What explains the 49ers 2022 second half of the season offensive explosion? Looking to the film and analytics for clues
In football, context matters. Today we look at the real catalysts of the offensive explosion in the 2022 season that is incorrectly attributed to one player.
One of the more pervasive and persistent narratives of the offseason is the amount of credit given to Brock Purdy for the sudden explosion in points scored by the 49ers offense. From Jimmy Garoppolo to Brock Purdy, the conventional wisdom goes, the offense went from 26 points per game with Garoppolo to 34 points per game with Purdy.
This is, on a surface level, factually true. No one can deny that it isn’t. But this is where the facts end and divergent opinions begin as to what really was the catalyst. It therefore stands to reason that because Purdy was the quarterback during that time frame, then it means he is primarily responsible for this output in efficiency, correct?
One cannot deny the reality that he played well at certain times, made a handful of high level throws, made plays out of structure when the in-structure stuff wasn’t there, and quarterbacked the team during this stretch of the season that had many believing they could unseat the Mahomes-led Chiefs in a potential rematch of Super Bowl 54.
But how much of this credit goes to Purdy and were there other factors set in motion before his arrival in week 13? I would argue that quarterback play mattered less during this stretch, in which the winning streak started in week 8 under Garoppolo, and was more a product of a truly team effort more so than just one player.
In order of importance:
Christian McCaffrey trade and its impact in three key areas: EPA/play, DVOA, receiver yards of separation, yards per play, and quarterbacks targeting the middle of the field.
The defense’s turnaround from forcing 8 turnovers (1.1 per game) before the bye week to forcing 24 turnovers after the bye week (2.1 per game) and never having a game after the bye week where they didn’t register a turnover.
A distant third, quarterback play, in which there just wasn’t a ton of difference between Garoppolo and Purdy.
The Primary Catalyst: The Christian McCaffrey Trade boosts offensive efficiency
EPA per play
One question that everyone has for me is that if Brock wasn’t the primary reason, then why did they go from 26 points per game under Jimmy to 32 points per game under Brock. Surely, "correlation implies causation", right? No. Correlation does not imply causation and the easy answer to the question of why the offense exploded is simple: The primary catalyst for the gains in offensive efficiency can be traced back to the McCaffrey trade.
Looking at the big picture, the 49ers ranked 18th in EPA per play in weeks 1 through 6.
The CMC trade happened at the beginning of week 7 as he was loosely integrated into that game plan versus the Chiefs. Their EPA per play rank after week 7 was 15th.
From weeks 7 through 12 (after the CMC trade but before Purdy took over), the 49ers offense added .061 in in EPA to their season total. From weeks 13 through the divisional round with Purdy, they added .040.
After week 7 (EPA -.001), from weeks 8 through 12 (pre Purdy), the 49ers offense rocketed all the way up to 5th in the NFL in EPA (.129) for a gain of .130. From weeks 13 through the divisional round, the 49ers climbed to 2nd overall in the NFL in EPA at .150, for a gain of just .021 when Purdy took over.
He undoubtedly played well enough to keep winning games. But the efficiency gain was already there, and it occurred primarily when CMC entered the lineup. It’s reasonable to assume the 49ers offense would’ve kept the same efficiency with Garoppolo.
Defense-Adjusted Value Over Average (DVOA)
In Football Outsiders “Defense Adjusted Value Over Average” (DVOA), a metric that seeks assign success on a per play basis, the 49ers experienced their greatest gain in DVOA after the McCaffrey trade, which saw them gain 21.3% more efficiency in play success.
With Brock Purdy inserted as the starter, from week 13 through the divisional round, they gained a total of just 9% more toward their season total.
Furthermore, the offense mostly peaked after week 15 where they reached 30.4% in total offensive DVOA, finishing with 33% after the divisional round.
Receiver yards of separation
On the field, one area directly contributing to the efficiency gains were receiver yards of separation, an NFL NextGenStats metric that measures yards of separation to the nearest defender at the time of the throw. The table below shows the pre and post-McCaffrey trade yards of separation and his impact on the offense.
Receiver yards of separation per game pre-McCaffrey trade:
Deebo: 3.4
Aiyuk: 2.5
Kittle: 2.8
Receiver yards of separation per game post-McCaffrey trade:
Deebo: 3.7 (+0.3)
Aiyuk: 3.3 (+0.8)
Kittle: 3.6 (+0.8)
While Deebo experienced minimal gain in separation (his yards of separation are usually generally higher than the rest of the receivers anyways with his usage in the screen game), Aiyuk and Kittle saw their yards of separation go up by nearly one yard post-CMC trade.
In the cut-up above, the 49ers and Jimmy Garoppolo struggled throwing the ball over the middle of the field as they saw their bread and butter pass concepts erased by split safety coverage shells and more defenders zone dropping into their preferred throwing lanes between the numbers.
