[et_pb_section admin_label=”section”]
[et_pb_row admin_label=”row”]
[et_pb_column type=”4_4″][et_pb_text admin_label=”Text”]
Now that the draft is well and truly in the books, we are at a dead point in the season. Yes, there are some storylines worth following in terms of OTAs and press conferences (which the TDU crew are eating up), but I was getting a bit sick of seeing rankings like ‘Who is the biggest baby daddy in the NFL?’ given Tyreek Hill’s exploits in the last couple of years, so I decided to delve into some PFF data on Justin Herbert, find some interesting comparisons, and see if I can make some justifications for what we’ve seen from a drafting and front office perspective, and also some predictions on what we might expect from the big signal caller in the upcoming season.
I am using Herbert’s PFF QB metrics in this article to provide insight and predictions. I am also using some PFF 2012 San Francisco data to help shape ideas and perspectives. I would love for people in the comments to add other perspectives or ideas based on my analysis (and yes, like most articles these days, I have put a ‘key takeaways’ section at the end so you do not have to read through it all).
Play Action Herbert?
I think the biggest question I have about Herbert fitting into a Roman scheme was ensuring that Herbert’s natural ability as a passer is not downplayed, or not utilized. One of Roman’s calling cards at San Francisco was the effective play-action game, so that’s where I started my research – is play-action actually a strength of Herbert’s? The results were interesting.
Herbert has shown considerable growth in his ability to be more comfortable in the play-action game, even with changes of the coaches and coordinators around him.
2023: Completion Rate: 70.2%, YPA: 6.6, Passer Rating: 110.8
2022: Completion Rate: 63.5%, YPA: 7.0, Passer Rating: 87.4
2021: Completion Rate: 65.9%, YPA: 7.5, Passer Rating: 97.7
As a bigger QB, Herbert has always shown he has the athleticism to incorporate play-action into his game. Herbert’s performance with play action in 2023 shows a notable improvement in completion rate and passer rating compared to previous years. Despite a slightly lower YPA, the higher completion rate and passer rating suggest better efficiency and decision-making under play action.
But what does this mean? Many of us are prognosticating and hypothesizing if a Roman-led offense can utilize Herbert’s ability as a passer given how run-centric Roman’s previous offenses have been. So, I was able to grab some data from 2012 Colin Kaepernick/San Francisco’s Super Bowl run to compare with Herbert’s 2023 data. The comparison is fascinating, and perhaps gives some credence to what a Herbert/Roman pairing might look like:
Completion Percentage: Herbert has a significantly higher completion percentage on play-action passes (70.2%) compared to Kaepernick (60.2%).
Yards per Attempt: Kaepernick averages more yards per attempt (8.7) than Herbert (6.6), indicating deeper throws in play-action scenarios.
Touchdowns to Interceptions Ratio: Herbert’s TD to INT ratio (9:1) is better than Kaepernick’s (4:3), showing better decision-making or execution.
Turnover-worthy Plays: Kaepernick has a higher percentage of turnover-worthy plays (6.8%) compared to Herbert (1.0%).
Average Depth of Target: Kaepernick has a higher ADOT (13.7) compared to Herbert (6.1), again indicating deeper throws.
Adjusted Completion Percentage: Herbert has a slightly higher adjusted completion percentage (77.8%) compared to Kaepernick (73.4%).
Dropped Passes: Herbert’s receivers dropped fewer passes (4) compared to Kaepernick’s (5), but Herbert’s drop percentage (4.8%) is lower than Kaepernick’s (8.6%).
Time to Throw: Kaepernick takes more time to throw (3.61 seconds) compared to Herbert (2.93 seconds).
NFL Rating: Herbert has a higher NFL rating (110.8) compared to Kaepernick (89.3).
In comparing their play-action passing, Justin Herbert in 2023 demonstrates higher efficiency with a 70.2% completion rate and a stellar 9:1 touchdown-to-interception ratio, significantly outperforming Colin Kaepernick’s 60.2% completion rate and 4:3 ratio from 2012. Kaepernick, however, capitalized on deeper throws, averaging 8.7 yards per attempt compared to Herbert’s 6.6, reflecting a more aggressive downfield approach. Interestingly, while Herbert’s quicker release (2.93 seconds vs. Kaepernick’s 3.61) contributes to his higher efficiency, Kaepernick’s higher-risk strategy yielded substantial yardage despite a greater propensity for turnovers.
For Justin Herbert in 2024, his efficient play-action passing from 2023 suggests a strong foundation for continued success if he maintains or further improves his performance metrics. The high completion rate and favourable touchdown-to-interception ratio indicate that he can consistently make smart, accurate throws, which bodes well for sustaining offensive drives and scoring opportunities. If Herbert can combine his efficiency with a slightly more aggressive downfield approach, similar to Kaepernick’s higher yards per attempt, it could enhance his overall effectiveness and make the Chargers’ offence more dynamic and difficult to defend. Herbert is a different (I believe far more talented) quarterback than Kaepernick or any other passer Harbaugh or Roman has worked with, so I think the sky’s the limit both figuratively and literally!
*Play-action was used 28.2% of the time in Kaepernick’s 2012 season, compared with 25.2% for Herbert in 2023, so it is not a huge leap in terms of what Herbert will be asked to do.
How do Alt and McConkey help Herbert?
