QB Rating by Year

[SkyDog]

Was this league started with fictional players? If so, that explains it more than anything else. Because of how cohesion works, you'll see this type of rise in all league offensive stats in the first 7-10 years of a career. Then also factor in the inflated offensive talent in the league, and it makes complete sense.

Here are the league QB ratings for the first 9 years of my most recent SP career.

2007: 72.4
2008: 76.1
2009: 76.5
2010: 76.0
2011: 78.5
2012: 78.0
2013: 79.2
2014: 80.5
2015: 83.2
2016: 82.9
2017: 85.4

For 30 seasons, starting at 2015, the rating has never fallen below 81.3, and has been as high as 87.9, but for the first eight seasons, it never got above 80.5. This isn't isolated to just QB rating, by the way. It's rushing, passing, scoring--pretty much every offensive category struggles. And it's not a case of them spiraling out of control: they're definitely too low in the early years, and rise to NFL levels and stay there (at least in SP).

As far as why it gets higher in MP, I suspect that game planning only explains this partially. Jim has said directly that the low stats early on are because of cohesion in fictional leagues. When you consider that both offenses *and* defenses are playing together for the first time in a fictional league, I can only assume this means that cohesion is more important for offense than defense. If I'm correct there, then that's also going to be a factor in why MP leagues' offensive numbers eventually get higher than those in SP: we tend to keep our teams together longer than all those AI teams do.

[Suicane]

And McMurray!

[Kobeck]

lol


were all waiting for Goddell to ask him to come to New York and pee in a bottle

[Suicane]

He can pee in a bottle in NY from St. Louis, that's how geneticly superior he is to the average human.

[Suicane]

Someone go on fofc and find the greatest year by a QB in MP.

[Hell Atlantic]

Dean Houston went undefeated one season while winning 4 championships in a row.

[Suicane]

Dean Houston went undefeated one season while winning 4 championships in a row.

SP is like playing with yourself while MP is like actually bedding a supermodel. Mark McMurray bags Victorias Secret. All of em.

[sovstar]

Was this league started with fictional players? If so, that explains it more than anything else. Because of how cohesion works, you'll see this type of rise in all league offensive stats in the first 7-10 years of a career. Then also factor in the inflated offensive talent in the league, and it makes complete sense.

Here are the league QB ratings for the first 9 years of my most recent SP career.

2007: 72.4
2008: 76.1
2009: 76.5
2010: 76.0
2011: 78.5
2012: 78.0
2013: 79.2
2014: 80.5
2015: 83.2
2016: 82.9
2017: 85.4

For 30 seasons, starting at 2015, the rating has never fallen below 81.3, and has been as high as 87.9, but for the first eight seasons, it never got above 80.5. This isn't isolated to just QB rating, by the way. It's rushing, passing, scoring--pretty much every offensive category struggles. And it's not a case of them spiraling out of control: they're definitely too low in the early years, and rise to NFL levels and stay there (at least in SP).

As far as why it gets higher in MP, I suspect that game planning only explains this partially. Jim has said directly that the low stats early on are because of cohesion in fictional leagues. When you consider that both offenses *and* defenses are playing together for the first time in a fictional league, I can only assume this means that cohesion is more important for offense than defense. If I'm correct there, then that's also going to be a factor in why MP leagues' offensive numbers eventually get higher than those in SP: we tend to keep our teams together longer than all those AI teams do.

i figured cohesion had something to do with the first couple of seasons. i don't know if i buy that it had much to do with the jump from 2010 to 2011 though. the rating actually dropped from 2008-2010. how would you explain that?

suicane's quarterback now has 22 touchdowns and no interceptions through 7 games. this guy never cracked a 90 QBR before and now he isn't making a single mistake. suicane might not have been using the best gameplans in the past, but this guy is putting up tom brady numbers with questionable talent around him. passing is screwed up in this game - to me that much is apparent.

[SkyDog]

i figured cohesion had something to do with the first couple of seasons. i don't know if i buy that it had much to do with the jump from 2010 to 2011 though. the rating actually dropped from 2008-2010. how would you explain that?

suicane's quarterback now has 22 touchdowns and no interceptions through 7 games. this guy never cracked a 90 QBR before and now he isn't making a single mistake. suicane might not have been using the best gameplans in the past, but this guy is putting up tom brady numbers with questionable talent around him. passing is screwed up in this game - to me that much is apparent.

I agree that it looks like passing works too well. I tend to disagree that it's an engine problem, per se. My take is that there are multiple factors at work here that get MP leagues up even higher.

1. COHESION--As mentioned above, based on Jim comments, fictional teams have lower cohesion. And yeah, the more I think about it, the more sense it makes that offenses in MP will do better. We don't get to see raw cohesion numbers, just relative ones. Looking at FOFL ('cause that's the league I happen to have open now), well over half the teams in the league have had their QB in place for 7 years or more, and 11 have had their QB in place for more than a decade. SP teams keep their *good* QBs for a long time now, but there just aren't that many good QBs in FOF2K7. It looks like less than a dozen have been in place more than 6 years in that league, which makes sense since less than a dozen are rated above 55. Which brings me to...

2. TALENT--I really, really, really wish Jim had had the chutzpah during the conversion to recalibrate leagues' talent to FOF2K7's expectations. All converted MP leagues have more talent at QB/WR than the FOF2K7 engine expects, and some (IHOF and FOFL and this league) had a *LOT* more. Some might say "well, the defense is better, too!" From my experience, I don't care how good your defense is in FOF; a studly QB/WR combo will outduel a studly secondary more often than not.