Coverages adapt to the 49er offense and yards per play
Also, pre trade, the offense was averaging 5.6 yards per play, a very respectable 10th in the NFL but those include at least two games where the 49ers gained yards on chunk plays in garbage time versus Atlanta. Post trade, the offense gained a half of a yard in this metric up to 6.1 yards per play, an average that was suppressed by a paltry 4.7 yards per play versus a top 5 defense in week 12 versus the Saints. Still, gaining a half yard in this metric is a huge boost.
The plays above shows why it was the case that both the receivers gained that separation yardage, why Garoppolo’s pass attempts went up in the middle of the field, and why the offense got a boost in the YPA metric. You can see the effect McCaffrey has on the offense with his ability to draw extra defenders to him in space, opening up the rest of the field for everyone else.
Per Sports Info Solutions, with Garoppolo under center, the 49ers offense faced split safety coverages on 45% of their passing snaps and 55% single high safety coverages. Post-McCaffrey trade, the number of single high safety coverages rose to 62% pre-week 13.
From week 13 until the end of the season, Brock Purdy saw a league high 71% single high safety coverages in the pass game. With more defenders in the box, more of those throwing lanes opened up. And it was a slog early in the season with teams largely sitting in the middle of the field taking away the bread and butter routes in the passing offense.
Per Pro Football Focus, for Garoppolo, pre CMC trade, his percentage of passes in the 0-9 yards downfield range was 42% of his pass attempts. Post trade, his percentage of pass attempts from 0-9 yards downfield went up to 50%.
His percentage at the 10-19 yard range dropped from 20% to 19%. However, his target share over the middle of the field from 0-19 yards went up by a full pass attempt. 5.6 per game to 6.7.
The points per game narrative
The 49ers offense was pretty visually stale from weeks 1 through 6 despite a handful of explosive plays. If you remove the 2 defensive touchdowns from the overall point total, the offense averaged a low 18 points per game during that 3-3 stretch. One game was played in a monsoon, another against a top 10 defense (Denver), but the 49ers offense seemed to save it’s worst output for the worst teams. 14 points against the Falcons just wouldn’t cut it and the Chiefs were next on the schedule.
Week 7 was McCaffrey’s first game in a 49ers uniform but he was used sparingly after getting into Santa Clara mid week and getting down a handful of plays in practice. The offense still didn’t perform much better even after getting off to a hot 10-0 start in the first quarter, scoring just 13 the rest of the game.
The offensive explosion happened immediately after that. The offense went on the road and put up 31 points on the Rams, just 22 against the Chargers (could’ve been more but receiver Brandon Aiyuk dropped a touchdown pass and fumbled on a 20 yard catch and run among several other player execution issues), and 38 on the road in Arizona in a Monday Night game in Mexico City, bringing the 49ers offense from 18 points per game to 29 points per game.
With Purdy under center, the offensive points per game went up to 32, a sizable 4 point gain, but there are other underlying reasons that directly contributed to this point total, which will be the subject of the next section.
Defensive takeaways give the offense more chances to score
The 49ers defense, while having a top 5 defense all season, was not generating the rate of turnovers in the first half of the season that they were in the 2nd half of the season. From weeks 1-8 (bye week in week 9), the defense generated just 9 turnovers, 1.1 turnovers per game.
In the second half of the season after the bye week, they generated 24 turnovers, 2.1 turnovers per game from week 10 through the divisional round. In Football Outsiders “turnovers per drive” metric, the defense went from 19th at the bye week to 2nd by the end of the season. That’s quite the turnaround.
In the 11 games that Lance and Garoppolo were under center, the defense was only able to give the offense a short field (opponent side of the 50 yard line) on five turnovers. Three of them came when the score was 7 points or less and one interception was returned for a pick-6. The offense was able to generate 17 points on those 3 drives where they started on the opponent side of the 50 yard line.
While Purdy was under center, the defense returned one fumble for a touchdown and gave the offense seven more drives where they started with shortened fields in the 8 games Purdy started or played. They scored 39 points on 7 short field drives when the score was within 10 points or less. They scored 66 points total on shortened fields from turnovers while Purdy was under center.
On all those turnovers on the opponent’s side of the 50, the offense started with the ball at the following yard lines: 25, 38, 6, 11, 23, 34, 42, 3, 18, 34, 28, 37, 21. Every offensive drive listed here resulted in some form of points, either a touchdown or a field goal. If you give Shanahan a short field with this much talent, it’s guaranteed to result in points no matter who the quarterback is.
The usual rebuttal to this is that “well yeah, with Purdy running the offense, opposing teams were forced to get more aggressive on offense and take chances.” On the surface, this seems plausible. The usual trope is when the 49ers offense scores, teams are forced to get aggressive and they make mistakes.
That’s not the case here. As I already showed above, 8 turnovers occurred in a game where the score differential was just 10 points or less, not considered a scenario where any team would panic. Ten point leads can be erased in a matter of minutes (Super Bowl 54?).
It sounds cliche, but giving a high powered offense a short field would generate more points than if they had started with the full field. In two of the games, versus Washington in week 16 and Arizona in week 18, the offense scored 36 total points on short fields (19 versus WAS and 17 vs AZ).