The next logical step in my analysis was to then look for trends in data to make sense of Hortiz’s first two draft picks Joe Alt and Ladd McConkey. Naturally, I investigated offensive line protection and how that impacted Herbert’s data.
When Herbert is kept clean:
2023: Completion Rate: 71.8%, YPA: 7.2, Passer Rating: 106.4, PFF Offensive Grade: 89.8
2022: Completion Rate: 67.7%, YPA: 6.8, Passer Rating: 92.7, PFF Offensive Grade: 77.9
2021: Completion Rate: 65.9%, YPA: 7.5, Passer Rating: 97.7, PFF Offensive Grade: 90.0
Herbert’s performance when kept clean improved in 2023, with the highest completion rate, passer rating, and PFF offensive grade compared to previous years. Herbert is an efficient quarterback. Harbaugh and Hortiz clearly understand this. These numbers highlight the critical importance of offensive line protection for not only maintaining Herbert’s high efficiency but also matching who Herbert is. Whilst Herbert can uncork the deep ball, it is his efficiency in moving the ball down the field that will win games. He is a naturally conservative QB who wants to make good decisions (in 2023 Herbert was tied for 5th in the league with a TWP% of only 2.4 alongside Mahomes and Goff). With both Slater and Alt as anchors on the line, this only plays into Herbert’s natural personality as a quarterback. Perhaps this also means Herbert’s average time to throw (2.93 seconds) might increase, though it won’t look like 2012 Kaepernick (3.61 seconds), opening the field on deep shots not by scrambling as Kaepernick liked to do, but with solid blocking up front. The more I look at the Joe Alt pick, the more it makes sense (and the more excited I get and I am not even covering the run game in this post).
But how does Ladd McConkey work? And why trade UP for him? (And, why did I scream in joy and scare my baby daughter when the pick came in?). Again, I feel like Harbaugh and Hortiz want to best support who Herbert is as a passer. Yes, the Bolts lost Keenan Allen and Mike Williams and the receiver room looked bare, but why McConkey? Why take Alt at #5 and not Odunze!!??? (Keep in mind how the Alt pick and offensive protection begins to play out as I go through these numbers, too).
In 2023, Justin Herbert demonstrated improved efficiency in both short and medium passes, achieving higher completion rates and YPA compared to the previous two seasons. The increased completion rate for short passes (76.4% from 75.1% in 2022 and 74.3% in 2021) and the rise in YPA for medium passes (9.4 from 9.1 in 2022 and 8.8 in 2021) indicate growth and improvement, reflecting better precision and decision-making, allowing the offence to sustain drives and create scoring opportunities.
2023:
Short Passes (0-9 yards): Completion Rate: 76.4%, YPA: 6.3
Medium Passes (10-19 yards): Completion Rate: 51.8%, YPA: 9.4
Deep Passes (20+ yards): Completion Rate: 36.4%, YPA: 11.2
2022:
Short Passes: Completion Rate: 75.1%, YPA: 5.9
Medium Passes: Completion Rate: 50.7%, YPA: 9.1
Deep Passes: Completion Rate: 34.7%, YPA: 10.9
2021:
Short Passes: Completion Rate: 74.3%, YPA: 6.1
Medium Passes: Completion Rate: 49.8%, YPA: 8.8
Deep Passes: Completion Rate: 35.2%, YPA: 11.0
Again, looking at the 2012 San Francisco Roman-led offensive data, we begin to see how and where Roman might want to improve some of Herbert’s numbers by using McConkey.
Short Passes:
Herbert 2023: Completion Rate: 76.4%, YPA: 6.3
San Francisco 2012: Completion Rate: 75.3%, Y/REC: 10.6
Herbert’s short pass efficiency in 2023 is slightly better than the 2012 San Francisco team in terms of completion rate. The Y/REC of 10.6 in 2012 highlights the importance of YAC, an area where McConkey (and hopefully QJ!) can excel, potentially increasing Herbert’s YPA.
Medium Passes:
Herbert 2023: Completion Rate: 51.8%, YPA: 9.4
San Francisco 2012: Completion Rate: 63.2%, Y/REC: 19.6
San Francisco’s 2012 team had a higher completion rate and Y/REC for medium passes. McConkey’s ability to create separation and gain YAC can help improve Herbert’s medium pass stats, closing the gap between these two metrics. I also think an improved Palmer will help in this aspect of Herbert’s game.
Deep Passes:
Herbert 2023: Completion Rate: 36.4%, YPA: 11.2
San Francisco 2012: Completion Rate: 50.0%, Y/REC: 34.0
While Herbert’s deep pass completion rate in 2023 is lower (MUCH lower) than San Francisco’s 2012 team, his YPA shows potential, and he clearly has the capability. McConkey’s deep threat capability alone is not going to improve this markedly, but the additions of certain FAs like Chark, and the utilization of Johnston, Palmer and Davis might have something to do with improving this area of Herbert’s game. This data also suggests that while a Roman-designed offence is perhaps more ‘conservative’, the 2012 San Francisco data shows there still a focus on big-playability and a want to capitalise on it!
Incorporating Ladd McConkey into the Chargers’ offence can amplify Justin Herbert’s strengths in the short and medium passing game. The comparisons with the 2012 San Francisco Super Bowl team underscore McConkey’s potential impact, and why Hortiz traded up for him. McConkey’s agility and route-running can enhance Herbert’s completion rates and YPA, particularly in short and medium passes. His skills in gaining YAC and creating mismatches will support a balanced attack, ensuring Herbert has reliable targets to sustain drives and maximize offensive efficiency. As the Chargers shift to a run-first offence, McConkey’s presence can help keep defences honest underneath, ensuring the opposition can’t stack the box on early downs, and also enhance Herbert’s abilities in the play-action game.
