Passing offense overtuned?

[markdub2]

Anyone else experiencing passing stats being too high? For example, my 56 rated QB (ie non-elite) just hit 76% (nfl record) completion percentage over 14 games that he played, and my rookie WR1 (mid 60s rating) caught 21 TDs (3nd most single season receiving TDs all time behind Moss and Rice). I am using a custom playbook, but I didnt game the playbook to try to break anything, although I know thats possible, I just added in real NFL passing plays. Are yall experiencing QBs/WRs playing a little too well?

That said, I appreciate that the sim output is much closer to modern NFL passing than FOF8 which had notoriously low passing offense, it just seems it might have gone a little too far, especially when Rex's potentially sub-optimal playbook generation isnt nerfing the overall output of offenses.

[scoman]

QB Completion Percentage - Front Office Football Central (https://forums.operationsports.com/fofc/showthread.php?t=99081)

[Ben E Lou]

Per Pro Football Reference, league-wide completion percentage the last five seasons:


2018: 64.9
2019: 63.5
2020: 65.2
2021: 64.8
2022: 64.2


I can't recall if Jim uses three-year or five-year averages off the top of my head. Either way, there's not a huge difference there. The baseline would be either 64.5 or 64.7. I just ran a 30-season sim with 9.0c, injuries at 100, starting with default rosters. Here's a screenie:


https://www.dropbox.com/scl/fi/0ec3oqyd3ad92cz5qq3qh/30seasons-90c.png?rlkey=clatq5wzvh3yzozer7m2v2eew&dl=1
Dropbox - 30seasons 90c.png - Simplify your life (https://www.dropbox.com/scl/fi/0ec3oqyd3ad92cz5qq3qh/30seasons-90c.png?rlkey=clatq5wzvh3yzozer7m2v2eew&dl=0)


Even if we go with the lower number, we get 13 seasons above it, 16 seasons below, and one right at 64.5. I don't see any indication here that overall percentage is too high. You're looking at individual statistical outliers.

[scoman]

Wrap it up however you want mid to high 70s is too high for the league leaders, how about showing your league leaders for those years,that is what we are talking about not the overall league average which seems to be fine.

[Ben E Lou]

Wrap it up however you want mid to high 70s is too high for the league leaders, how about showing your league leaders for those years,that is what we are talking about not the overall league average which seems to be fine.
Ummm...I don't LIKE the current stats in the NFL, but they are what they are. Mid-70s has happened twice in the last six seasons. A statistical outlier creeping into the upper 70s on an individual basis seems completely believable. League leaders should not be absolutely topped at the current record. There needs to be the possibility of new records being set..and a new record that's just 3-4 percent above the current one seems completely plausible to me. That's why I find the overall numbers to be far more relevant to the discussion of tuning the engine, not any individual performance.

[scoman]

Ummm...74% has been done twice in the history of the NFL, by the same guy, I prefer realistic results. Unrealistic results result in a immersion killer. The game is producing those unrealistic results on a yearly basis...much more than a outlier.

[markdub2]

Even if we go with the lower number, we get 13 seasons above it, 16 seasons below, and one right at 64.5. I don't see any indication here that overall percentage is too high. You're looking at individual statistical outliers.

Thanks for the reply and the data, I am open to it being an outlier, and am happy that NFL records can be broken by the game engine in a natural way, I was just shocked that a middling QB was able to smash it like that.

[NawlinsFan]

It's interesting that the discussion revolves around the QB and not the setting the QB is in. A lot of any QB's success is the receiving group they have supporting them so in part what is seen both in real life and in game is due to the roster as well. With that a mid-range QB just may have higher ratings in accuracy and timing that contribute as well.

In the end it's a game that is "representative" of a professional league. I think we just need to appreciate what it can do and how it draws us in with it's addictive play and not get to wrapped up in the one , or few, instances when it isn't just right.


Note: Now that being said I am still frustrated that I can't set the stadium configurations in a non-default universe. My peeve. :banghead: :) :lol:

[RGVicedo]

Anyone else experiencing passing stats being too high? For example, my 56 rated QB (ie non-elite) just hit 76% (nfl record) completion percentage over 14 games that he played, and my rookie WR1 (mid 60s rating) caught 21 TDs (3nd most single season receiving TDs all time behind Moss and Rice). I am using a custom playbook, but I didnt game the playbook to try to break anything, although I know thats possible, I just added in real NFL passing plays. Are yall experiencing QBs/WRs playing a little too well?

That said, I appreciate that the sim output is much closer to modern NFL passing than FOF8 which had notoriously low passing offense, it just seems it might have gone a little too far, especially when Rex's potentially sub-optimal playbook generation isnt nerfing the overall output of offenses.

