Week 10 DVOA: Are 49ers Doomed to Underachieve?

Week 10 in the NFL was a week full of good teams escaping with close victories. The Lions needed the Texans to miss a 58-yard field goal, then hit a game-winner as time expired. The Chiefs needed a blocked field goal to win. The Ravens needed a missed 2-point conversion.

As a result, we see the same DVOA rankings as we did a week ago, but with lower numbers. The top seven teams in DVOA remain in the same order, but six of those seven teams have dropped in DVOA. The exception, surprisingly, is San Francisco, still fifth with a 5-4 record.

The Detroit Lions remain our No. 1 team for the third straight week, still with a healthy lead over the rest of the league and especially the rest of the NFC. The Lions are no longer one of the “top 10 teams through X games” in DVOA history because of this week’s close game, but they’re right below that. They now come in as the 13th-best team through nine games since 1979.

The Ravens are still the No. 2 team and they still have by far the best offense in the league according to DVOA. Baltimore’s offensive DVOA of 38.6% is nearly twice the rating for the Washington Commanders, who are second on offense at 19.6% DVOA. The Ravens remain one of the top offenses in DVOA history. For this week’s table, I thought it would be interesting to also compare these teams in the area of Baltimore’s obvious weakness, pass defense. The Ravens do have one of the worst pass defenses among these historic offensive teams.

The undefeated Chiefs remain No. 3 in DVOA. The Chiefs have the worst point differential among all the 9-0 teams in NFL history, and they are the fifth-lowest 9-0 team by DVOA. That’s still a list of very good teams, of course. Just because the Chiefs have not played quite as well as their win-loss record does not mean they are not a leading Super Bowl contender! Nonetheless, the Chiefs beat the Broncos despite a PGWE (Post-Game Win Expectancy) of just 20%. Here’s a list of the lowest 9-0 teams since 1979 by DVOA:

Now that we’re 10 games into the season, you’re starting to see a little less movement up and down the rankings each week. That’s what happens when you’ve got a bigger sample size of games in the past. This week’s big movers are the Chargers and Cardinals, who each move up three spots. The Chargers are now 10th and the Cardinals are 12th. Another interesting thing about the DVOA ratings this year is that there are no “fraud” teams. Every team with a winning record has a positive DVOA, and only one team with a winning record (Atlanta) appears below any teams with a losing record.

Do Certain Coaches Outperform PGWE?

I apologize for not having the original tweet to reference here, but when Dennis Allen was fired by the Saints a week ago, I had a reader ask me if Allen had a history of his teams underperforming their Post-Game Win Expectancy (PGWE). After all, the Saints had two games this year where they lost despite high PGWE: Week 4’s 26-24 loss to Atlanta (75% PGWE) and Week 9’s 23-22 loss to Carolina (92% PGWE).

Post-Game Win Expectancy, for those who are new to my stats, is a metric that attempts to figure out “based on the stats of this game, how often would team X have won?” It is based on the stats of the game itself, not any ratings going into the game. It incorporates VOA (without opponent adjustment) along with the number of plays each team has (more plays is better) and the number of penalties each team has (fewer penalties is better). You can find PGWE for every game this year on a page in the Premium Splits workbook that subscribers can download from the DVOA Downloads page.

We’ve compared a team’s wins and losses to their Pythagorean projection before, based on points scored and allowed. Comparing wins and losses to total PGWE is just a more complex version of that. And yes, it does look like Dennis Allen has a fairly consistent record of underperforming his team’s PGWE.

• 2024 Weeks 1-9: 2 wins vs. 3.9 expected (-1.9)
• 2023: 9 wins vs. 10.2 expected (-1.2)
• 2022: 7 wins vs. 8.3 expected (-1.3)
• 2013 Raiders: 4 wins vs. 3.8 expected (+0.2)
• 2012 Raiders: 4 wins vs. 5.0 expected (-1.0)

The real question is how “real” this is. How much is this just random variation, where you will naturally have some coaches with a record of frequent overperformance and other coaches with a record of frequent underperformance? Or is it real when a coach has a long career and commonly overperforms or underperforms the PGWE? We might want to ask this same question about quarterbacks as well.