My key takeaways:
- Herbert’s play-action performance in 2023 shows significant improvement, with a 70.2% completion rate and a 9:1 touchdown-to-interception ratio, indicating his suitability for a Roman-led offence.
- The addition of Joe Alt strengthens the offensive line, which is crucial for maintaining Herbert’s high efficiency when kept clean, as seen in his improved 2023 metrics.
- Herbert’s improved efficiency in short and medium passes in 2023 can be further enhanced by Ladd McConkey’s agility and route-running, increasing YAC and overall offensive effectiveness.
- Herbert’s deep passing game shows potential with a YPA of 11.2 in 2023. McConkey’s addition, along with other receiving options, can help capitalize on this potential but it is the biggest area of focus for Herbert in 2024.
- And through all of this, despite 2023 being perceived as a ‘down year,’ Herbert’s metrics indicate growth and improvement across various aspects of his game. I beleive Joe Alt and Ladd McConkey directly lead to a continued growth in certain metrics for Herbert, which is a scary though for opponents. The guy just keeps getting better and better!
In conclusion, while I may have screamed with joy and scared my baby daughter when the Chargers traded up for McConkey, it’s clear that the front office knows how to make Herbert-friendly decisions better than I know how to handle draft day excitement. If this season doesn’t prove that, then at least we can all agree that hopefully, Herbert’s efficiency makes for a smoother rollercoaster ride than my parenting skills. Here’s to hoping my analysis is as spot on as Herbert’s play-action passing—because if not, I’ll be the one needing protection.
Another banging article, mate!
As you point out, some fans would argue Herbert regressed slightly in 2023 (v 2021-2022, and I’m sure for certain metrics this may be true), but I think your article beautifully made the case that Herbert also showed continued development in other areas last season. The future is very bright indeed.
And never forget…Harbaugh’s a great coach and family man despite all his theatrics on the sideline. I’m sure baby M will forgive your Draft Day enthusiasm and would never expect Dad to restrain himself in matters pertaining to the Powder Blues 😅
Haha indeed, sir! Yeah I am not saying that Herbert didn’t regress in some metrics, but on the whole there is always a push toward growth whenever I delve into most of the stats. Incredible given the changing schemes coordinators, and just general f-around of the last 4 years. Incredible!
Love the in depth article!! Will share it on twitter! #Boltup!
Thank you sir! And thank you for sharing! I’ll keep em’ comin’ if you keep findin’ them interesting 
Loving these in-depth articles. As Hortiz continues to build a stronger team around Herbert I believe that the numbers will only improve.
One thing you didn’t account for in that 3-year lookback was the two offensive coordinators and two new systems that Herbert had to learn. Roman will be his fourth offensive coordinator and fourth system in 5-years. No way that he can show much improvement if that continues.
Thank you for the kind words.
In terms of the coordinators changing and schemes shifting, it is actually incredible to think that Herbert has improved in SPITE of all the calamitous decisions made around him. Now not all metrics are big improvements, and I’ve cherry picked certain data points to make my analysis worthy of thought. However, it is incredible he has not had some major regressive episode, or just said ‘trade me’ 😂😂😂
Keep em coming Jack….nice article. I’ve seen a few recent articles dinging Herbie because of lack of postseason success. I think that’s going to change real soon.
@buck-melanoma Thank you for your kind words. I have a positive approach to most things, both a blessing and a curse. This article is definitely more positive. As you state from other articles, and to put it bluntly, Herbert has not won in the post season therefore he cannot be classed a ‘excellent’ QB. I get that. But those articles are simplistic. Winning is more than just one player, however the QB is incredibly vital to winning in the NFL. I like to see positives out of the negatives, and for those people dinging Herbert for a ‘lack of something’, I choose to see growth. I am a teacher, so perhaps that clouds my judgement haha.
Looking forward to more interactions, Buck!
Great article Jack – Where has all this amazing content been hiding? Great to see you find a second outlet for your talents! A few points:
- Comparing Herbert to Kap at first seemed to me like comparing chicken salad with chicken shit. The metrics were pretty eye-opening. If Harbaugh/Roman can make chicken salad out of Kap, it’s very exciting to think what they can do with Herbert
- The Chargers likely have the best OT group in the NFL: Slater (Elite!), Alt (Elite?), Salyer (above average), Pipkins (average). That is a ton of talent with quality depth
- My concern with the OL is the interior:
- Bozeman is an upgrade over Clapp (who I thought played OK, but was shite at communicating, making everyone around him worse). At best, he’s an average (but physical) center
- I keep waiting for Zion to turn the corner. You could make an argument that he was actually worse in 2023 vs. 2022. This is the year we will see if quality coaching can turn him around
- I still think Salyer’s best position is tackle. I will not be convinced that he can play the interior until I see it.
- As Kyle has pointed out, Pipkins strength is his agility. His lack of strength will limit his success in the interior
- I thought McFadden looked decent in his limited snaps
- I thought Jaimes was OK at center. His play was similar to Clapp, but he was a much better communicator. Being a converted tackle gives him a ton of position flexibility, giving him a shot at a backup role.
