RFF's M18 32-man CFM USER v USER Sliders

[Aestis]

Major update.

I've been holding off posting the past few weeks as I've been waiting for more complete results. I've made a few fairly sizable adjustments which has improved our #s substantially and the feel of the offense greatly.

Full context, the pass rush had me in a desperate state. Even at 100 PBK, the pressure was coming too fast & sack rates were way too high, like 3x the NFL. I was trying everything--penalty sliders, etc. Nothing made a big difference. User PBK seemed like it did nothing. So I tried something even more desperate--experimented with CPU PBK. And lo & behold, both our league and another league using our sliders got the 'beta' version and also agreed....

Raising CPU PBK seems to affect User v User games. I can't explain it. I can't prove 100% it made the difference, but both league data AND the "feel" of the pass game changed overnight when I bumped CPU PBK from ~35 to 75. As far as I know, this doesn't apply to other sliders. Certainly not things like Catch, QB Acc, RBK, etc.


So without further ado, here is how we fixed the passing game:
(Optional) 1) Lower Holding slider to 20 & tell everyone in the league they can set Blocking to Aggressive.
(Optional) 2) Moved down to Slow game speed. Can barely tell the difference, but that extra half-second helps guys make smarter decisions.
(The big one) 3) Raise CPU Pass Block to 75.

With the CPU PBK change, our pass game exploded:
- 4 week avg sack rate before the change: 15.4% (meaning users were getting sacked on nearly 1 out of 6 drop-backs)
- 4 week avg sack rate after the change: 10.9% -- i.e. sacks dropped nearly 30%.

Sacks are much higher than NFL average, because if you give users an extra second or two, they won't just get the ball out quicker... they'll just wait in the pocket longer & still get sacked, lol. BUT in addition to sacks dropping...

- Thru the first 9 weeks, the highest median pass yardage we'd ever seen was 200 (target = 235) and usually the median pass yards in a week was in the 160-190 range. In the 4 weeks since the change: 233 avg, and 3 of the 4 weeks our median pass yardage was > 240 yds. That's a staggering change.

- Our Avg Yds per Att went up from 7.5 (4 weeks prior) to 8.3, a 10% increase after mostly hovering in the 6s & 7s through 9 weeks.

At least for us, this was a huge unlock. It's such a surprising find, that I wanted to make sure it wasn't a 1-2 week fluke. If anything, our pass game might be TOO good now and could even require a small tweak back down.

I've also updated RBK/TKL sliders with blocking & tackling both getting a boost, it's helped a bit with the boom-or-bust run game... higher blocking = fewer negative runs. Higher tackling = fewer giant 50+ yd runs as a result of juke/spin moves. Doesn't remove the issue, but it has helped. More 3-5 ypc run games instead of half below 2 and half above 7.

A few other small tweaks to get the data a little bit closer to NFL levels (+FG Acc, -Punt Power, +Catching, etc)


10/17 Slider Changes (OP updated as well)

Difficulty: All-Madden
Game Speed: Slow
Injury: 22

QB Accuracy: 60
Pass Block: User 100, CPU 75 (yes CPU PBK required)
Catching: 80
Run Block: 59
Fumble: 77

Pass D Reaction: 0
Interceptions: 25-33 (33 if you hate dropped pick animations and don't mind a few extra INTs to punish bad throws, lower if you can live with them to keep INTs a bit lower)
Pass Cov: 0
Tackle: 65

FG Power: 45
FG Acc: 100
Punt Power: 53
Punt Acc: 49
Kickoff Power: 51

Offsides: 45
False Start: 10
Holding: 20 if you want Aggressive Blocking to be used regularly, otherwise 60
Def Holding: 85
Facemask: 51
Def Pass INT: 65
Illegal Block: 50
Roughing: at 51 it's called often, at 50 it's called rarely.

[Cajungodfather]

Have you tinkered with a lower speed threshold at all? We did find that going to 35, helped wrs get more separation, and as far as I can tell, with the AM
Difficulty, speed doesn’t become the king attribute, but does have value. Our offensive yardage went up when we lowered it.

Side note, everything you said about all pro is correct. In my other league, they finally decided to go to AM, after maxed out passing sliders, there was still a 70-42 game, in the preseason. We are going to try this new set you have in PFL, but with lower threshold. If speed is OP again, I’ll go back up.