3. GAME PLANS--There's no question in my mind that as a whole, the FOF community is better at offensive game planning than defensive game planning. In retrospect, I'm not certain that game plan library was a great idea. They weren't *great* game plans, but they're good ones, usually better than Rex. I suspect that thread turned a good number of below-average/Rex game planners into average-to-good game planners. On the flip side, defense game planning is quite difficult, and has at least one significant bug that a lot of people probably aren't game planning around.

I guess what I'm saying in short is that I think passing was too good in FOF2K4 and in FOF2K7 up to and including version 6.0d. I'm not convinced that's so much the case now as other outside factors. (It may be a *little* too good, and that effect gets heightened by the other stuff.) I've picked up some stuff on defensive game planning quite recently from SP that I'm starting to use in MP. We'll see how it goes, but so far, my players are outperforming their talent on pass defense and finally doing better than just using Rex. In FOFL last year, I started a CB rated something like 18/20 (at best) all year long, and another good-but-not-great one, and had safeties with 42/42 and 56/56 ratings. We allowed an 82.4 pass rating against (league average=87.7). The problem with this is that I play a BUTTLOAD of SP and am in four MP leagues, and after over a year, I'm *just* to the point where I can create a defensive game plan that does better than Rex.


As for your 2008-2010 comment in particular, fictional leagues don't always go straight up as nicely as the one I mentioned. Some do fluctuate like that. I'd write that off as just "noise," in all likelihood. You should see passing level off over the next few seasons.


Regarding the Suicane QB example, what are the ratings of the QB and receivers? As best as I can tell, those "Excellent/Very Good/Good" designations actually apply now. (The engine looks at a 50/50 WR as "good.")

[Hell Atlantic]

Mark McMurray, Suicane's QB, is rated excellent at using PEDs.

[Suicane]

I get no respect I tell ya. If i wasn't so insane about McMurrays numbers i'd set my sights on averaging 200 yards on the ground from here on out.

[JesterBlaze]

Jim has said directly that the low stats early on are because of cohesion in fictional leagues. When you consider that both offenses *and* defenses are playing together for the first time in a fictional league, I can only assume this means that cohesion is more important for offense than defense.

Though the game seems to me offensively biased it should be tougher for an offense at the lower talent/cohesion (offense has to make the play regardless of defense) yet they should have that "perfect pass" window at the top for the best ones. Maybe FOF actually does this part right, but just with too low a threshold for the "perfect pass".

[SkyDog]

Though the game seems to me offensively biased it should be tougher for an offense at the lower talent/cohesion (offense has to make the play regardless of defense) yet they should have that "perfect pass" window at the top for the best ones. Maybe FOF actually does this part right, but just with too low a threshold for the "perfect pass".

{Smacks head}

Goodness. Ben, you dummy. I can't believe I never put two and two together here. I had all the information I needed to see it, but reading this post made me think it through. The more I think about it, the more I realize that it's likely that our issue with MP leagues and too much passing offense has a *lot* to do with talent relative to the engine's expectations. (I'm sure there's game plan component, too.) I say this because it's unquestionably much, much, much tougher for the passing game when you equalize every team with very low talent. If you want to see it, just create a league from a one-player universe. I have such a league that I played in SP. Here were the QB ratings and ypc for the first few seasons (before the normal-talent draftees showed up).

2006: 65.0, 4.05
2007: 67.5, 4.08
2008: 70.7, 4.01
2009: 74.1, 3.97
2010: 73.8, 3.99
2011: 76.7, 4.01
2012: 75.0, 4.07
2013: 80.3, 3.98
2014: 81.3, 4.09
2015: 79.3, 4.07
2016: 80.5, 4.10
2017: 79.7, 4.02

(This was with 6.0e, before offense got raised a hair in 6.1.)

It's pretty clear what's going on here. The running game--with crappy offensive players with no cohesion versus crappy defensive players with no cohesion--is pretty much the same in terms of effectiveness as when good offensive players with cohesion go up against good defensive players with cohesion. But the passing game SUCKS when both talent and cohesion sucks. A 65.0 rating is much lower than anything I've seen with a normal pool and no-cohesion players.

I'm going to test this out in the latest version by creating two universes-- one with all players with 0 ratings, and another with all players with 9s--and seeing what happens with stats.

[SkyDog]

Before I get all scientific (which I suspect I won't have to do). I just want to do a quick one-season "sniff test."

FIRST RUN
Default player file, but all players' overall talent set to 0. Preference draft (so lowest possible cohesion without using all rookies). X-factor off.

League-wide ypc: 4.14
League-wide QB Rating: 67.7


SECOND RUN
Default player file, but all players' overall talent set to 0. Preference draft (so lowest possible cohesion without using all rookies). X-factor off.

League-wide ypc: 4.00
League-wide QB Rating: 84.6





Yup. I don't feel a need to get scientific on this at all. Rushing ypc in both instances was within normal 6.1a range for year 1. Passer rating in both instances was outside of normal 6.1a range for year 1. That convinces me.

Just to satisfy my own curiosity, I'm going to do this again without a preference draft. Players will be assigned to their real-life teams, therefore giving a "normal" cohesion distribution...

[SkyDog]

Star league: 88.8
Default player file league: 80.0
Crap league: 70.9


(Note: in all cases in both posts, I had the X-Factor turned off, and Main Street settings, for minimal scout error.)


So, without running it a bunch of times, I think it's fairly safe to say that talent level can make an 8 to 12 point difference from the norm in QB rating, and cohesion can account for roughly half of that.