Overall, that’s a whole bunch of wishing and hoping in the interior. Hoping that Ben Herbert can physically get this group ready to play & that Harbaugh/Roman have a plan (but hope is not a plan).
Its been a minute since the Chargers have had a WR group with true YAC ability. The Telesco era had a handful of ST players who they could engineer a few gadget plays for. McConkey being able to actually play the WR position AND be a YAC threat is something Herbert has never had.
Thanks, Kev! Yes, it is something that I’ve been wanting to try my hand at for a while. I have been practicing writing professionally about education on LinkedIn to develop my voice and skills, so now transitioning that into the Chargers sphere. It is also a good way to keep my football mind going as we take a break from TDU in the offseason.
Some of my thoughts responding to yours:
- Haha yeah, the comparison to Kaepernick was one of the more logical places to start. I did flirt with looking at Lamar’s MVP season under Roman, but I just thought that Lamar was SUCH a different QB to Herbert. Kap is vastly different to Herb, but not as much as I thought Lamar was.
- The concern over the interior is valid. However, we have never seen an ‘angry’ interior. What I mean by that is a focus on gap/power that Roman is known to run. I’d love to see what Johnson is like running down hill on pulls, getting out and using his big frame.
- Who knows what quality coaching and a clear identity on the offensive line might mean for some of our later round interior guys – McFadden could easily become a starter as you say.
- Salyer is a tackle. I agree with that. He has agility for a man his size. Could we see certain plays where we have SIX lineman on the field? Alt lined up in the TE position? I’d love to see it.
- The YAC stuff is purely hypothesis, but trading UP to take a guy who enjoys getting those extra yards after catch like McConkey (I’ll miss Keenan Allen, but I won’t necessarily miss him going to ground when contact was coming even though it was to preserve his body), and having QJ on the roster surely means Roman is aiming to get those guys running routes that enable and champion their abilities, rather than just doing shitty gadget-type stuff and passing laterally, totally communicating to the opposition what was happening.
All a wait and see…:)
@wrefordreed89 Lamar would be interesting. As an aside, I think Jackson is a great regular-season QB that can win games with his freak athleticism. I don’t think he’s good enough at passing to win in the post-season. Again, just my personal opinion, but I think he’s holding the Ravens back. Will be interesting to see how Herbert performs in a similar offense (my admittedly biased opinion is that Justin is going to be much better).
Love the point on having an “angry” interior. The more I think about it, the more upset I am about supporting Staley. The Harbaugh staff is so much better. Can’t wait to see how that translates to on-field performance.
@kevdiego I did a whole research reply for you comparing Lamar to Herbert, but it didn’t save!!!
Good stuff. If people who believe that Roman was brought in to stifle was JH10 does best were to read this, they might change their minds.
@mongotesla indeed. I still have my reservations, but watching a few Roman interviews lately from his time during and after his Baltimore tenure has been illuminating. The way he speaks about QBs specifically, one quote from him I watched last night was ‘a signal caller who can stand in the pocket and deliver will ALWAYS be of value in this league’ (when asked about mobile and dual threat QBs). He goes on to state that that ‘style’ of QB play means you ‘win consistently’. Roman has worked with Luck, Kap, Lamar, Smith, and to think that out of all those player Alex Smith was probably the most ‘pure passer’. Luck, of course, was the best throwing mobile QB of the bunch, and Lamar is Lamar. So to think that Roman will be working with perhaps the best ‘arm’ he’s ever worked with SHOULD lead people to think differently about what JH10 will be like going forward.
Yeah, Jack. Each QB that Roman has worked with has a unique way of playing. What this says to me is that Roman can keep his plan in place and make adjustments to make sure that the signal caller doesn’t fade out or become a backseat option. I think that he’s looking to make Herbert shine in the way that only Herbert can do.
Let’s hope the experience comes to the fore with Herbert. Other than Lombardi (who admittedly had a great first year and helped Herbert to the OROY award), Moore and Steichen were relatively inexperienced comparatively. Whisenhunt has experience in the mid-20teens.
Wherever Roman has gone his first number of years have been very strong. Who knows. It is all speculation. I just want to see the first snap tomorrow haha.
Once again I find myself thinking laterally off the back of one of the TDU’s team excellent think pieces! Nothing inspires creative thought like better creative thought! Great work man.
I have a few thoughts I wanted to run by you (and the rest of the forum);
- I am one of those fans who think there’s barely anything Herbert does wrong and his failings are due to the talent around him, but the area of his game I am most critical of is his play action game. He has all the tools you’d want to see in someone to succeed in this concept so I actually think the biggest part of Herbert improving his play action game is his mentality. As we’ve seen with his conservative approach to turning the ball over he doesn’t like to throw off-platform unless the game is on the line and it costs him opportunities. There are times, especially early last season, where he had open guys in PA Flood and he didn’t pull the trigger on any of them because they weren’t high percentage throws. This for me was symptomatic of his lack of confidence overall but like you I looked back on previous years and noticed he doesn’t excel in this area in general despite having the tools to do so. With his preference of QB coach on board and play action becoming a bigger part of his passing share this should be something he gets more practice time to correct.
- I wanted Brock Bowers to be Charger specifically to improve this area of the game but I agree that Ladd is a significant upgrade over Mike W or Keenan. The key is going to be building in options across all three levels, Ladd can fill the intermediate very nicely and I think Hayden Hurst will be the primary short target with his safe hands but the deep area is the biggest question mark, I think Chark can excel there if he finds his way back to his LSU days.