I agree a tune-up may be in order :devil:

I just ran a sim using the Eagles and Jalen Hurts had 5800 yds passing 70 TDs 8 INTs and a 75.7 comp.

As a team the Eagles had 6257 yds passing and 2865 yds rushing.

[jcard]

Ummm...74% has been done twice in the history of the NFL, by the same guy, I prefer realistic results. Unrealistic results result in a immersion killer. The game is producing those unrealistic results on a yearly basis...much more than a outlier.

Agreed. It is fine (actually preferable) that the game engine allows for performances at or beyond historical extremes. At the same time, such outliers were the result of a propitious confluence of skill, coaching, opposition, teammates, play calling, and luck. In particular, play calling can play a role beyond just calling the “right” plays— fealty to a certain scheme that facilitates outliers in a particular statistic often comes at the expense of a trade-off of a lower ceiling on other metrics— so there are diminishing marginal returns overall from maximizing one statistic.

[scoman]

Im all for a chance to have records broken,but it should happen once in a generation type player not on a yearly basis.




Example,I just finished my season in a fictional league 2030 32 team 16 game format.


Top 3 in Passing Percentage 77.5,74.6, 74.2 which is simply too high for the leaders.






Check your season stats ,its pretty much a yearly event.

[gfrench]

I turned up the injury rate to help deal with this.

[scoman]

I turned up the injury rate to help deal with this.




Yeah ,pretty shocked this isn't getting addressed. So how does that help?




And if I remember right 200 was the most realistic injury setting in FOF 8 ,wonder if its the same with FOF 9?

[Ben E Lou]

Yeah ,pretty shocked this isn't getting addressed.How do you know that? Did you report it and hear back that it will not change or is that just speculation? (Curious because I'm working on a customization that would be impacted by a chance in stats output.) So how does that help?In every version of FOF since it was possible to test it, when talent is matched on both sides of the ball, the greater the talent, the higher the offensive output. Injuring more players drops the talent level and results in decreased offense. Haven't checked this in FOF9, but I'd be surprised if it isn't still the case.

[henry296]

I wonder if this is a related to passing depth of target. Not sure if I'm understanding this correctly by looking at my most recent season in advanced passing, most QBs are around 4 yards in the air yards per target metric.

Looking at football-ref they have intended air yards per target which seems to be somewhere around 8.

Are those the same calculation?

About 57% of attemps are considered short in the game. Not sure how that compares.

[scoman]

Was hoping this would be addressed in 9.0d .




Ran a test league 32 team 16 game default setting ...Top 2 were 77.8 and 77.0 and the 77.0 was in over 500 attempts



So apparently the developer doesn't deem this to be problem.


Too bad the user couldn't adjust league totals ala OOTP then everybody would be happy.


Adjusting the injury setting doesn't work as it leads to unrealistic individual season totals.


So I guess it it what it is.

[NawlinsFan]

Was hoping this would be addressed is 9.0d .




Ran a test league 32 team 16 game default setting ...Top 2 were 77.8 and 77.0 and the 77.0 was in over 500 attempts



So apparently the developer doesn't deem this to be problem.


Too bad the user couldn't adjust league totals ala OOTP then everybody would be happy.


Adjusting the injury setting doesn't work as it leads to unrealistic individual season totals.


So I guess it it what it is.

What is interesting is a I did a similar run and the highest result I had was 71.3 and another at 70.4. I would assume that the results from universe to universe would vary and that in some instances they may be elevated and in others not so much. Could be that this is why this isn't considered a high priority at the moment.

[Ben E Lou]

Was hoping this would be addressed is 9.0d .
{snip}
So apparently the developer doesn't deem this to be problem.So....asking this question again from above: did you report it? If not, why assume that the developer is looking at it or even aware of it as a concern?

[scoman]

What is interesting is a I did a similar run and the highest result I had was 71.3 and another at 70.4. I would assume that the results from universe to universe would vary and that in some instances they may be elevated and in others not so much. Could be that this is why this isn't considered a high priority at the moment.




I bet your injury setting wasnt 100

[NawlinsFan]

I bet your injury setting wasnt 100

Nope, 75

[scoman]

Nope, 75




I simmed a few more years and it did look better,maybe i jumped the gun.

[Habsfan18]

It does seem a bit high, even in the new patch. Keep in mind it’s a small sample size but I simmed 5 seasons tonight in my default settings 32 team long-term test file and I’m seeing anywhere from 7-10 QB’s with a completion percentage of over 70% at the end of each season. The top guy is usually around 75%, and at least 3-4 with higher than 72%.

So it’s a bit high for sure, maybe 2-3% higher than average for the top guys across the board, but that’s a ridiculously small sample size. I’ll likely take some time over the weekend to sim like 100 years and see what we’re looking at.