (7-9) – Panthers W (6-10)

Saints L (6-10) – Eagles W

Cardinals L – Falcons W (6-10)

NFC West

49ers W (12-4) – Raiders L

Jets W – Seahawks L (7-9)

Cardinals L (4-12) – Falcons W

Rams L (4-12) – Chargers W

Week 18 Results

NFC East

Giants L (9-7-1) – Eagles W (15-2)

Cowboys W (13-4) – Commanders L (8-8-1)

Commanders – Cowboys Matchup: The Cowboys (0.154 EPA/play) just have a far superior offense to Washington (-0.021). The Commanders have a slight edge of defense (-0.099 EPA/play vs Dallas -0.088 EPA/play). However, unless the Commanders’ defense returns to the form it was showing a few weeks ago, or the Cowboys rest their starters, this could be as lopsided a game as the first time these teams met in Week 4.

NFC North

Vikings W (13-4) – Bears L (3-14)

Lions W (10-7) – Packers L (7-10)

NFC South

Bucs W (8-9) – Falcons L (6-11)

Panthers W (7-10) – Saints L (6-11)

NFC West

Cardinals L (4-13) – 49ers W (13-4)

Rams L (4-13) – Seahawks W (8-9)

Photo by Al Bello/Getty Images

Final NFC Playoff Seeding

NFC East Champion – Eagles (15-2), 1^st seed

NFC West Champion – 49ers (13-4), 2^nd seed – Conference record tiebreak over Minnesota

NFC North Champion – Vikings (13-4), 3^rd seed

NFC South Champion – Buccaneers (8-9), 4^th seed

Wild Card 5^th seed – Cowboys (13-4)

Wild Card 6^th seed – Lions (10-7)

Wild Card 7^th seed – Giants (9-7-1)

Like last week, the Eagles once again clinch the first seed, earning a bye in the Wild Card round and home field advantage throughout the playoffs. The 49ers edge out the Vikings for the second seed via the conference record tiebreak (10-2 vs 8-4), since they have not played head-to-head. The 8-9 Buccaneers win the NFC South, to claim the fourth seed as the division leader with the worst record.

The Wild Card seeds are determined by Win-Loss record alone, with no need to rely on tiebreaks. The Cowboys win out to finish with a 13-4 record and claim the 5^th playoff seed. Detroit also wins their final three games against a soft schedule of Carolina, Chicago and Green Bay. They finish with a 10-7 record to claim the 6^th Wild Card seed. Thanks largely to the Week 15 victory over the Commanders, the Giants finish with a 9-7-1 record to claim the 7^th and final Wild Card seed.

Having won only 7 games, with three remaining, the Commanders face an uphill battle to finish ahead of either Detroit or the Giants to make it into the post-season. Of the three teams, the Commanders have the toughest remaining schedule, consisting of the 49ers, Browns and Cowboys. New York has a less formidable schedule with remaining games against the Vikings, Colts and Eagles. The Lions have it the easiest, as they do not face a team with a winning record for the rest of the season. At this point in time, it would be very difficult for the Commanders to finish ahead of Detroit (7-7) or tie their win total. If they can manage 10 wins, they are in. They just have to beat the 49ers on the road and Cowboys to pull that off.

The Commanders’ best chance would seem to be overtaking the 8-5-1 Giants. However, if the Giants win one more game, that would require the Commanders to win all three of their remaining games, since the Giants also hold the head-to-head tiebreak.

While the Commanders can still make the playoffs, thwy need to stay ahead of the Lions and Seahawks or overtake the Giants. That is the price they paid for failing to rise to the occasion in the rematch with their division rival on Sunday.

Acknowledgement: Thanks to James Dorsett for his usual expert editorial input.

---