- I made the same observation in my Joe Alt profile that I think having a strong set of agile tackles and prioritizing that over the interior shows that play action on deep pass sets is going to be a bigger feature of Greg Roman’s offense than people realize. Pipkins was solid in most areas but deep drop backs were where his flaws showed the strongest and it actually got Herbert hurt as he had to step up into the interior rush to avoid being sacked (this was the exact play he broke his finger). I am confident this was an area Joe/Jim thought they HAD to improve in order to have a run first offense and simultaneously not waste having a top 5 QB.
Again, appreciate the great content you and the other TDU boys are putting out and I look forward to reading more!
@ryanwatkins excellent points Ryan! I agree that the value of getting his QB coach back can’t be underestimated. I believe strongly that Herbert has suffered developmentally because of the revolving door of coaches and systems that he has had to endure. This shouldn’t be understated. It has been the complete opposite of the environment that you would want to provide a young franchise QB.
It is historically bad to say the least. Heading into his fifth season he has now had to adapt to the following:
- 2- GMs
- 3- Head Coaches
- 4- OCs
- Chronically bad pass protection.
- Chronically bad running game.
- Chronically bad Defense.
The Spanos family has created an absolutely awful environment for Justin Herbert to start his career. Imagine his career trajectory surrounded by stability in coaching & systems. Imagine his career trajectory with even a mediocre running game and mediocre pass protection.
Sorry as I went full SMG mode right there and probably now lead Stormcloud in super negative posts at 1!
It’s not just our Chargers that let him down in those aspects, Oregon gave him a new OC every year but one so Justin has not had any kind of consistency since he was the president of the fishing club at HS!
@ryanwatkins All credit to Jack! He’s definitely inspiring me to do a research piece at some point this offseason.
On your thoughts:
- Herbert’s pretty close to flawless, I agree. There are instances I’ve seen on All-22 where he processes so quickly that arguably he doesn’t give the deeper routes in the progression time to get open before the ball’s already out. Sometimes it can be frustrating when you see on tape that the deeper route has uncovered yet Herbert took the short completion instead. However, without being in the QB room or knowing the progression/play design, it’s tough to critique that element of his game. I also think you’re right that he doesn’t uncork the ball off play action with the regularity you’d expect. But again, it’s hard to know how much of that aesthetic frustration is legitimate. It might be that the conservative call was the correct call on most of those. Live to fight another day.
- I would have done bad things for Brock Bowers to have wound up a Charger. Watching him and other favourites of mine (eg, Hunter Nourzad, Audric Estime) land in the AFC West is the part of draft prep I enjoy least. I want to see more Gruden/Mayock-era esque reaches from our division rivals, hopefully starting in 2025.
- “Nail meet head” on your Joe Alt v Pipkins commentary 👍
Thank you, @ryanwatkins!
Per your first point, Justin Herbert’s performance in play action is competent but highlights areas where improvement could elevate his game. With a completion percentage of 70.2% and 6.3 yards per attempt (YPA), Herbert showcased accuracy but lagged in maximizing yardage. In contrast, Brock Purdy led the pack with a YPA of 10.0 and a completion rate of 77.3%, demonstrating higher efficiency in play action scenarios. Similarly, Josh Allen posted impressive numbers with a YPA of 9.6 and a completion percentage of 70.9%, indicating his ability to convert play action into significant gains. Herbert’s cautious approach was evident in his low turnover-worthy play percentage (TWP%) of 2.4% and just one interception, underscoring his focus on ball security. However, this caution perhaps limited his effectiveness, reflected in his average depth of target (ADOT) of 8.6 yards. In comparison, Geno Smith boasted an ADOT of 9.4 and YPA of 9.1, taking more downfield risks that paid off with higher yardage, and had a TWP% of 1.4%, indicating both aggression and efficiency. I didn’t realize how good Geno was throwing the ball out of play action!
Lamar Jackson, with an ADOT of 9.8 and YPA of 9.9, also illustrated the benefits of a more aggressive play action strategy, with a TWP% of 3.7%, reflecting a higher risk-taking approach. Herbert’s play action touchdowns (6) were modest, especially when compared to quarterbacks like Jared Goff, who had 10 touchdowns and a YPA of 8.7, blending efficiency with scoring ability and a TWP% of 2.7%. I think with a Roman-led offensive scheme, and more security at the head coach position, we will see Herbert’s numbers go up, but I wouldn’t mind seeing his TWP% go up, too. Taking more risks, being more confident, and also better balance with the run game and better protection.
As for the second point, I am not sure if I said McConkey would be an upgrade on Keenan. KA13 was an incredible receiver for the Bolts over many years, but looking at the numbers in 2023 (I know, a Herbert injury season), I think McConkey (and QJ if used properly) SHOULD provide a different avenue to getting yards in the YAC game. (Take into account KA’s age, too. YAC is generally a young man’s game.) In terms of how incredibly safe KA13 was for Herbert, if McConkey came half as close as some of these numbers, then he’s doing extremely well. Here is my analysis:
KA13’s performance in the short passing game during the 2023 season placed him among the top 15 receivers. With 145 targets and a reception rate (REC%) of 81.9%, Allen showcased reliability in catching short passes. He accumulated 190 yards after catch (YAC) with an average of 2.3 yards per reception (YAC/REC). This ranked him lower compared to leaders like Puka Nacua, who had a YAC of 406 yards and a YAC/REC of 6.8, and DJ Moore with a YAC of 292 yards and a YAC/REC of 6.5. Allen’s drop rate (DRP%) of 5.6% was moderate, and his catchable target percentage (CTC%) was 57.1%, indicating he made the most of his opportunities. However, compared to A.J. Brown, who had a CTC% of 76.9% and a higher YAC/REC of 6.2, Allen’s YAC is clearly not league-leading (but that is not his game!). Chris Godwin, with a similar YAC of 204 yards but a higher YAC/REC of 4.2, also demonstrated more effectiveness in maximizing short passes, but again all these receivers I am mentioning are different types.