Those totals alone are enough to inflate the points per game metric frequently cited. With proper context, it’s easier to see why it’s inflated and truly points to a total team effort. Does this mean Purdy wasn’t good? Absolutely not. The fact that the offense could score that much on short fields points to a healthy, functional offense that Purdy was effortlessly able to execute regardless of the field position. However, the defensive metrics help add a layer of context that just isn’t discussed enough.
Were the quarterbacks contributing anything at all?
Statistically, there is almost no difference between the two quarterbacks last season. I do think Purdy offered a little more upside in terms of out of structure play-making ability but outside of that, both quarterbacks’ on-field play was nearly identical in both film and stats.
With recency bias a huge factor here though, I think fans tend to forget that Jimmy made a ton of good throws last season, both in and out of structure, both from clean pockets and under pressure, just like Purdy did. It’s not unreasonable to think that the efficiency would have continued to climb as much as did when Purdy took over. And I am far from Garoppolo’s biggest fan. I’ve probably been one of his biggest critics. Still, their play was nearly identical.
Garoppolo and Purdy both led the league in NextGenStats “completion percentage over expected (CPOE),” a model that takes into account difficulty of throw and probability of a completion based on distance the ball traveled in the air and separation of the wide receiver.
A CPOE greater than the raw completion percentage, as in Garoppolo and Purdy’s case, means they are performing under expectations while getting the highest amount of open throwing lanes.
Garoppolo, with a comp% of 67.2 and a CPOE of 69.3 (-2.1%) and Purdy, with a 67.1 comp% and a 69 CPOE (-2%) were not hitting as many open throws as they should have been hitting in this offense.
Depending on who’s metric you look at, the results are not favorable for either quarterback. Per RBSDM, Purdy has a CPOE of +0.5 (68.8 comp% to 68.3 CPOE) for 21st in the NFL and Garoppolo has a CPOE of -0.7 (67.9 comp% to 68.6 CPOE) for 29th in the NFL.
Some of the league’s best quarterbacks have been in the top 5 or top 10 in this category since 2019 like Patrick Mahomes (2 super bowl wins), Matt Stafford (1 super bowl win) and Tom Brady (1 super bowl win in that time) but no one is mistaking Purdy and Garoppolo for those quarterbacks, neither of which have shown they are capable of hitting throws at all levels consistently when needed with elite anticipation and timing in tight coverage.
Neither quarterback can regulary or consistently hit the throws in the above Matthew Stafford cut-up. Yet Garoppolo’s “aggressiveness%” (the amount of throws that NextGenStats considers into tight coverage with a defender at least 1 yard or less away from the intended receiver) was at 14% for the games he played in 2022. Brock Purdy was second from the bottom in the league with just 9.4% of his throws into tight coverage.
Outlook
The quarterback’s role in this offense, while not making them the central focus, is still important and it still takes a certain level of competency to run what is otherwise considered one of the league’s most complicated schemes. The obvious evidence for this is the lack of success others like Brian Hoyer, Nick Mullens, and CJ Beathard had. Yes, the rosters were relatively weaker than it is now, but those quarterbacks were not capable of making some of the required throws to be successful like Garoppolo or Purdy could make.
Where I think Purdy edges out Garoppolo though, is the ability to extend plays. He’s just a more naturally athletic player than Jimmy. Purdy was even told by his coaches that he went off script too much and didn’t take the plays that were there for him within structure. Purdy admitted to it as well. Even if that’s the case, he was able to avoid sacks at the very least that Jimmy wouldn’t have been able to avoid.
But statistically, their numbers were nearly identical with Purdy having the advantage of coming in at a time when the defense was hitting its peak and the offense was seamlessly integrating yet another elite offensive weapon, so much so that it made his job 10 times easier. The issue with all of this is, what is this offense going to look like again when the defense doesn’t generate the favorable turnovers or teams figure out ways to scheme against the interior of the field (Seattle laid a blueprint), or, knock on wood, the offense loses 1 or 2 key skill players?
Either way you slice it, neither quarterback was contributing as much as they were perceived to be contributing to the overall offensive efficiency gains. However, and this shouldn’t be taken for granted, it does require competent quarterback play. I do not think that Nick Mullens, CJ Beathard, or Brian Hoyer could run the offense to the level Jimmy and Brock did even with the talent around them right now.
I’m hesitant to put Lance in there because we have not seen him play in consecutive games but even in the ones he did play, he looked supremely more talented than the previous backups prior to him. But Purdy does edge him out in experience. Lance is going to require time to develop and that is something many feel this roster doesn’t have. Purdy makes the most sense right now from a 49ers front office perspective.
However, there is no guarantee that even with Purdy, the offense continues this level of output. There were other underlying factors that contributed to the second half explosion that have largely gone unexamined due to the nature of the offseason narratives surrounding the quarterback position. Hopefully this helps add more context and shows how truly dominate the entire team was in 2022.