After weeks of pulverizing opponents into submission, the San Francisco 49ers defense suddenly sprung a leak. Until the Thursday night game against the Arizona Cardinals three weeks ago, San Francisco hadn’t given up more than 20 points all season. As recently as Halloween, Football Outsiders’ numbers put them in generational company as an all-time great unit. Now, after three straight games allowing 24 or more including the team’s first defeat, questions arise about the long-term viability of this group as a dominant force.
FiveThirtyEight’s Josh Hermsmeyer wrote about the variability of defense last January before the Super Bowl, pointing to its volatility and unreliability as a predictive metric. Hermsmeyer believes this stems, in particular, to our poor mainstream stats related to measuring the true quality of a defense.
A week ago, The Athletic’s Ben Baldwin took this idea and framed it through the lens of the 49ers. He found that the success of a defense in the first half of the season doesn’t have a strong correlation to how the defense plays in the latter half of the season. This fits with Hermsmeyer’s work about the unpredictability of defense over time.
Opponent quality drives much of what we see with defensive success. This presents an intuitive solution: playing bad offenses tend to make a defense look better. To wit, here’s a list of the offensive DVOA of the teams the 49ers have faced this season:
In other words, the 49ers have faced three top-half offenses and one elite offense. They lost the game against the elite offense and faced their toughest tests against the two other best offenses they faced, including having to come back against the Cardinals twice. Though Russell Wilson didn’t exactly light up the San Francisco defense, getting a defensive touchdown from Jadeveon Clowney and benefitting from a pair of short fields off turnovers, that’s precisely the point.
By its literal definition, DVOA accounts for this. It’s the reason the stat exists. So those teams are bad relative to average, not just bad because of how they fared against the 49ers. On the other hand, DVOA also factors in quality of opponent in San Fran’s No. 2 defensive DVOA ranking, head and shoulders clear of the Steelers at No. 3. Their offensive schedule faced is predictably last in the league, but they destroyed those bad teams.
That matters too.
At this point, we should include the 49ers offense in this discussion, particularly because of how the quarterback has played this season. Against bad teams, one’s mistakes cost less. Turn the ball over against Washington, and it likely won’t be nearly as injurious as doing so against Seattle or Green Bay. In other words, it’s not just what the 49ers defense does, but what the offense does as well. Jimmy Garoppolo threw two picks against the Cardinals and was utterly useless most of the game against the Seahawks, with a pick and three or four would-be interceptions bouncing off the helpless arms of defenders.
As Hermsmeyer notes in his piece, turnovers — specifically winning the turnover battle —provides a clear path to winning, another intuitive notion supported by data. It should likewise be obvious the delta between turning over a bad quarterback compared to a good one. The Buccaneers boast an explosive offense and they were able to get yards on San Francisco, but Jameis Winston will turn the ball over. Here is the DVOA of the quarterbacks the 49ers have faced this season:
This tells us something insightful about the 49ers defense. Sure, the statistics look nice when they dominate teams all season, but those same quarterbacks have been bad against other opponents as well. Robert Saleh’s group previously faced just one top-half quarterback by DVOA this season and they lost that game. Rodgers comes into Sunday’s primetime tilt 8th in DVOA, the only quarterback the 49ers have faced even close to keeping company with Wilson.
This should tell us something about the Packers’ defense as well, having faced the toughest slate of offenses in the league to date. On the other hand, the 49ers come into the game 17th (with statistics not yet available to include the Cardinals game Sunday), presenting a turnover and sack-hungry defense an opportunity to turn the tables.
Garoppolo’s penchant for turning the ball over presents a bigger problem this week against a top-10 Packers offense and quarterback. Even if the Green Bay defense plays an average game for them and an average game against the 49ers, it will mean at least one Garoppolo turnover and certainly more than one opportunity. Those marginal differences matter more against a good offense like the Packers possess.
The loss of Kwon Alexander, the ultra-athletic, highly-paid offseason linebacker acquisition likely contributes to this downward trend, another important bit of evidence that defensive ability stems more from talent than scheme, unlike offense which can be more effectively game-planned with inferior talent. But Alexander won’t be back this season, which means this is the version of the 49ers that the Packers and their other opponents will face this season.
As Baldwin wrote before the Seahawks game, we likely won’t know how good this 49ers defense truly is until after they’ve played good offenses. Now Rodgers is on deck before matchups with Drew Brees, Lamar Jackson, Russell Wilson, Matt Ryan and Jared Goff. If the Niners finish the year the way they started, we’ll know this defense truly belongs in such elite company. If not, we can point to this stretch as the touchstone moment when we realized this team revealed itself to be some level of paper tiger.
Just how big? That will be up to Rodgers starting this weekend.