In the intermediate passing game, Allen’s performance was again consistent. He had 35 receptions from 145 targets, resulting in a REC% of 60.0%. His YAC totaled 96 yards, with a YAC/REC of 2.7, indicating fewer yards gained after the catch compared to others in this category. For instance, CeeDee Lamb led with a YAC of 194 yards and a YAC/REC of 6.1, and Brandon Aiyuk had a YAC of 183 yards and a YAC/REC of 4.8. Allen’s average depth of target (ADOT) in the intermediate range was 15.1 yards, aligning with other receivers, but his ability to generate additional yards after the catch lagged behind leaders like George Pickens, who had a YAC/REC of 7.9 and an ADOT of 14.4 yards.
But as I said, all of these YAC monsters are different types of receivers. I am not suggesting KA13 is George Pickens, Puka Nacua, or DJ Moore. We will absolutely miss Keenan in the short and intermediate game with those reliable hands! Herbert will miss him being consistently open in pressure situations. I will miss him! What I was saying in the article is that I think McConkey and a Roman offense will mean Herbert’s yards will be made in different ways. It is going to have to be considering the talent that walked out of the receiving room.
Love the discourse on here already.
@wrefordreed89 Thank you for providing the numbers to add some better depth to the discourse! KA13 was a dream partner for a young quarterback’s development, you can equate it to Travis Kelce being there for Patrick Mahomes in terms of reliability. You mention the safety net he offered Justin and it’s here that I think we will see the biggest difference, the Chargers of the last few years could live with not reducing the distance to the down marker on first and second down because they had Slayer in his 7/11 mode, Mike Williams 80/20 deep ball work and Herbert who could find them on those got to have it situations no matter what kind of pressure was in his face. We all remember the incredible 19 play drive in the 2021 Week 18 Raiders’ game and other insane performances where the impossible was made possible by that group, but to win and win consistently those heroics shouldn’t be the primary means of attack. The difference will be seen in that Justin should not be having to force the ball to #13 in every single game, targets should be spread out to where the defense’s weaknesses are on both the ground and through the air by reducing the deficit on all three downs.
Converting 3rd down used to be a priority of how the offensive scheme was designed and the success of the team lived and died on that metric; in 2021 the bolts had a top 5 offense with a 3rd down success rate of 45% (3rd highest in the league), in 2022 they had a top 10 offense with a 3rd down success rate of 43% (8th in the league) but in 2023 that figure went down to 38% at 16th in the league. Now I am simplifying of course and there are a lot more components that go into winning football games but that over reliance on a highly volatile metric is what has cost the Chargers success for the past 20 years. It’s time to play football a different way and moving on from Keenan is part of that due to the cold world of cost benefit impact analysis; he simply would have been too expensive to have in the team playing in a less significant role.
At first I struggled with seeing the best receiver in modern Chargers history walk out the door but the more time passes the more I see Harbaugh’s vision for this team and in particular this offense. I can go on about the benefits of having a TEAM focused scheme rather than a STAR focused one but we’d be here all day and frankly I have work I really should be doing lol.
Looking forward to more discourse with you and the rest of the Stormcloud community!
@wrefordreed89 great responses and content happening on this site. Feels completely respectful and thought out! Kudos!!
Play action passing and the running game go hand in hand. If your running game is putrid (Chargers) it only makes play action that much harder to sell. I think that Linebackers having no fear of the running game get healthy drops due to the complete lack of respect for the Chargers running game. The three teams you mention Buffalo, San Francisco, and Seattle had the following PFF running game numbers:
San Francisco – 92.1 (2)
Seattle – 85.3 (11)
Buffalo – 91.3 (4)
Chargers – 69.4 (32)
Hahahaha when you put it like that, no AI data crunching needed.
Chargers were shithouse running the rock last year 😂😂😂
It continues to boggle my mind that Herbert’s numbers are not WORSE given the issues around him. The guy is a freak in terms of producing numbers that are top 10 worthy in almost every category. I just hope to lord that we see Herbert with some good playcalling and support around him. I’m so sick of watching HoF talent QBs fail to have a team built for them given their skill sets.
Loving the discourse. I aim and want to provide discussion/thought pieces rather than ‘tell you how it is’ articles. I’ve always enjoyed reading the opinions of others in blogs and articles. It’s just nice to have so many wonderful people here already. Thank you! 🙏
Ladd is a significant upgrade over Mike W or Keenan
Ladd is a significant upgrade over Keenan? Blasphemy. WTF are you talking about?
He may very well be a “poor man’s Keenan” but that is not at all the same thing as “upgrade over Keenan”. The Chargers will be fortunate to ever have a WR again as good or better than Keenan. That is not something to take lightly.