It turns out there’s a current head coach with a much worse record than Allen when it comes to underperforming PGWE. Although some of the differences are small in certain years, this head coach has underperformed PGWE for eight straight seasons. It’s Kyle Shanahan of the San Francisco 49ers. These numbers include the playoffs:

• 2024 Weeks 1-10: 5 wins vs. 6.0 expected (-1.0)
• 2023: 14 wins vs. 15.0 expected (-1.0)
• 2022: 15 wins vs. 15.3 expected (-0.3)
• 2021: 12 wins vs. 13.2 expected (-1.2)
• 2020: 6 wins vs. 7.7 expected (-1.7)
• 2019: 15 wins vs. 15.4 expected (-0.4)
• 2018: 4 wins vs. 5.7 expected (-1.7)
• 2017: 6 wins vs. 6.1 expected (-0.1)

I’ve computed PGWE vs. actual wins going back to 2012 and taken a look, and there doesn’t seem to be any head coach with a positive track record of overperformance that quite matches what we have here for Kyle Shanahan. There are a few who are positive a lot more than negative, however. That includes Andy Reid. Since Patrick Mahomes became his quarterback, Reid’s only two negative years were at -0.3 compared to expected (2018) and -0.2 compared to expected (2021). Perhaps that’s a suggestion that Kansas City’s ability to win more games than DVOA would usually indicate is something “real.” Another head coach with a very good record over a short time is Mike Vrabel. In each of his first four seasons, the Titans won at least 1.3 more games than PGWE would suggest.

Here are the records of some head coaches and quarterbacks you may be curious about, going from 2012 through Week 10 of this season. “Even year” here is any year between -0.3 and 0.3 when compared to total PGWE.

Andy Reid total: +6.7 (7 positive years, 3 negative years, 3 even years)

Andy Reid since Patrick Mahomes became QB: +6.5 (5 positive years, 2 even years)

Bill Belichick total: +4.5 (7 positive years, 4 negative years, 1 even year)

Tom Brady no matter who his coach was: +5.7 (7 positive years, 2 negative years, 2 even years)

Mike Tomlin total: +6.8 (8 positive years, 2 negative years, 3 even years)

Mike Vrabel total: +5.8 (4 positive years, 1 negative year, 1 even year)

Pete Carroll total: +3.0 (7 positive years, 3 negative years, 2 even years)

Sean McVay total: +2.7 (3 positive years, 2 negative years, 3 even years)

Sean McDermott total: -0.5 (3 positive years, 4 negative years, 1 even year)

Sean Payton total: -1.7 (4 positive years, 6 negative years, 1 even year)

John Harbaugh total: -2.8 (4 positive years, 7 negative years, 2 even years) 

Full Week 10 DVOA Table

These are the FTN DVOA and DAVE ratings through 10 weeks of 2024. Our Defense-adjusted Value Over Average (DVOA) system breaks down every single play and compares a team’s performance to the league average based on situation in order to determine value over average. (Explained further here.) Beginning this week, opponent adjustments are at full strength.

OFFENSE and DEFENSE DVOA are adjusted for performance indoors and consider all fumbles, kept or lost, as equal value. SPECIAL TEAMS DVOA is adjusted for type of stadium (warm, cold, dome, Denver) and week of season. As always, positive numbers represent more points so DEFENSE is better when it is NEGATIVE.

DAVE is a formula which combines our preseason forecast with current DVOA to get a more accurate projection of how a team will play the rest of the season. The projections are adjusted for a few quarterback changes as well as injuries and trades. DAVE is currently 24% preseason forecast and 76% actual performance for teams with 10 games and 36% forecast with 64% actual performance for teams with nine games.