@tau837 I also did a double-take at that statement, but maybe it’s in the specific context of PA Flood working the out route from the slot (which Ladd could be great at), or other PA concepts where it helps to have Ladd’s speed off the mark on crossing/over routes off boot action.
I would still say Keenan was pretty good at uncovering over the middle of the field myself. But Keenan is a traitor, so…😜
Woah okay taken out of context that looks awful haha. That is not what I meant at all. Slayer is one of the best Chargers of my lifetime and a healthy Mike Williams is a top 20 WR in the league by no means do I under appreciate them just because they’re on other teams now. My comment was in reference to playing with the ball in hands out of the intermediate out route in play action flood, they’d be heading towards the sideline and maybe 2016 KA was a threat there but Ladd is a significant upgrade over those two guys at this current time, in this current side. He’s got some YAC ability and has the footwork to make tight cuts to expose any lane the defense offers him in this situation and this is a different skillset than 2024 slayer or Mike dub offered.
You can read through my lack of contextual writing already, we’ll get along just fine 😂
😂
Keenan was a really fun YAC dude early in his career (that hurdle over someone for a touchdown his rookie year was legendary) but he definitely lost that wrinkle over time.
I honestly think it was more of a business decision though. Almost like LT running out of bounds when he was close to the sideline, Allen seemed more content going down with contact later on, and it worked, he definitely stayed more healthy than I think he would have guessed based on the first half of his career.
@tau837 😁😁😁 Nice! Right after I make my “Feels completely respectful” site quote, I scroll to the very next post and Tau launches the “WTF are you talking about”! Funny shit! Talk about bad timing by me.
Jack: I am late to the party but another really good well researched insightful article. Thank you.
i may have missed this but to me there is an elephant in the room within this article and the comments not discussed. The success or failure of play action is in some significant part based on the threat the running game plays into the defense’s mindset. If the defense is not worried about the run (they weren’t the last several years for the Chargers), play action has less impact than in the case where a team has a very effective run game (the Niners in the Kaepernick years). This to me glaring difference explains why the play action stats for Kaep and Herbert are not as divergent as I would expect. Herbert is so much more accurate and effective as a passer than Kaep. And Kaep was a very effective runner in his day which Herbert may have but rarely used. Defenses had to protect against the run much more when playing those Niner teams than the Chargers of the last few years.
Consequently, my prediction is if Harbaugh, Roman and Co. can unlock an effective run game, including having Herbert run a bit more, you will see Herbert’s play action passing stats begin to explode. The proof will be in the pudding as they say, but this is what I expect under this new offensive scheme and mindset.
@gbgh This is one of those topics I think about from time to time but stay away from, GBGH.
Does a good running game improve play action efficiency?
I know this is one of the oldest adages in football. But I also know that the analytics community and data nerds have relished showing, using data, that there is no correlation between run game success and play action success. The true analytics types believe wholeheartedly that you don’t need to establish the run to be good at play action, and you can make the Defense bite equally effectively without this threat (through motion, RPOs, and just the nature of play action itself being deceptive enough to move defenders).
I’m not qualified enough to weigh in, but just wanted to call out that it’s a controversial topic, and the data nerds think the evidence is on their side.
@alisterlloyd I was unaware that this was controversial so thanks for clueing me in. While I understand how this would be difficult to prove either way simply with data as the game is complex and lots of factors weigh in making isolating just one component hard, my experience both playing and watching is decidedly on the side of it matters. To use a somewhat extreme example to make my point, when a team like the Chargers struggles game after game in the fourth quarter to gain one or two yards in short yardage situations, Herbert is going to have a more difficult time converting on third or fourth and short using play action than say the Jaylen Hurts given the Eagle’s incredible success in those situations.
I think it is pretty clear, but definitely disagree with me, that the way a defense plays the Eagles on third or fourth and short and the way they play the Chargers (especially late in games) differs somewhat markedly. Defenses are selling out and taking all kinds of risk to stop the Eagles that they do not take against the Chargers. Defenses against the Chargers will play more balanced and protect against both pass and run since it is easier to stop the run while committing fewer resources. In that case, Herbert will always have the harder time selling the play action and completing the pass than Jaylen Hurts in the same game situation.
@gbgh thanks for commenting! Ok I have gone full sicko mode here and done a bit analysis with some of the AI tools I have at my disposal here at work. This is the result that I got trying to look at whether a strong run game correlates to a stronger play-action pass game.
Now the data sets are from across 2022 and 2023 to try and incorporate full body of data from Herbert and the running backs at the Bolts. Here is what I’ve found.
*Note – this is my own choice of data sets put into an AI algorithm, there will be some issues. By no means is this a comprehensive conclusion. @alisterlloyd said, this whole topic is controversial in the analytics world. But the process is highly interesting!
**I am only just starting to use AI to analyse large data sets from PFF and other data readily available. I highly recommend people give it a go if you’re keen. It is a lot of fun!
Goal:
To analyze the relationship between the success of the running game and the effectiveness of play action passes, using comprehensive data for Justin Herbert and the Los Angeles Chargers from the 2022 NFL season, as well as league-wide metrics.
Data Collected
Justin Herbert’s Passing Concept Data 2022:
- Play action passing: 7.7 YPA, 71.9% completion.
- Non-play action passing: 6.4 YPA, 66.0% completion.
Chargers’ Rushing Data 2022:
- Team rushing EPA: -0.242
- Team rushing success rate: 33.1%
- Key players: Austin Ekeler (4.4 YPA), Joshua Kelley (4.0 YPA).
Blocking Grades 2022:
- Mixed performance with some strong individual grades (e.g., Corey Linsley: 90.2 pass blocking grade).
Defensive Metrics:
- Opponent defensive rushing EPA: -0.102
- Opponent defensive rushing success rate: 37.5%
League-Wide Data:
- EPA and success rates for all teams’ rushing and passing games, and their respective defensive metrics.
Analysis Steps
Correlation Analysis:
- Initial analysis showed moderate positive correlations between rushing success and play action passing success.
Incorporation of Defensive Metrics:
- Adjusted analysis accounted for the quality of defenses faced by each team.
Comprehensive Metrics:
- Combined offensive and defensive metrics provided a holistic view of performance.
Key Findings
Moderate Positive Correlation:
- Rush EPA and Play Action YPA: Correlation of 0.519.
- Rush Success Rate and Play Action Completion Percentage: Correlation of 0.554.
- Indicates that a more efficient running game correlates with higher effectiveness in play action passes.
Impact of Defensive Quality:
- Opposing defenses’ rush EPA and success rate influence the effectiveness of play action passing.
- Rush EPA (Def) and Play Action YPA: Correlation of 0.262.
- Rush Success Rate (Def) and Play Action Completion Percentage: Correlation of 0.358.
Offensive vs. Defensive Dynamics:
- Teams with a strong rushing attack and facing weaker rush defenses tend to perform better in play action passing.
Conclusion
THIS analysis supports the hypothesis that success in the running game is positively correlated with success in the play action passing game. This relationship holds even when accounting for the quality of defenses faced. Therefore, a productive running game can enhance the effectiveness of play action passes, though the quality of the opposing defense also plays a crucial role.
So, that’s something I guess? And really it’s pretty simple. I don’t think we need a deep statistical analysis to say: “Good teams find ways to run the ball effectively. When good teams run the ball effectively, the play-action pass game is effective’. An eye test tells us that. But for the purpose of the exercise, there are some nuggets of interest!
Again, awesome observation which has allowed me to deepen my understanding of the game (and team) we love!
On top of that, I’d love to see a few things from Roman. Along with play action, incorporate screen passes and other misdirection plays to keep defenses guessing. We’ve lost Ekeler who was a master at that, I hope we can still have some of the screen game production. Herbert’s accuracy make these extremely effective. This has always helped neutralize aggressive defenses who want to smash Herbert to open up more opportunities for big plays.
I’d also love to see Roman use play action passes on first and second downs where the defense is more likely to be set up to stop the run, exploiting their tendency to bite on run fakes, rather than using them on later downs when defences can easily predict and read the offence.
Justin Herbert’s Passing Concept Data 2022:
- Play action passing: 7.7 YPA, 71.9% completion.
- Non-play action passing: 6.4 YPA, 66.0% completion.
Third and long very few teams would ever run so to even go through play action would be meaningless, in fact they may be in shotgun with empty backfield. The defense playing pass 100%.
Third and short and the defense is playing both. Even 3rd and 3 they have to respect the run.
I would expect that when using play action that it is due to down and distance and that the ability of the team to run the ball has some correlation as evidenced above to a higher completion percentage and YPA.
The numbers/stats may not all align, but my football brain and gut tell me that a solid run game and designing plays that look like runs, but are passes lead to wins.
@wrefordreed89 Amazing that AI can attempt to solve a complex problem in such short time!
The correlation coefficients for the defensive quality metrics are weaker <0.35) but 0.5+ for the Rush/PA relationship is a fairly decent positive correlation (as you point out!). it’s not blow your doors down evidence, but assuming the AI got it right, the findings suggest that a good run game indeed helps, which is contrary to what I’ve heard PFF people like Austin Gayle say in the past.
Thanks!
First off, you are going to give Tau a run for his money with this AI enhanced data analysis. Great stuff!
It will be interesting to see how AI impacts football analysis, strategy, data selection in the years to come. My own thought is that some combination of virtual reality and AI will radically improve the quality of play in time as you can get so many more realistic mental reps in without injury risk. Especially for QBs in being able to process quickly and correctly.
Of course I love that your analysis backs up my opinion but I suspect others will find more variables to add in showing a different conclusion. One of the great things about sports is it has a discrete set of rules but still remains incredibly complex. Little things, hard to capture or quantify, can have a profound impact on the result. My own view is the data informs but experience with the input from ever improving data and analysis is the best way for GMs, coaches and players to make decisions, especially in real time. Too much reliance on data is a mistake IMO. There is something intangible about an experienced mind that cannot be replicated by software and data.
For example, I watch a guy like Kenneth Murray and say he has no business being on an NFL football field, but PFF not infrequently graded him as something toward mediocre. Sometimes even gets asking him as playing better than Derwin which I can assure you never happened in any game I watched. Their methodology using a consistent process has its merits but it is often wrong IMO because there is so much more going on in context that the data cannot capture.
I think the new regime agreed with me letting Murray walk. He will not last long with the Titans.
I guess what I am saying is the data analysis is great but never stop trusting your eyes and your own common sense. If the data indicates your eyes and common sense are wrong, there is more likely a problem with the data or its analysis than with you.
[/et_pb_row]
[/et_pb_section]