Sent from my iPhone using Operation Sports

[Aestis]

Quote: Originally Posted by Cajungodfather

Have you tinkered with a lower speed threshold at all? We did find that going to 35, helped wrs get more separation, and as far as I can tell, with the AM Difficulty, speed doesn’t become the king attribute, but does have value. Our offensive yardage went up when we lowered it. Side note, everything you said about all pro is correct. In my other league, they finally decided to go to AM, after maxed out passing sliders, there was still a 70-42 game, in the preseason. We are going to try this new set you have in PFL, but with lower threshold. If speed is OP again, I’ll go back up. Sent from my iPhone using Operation Sports

Threshold is good where it is. Speed is still absolutely a factor, I've gotten several bombs deep with Will Fuller barely catching a ball over a slower CB who took a false step, as an example, but it doesn't get yards of easy separation on every route all the time. You don't see many pure streaks = automatic TD vs man/press unless it's a matchup disadvantage. Streaks are subtly available vs, say, Cover 3 when the defense is playing closer to LOS and the deep third S takes a false step. A fast WR can get behind the defense but requires a good read to ID this and a lob leading away from the S help, not just: oh it's press/man, guess I get the TD now.

Big bodied WRs with good catching attributes have a place in this set. The animations are great. At 35 I feel things would just go back to speedsters > skilled WRs. But you also have to do what's best for your league. Lower speed threshold = will work a bit better for less skilled franchises, easier reads because separation is so much greater. I don't say that condescendingly, the first step is a commish team has to know their league & be honest about what's going to fit both their goals AND league skill to execute on this.

Our passing stats with the set I posted are actually a tiny bit *too* good at the moment, which is a great and refreshing thing the past 4 weeks. The last thing I'd want to do is lower SPD threshold to make them even bigger. Scoring is in a good place.

If you want to see a very big passing game on these sliders (23-26, 88% comp, 339 yds) that did not even lead to much scoring, here you go... note this represents a huge passing game, not league avg. But just showing what it looks like and how it's still VERY possible:
https://go.twitch.tv/videos/182791216
Big downfield pass plays at:
4:00
1:03
1:04

[Redhawk05]

are pass block, pass cov, or reaction time sliders reversed? I read on another slider thread that someone really believes the pass cov and reaction sliders are reversed.

so a pass cover slider setting of 1 would be better pass cov than if it were set to 99.

[Aestis]

Quote: Originally Posted by Redhawk05

are pass block, pass cov, or reaction time sliders reversed? I read on another slider thread that someone really believes the pass cov and reaction sliders are reversed. so a pass cover slider setting of 1 would be better pass cov than if it were set to 99.

He's not alone in thinking that, but for every controlled practice setting I've seen suggesting they may be reversed, I've seen another test that I considered to be better evidence they are not reversed.

I can't give you a 100% answer, and my advice is not to trust anyone claiming 100% proof of ANY slider without showing it with data. A lot of folks believe practice mode isn't even tuned the same as in-game. I tend to believe there's some truth in this based on my experience, but I can't say for sure.

So, color me extremely dubious that they are actually reversed. Our league passing #s are quite strong at 0 Pass Coverage with the set I've posted. The middle of our league is averaging 8+ yds per attempt each week, over 60% completion rate, & 210-240 yds passing. Again, those are the middle-of-the-pack numbers (weekly medians). Sacks & INTs are still high but both are much better than when we first started, and frankly they will always be high because of user decision-making, not sliders.

I am MOST dubious that coverage is reversed. Our completion rate gradually rose over the course of ~5 weeks as I was gradually dropping the coverage slider each week to get our passing game working better. Meanwhile our INT rate gradually dropped. The pass coverage slider dropped from 75 to 0 during this time, and the pass reaction slider dropped from I think 60 to 0. Now, you could argue the improvements were from users getting used to All Madden & a new game and not because of sliders dropping, and you might be right.

I therefore can't 100% say the slider changes are what caused this gradual improvement--modest but still noticeable in the numbers. But I can 100% say that as the coverage sliders dropped all the way down to 0, our passing numbers gradually went UP, not DOWN. That's not based on an eyeball impression or a single play in practice mode--that's based on real data across ~70 USER v USER games (=140 teams worth of data points).

So like I said. You may interpret this data how you will, but color me extremely dubious.

[Cajungodfather]

Quote: Originally Posted by Aestis

Threshold is good where it is. Speed is still absolutely a factor, I've gotten several bombs deep with Will Fuller barely catching a ball over a slower CB who took a false step, as an example, but it doesn't get yards of easy separation on every route all the time. You don't see many pure streaks = automatic TD vs man/press unless it's a matchup disadvantage. Streaks are subtly available vs, say, Cover 3 when the defense is playing closer to LOS and the deep third S takes a false step. A fast WR can get behind the defense but requires a good read to ID this and a lob leading away from the S help, not just: oh it's press/man, guess I get the TD now. Big bodied WRs with good catching attributes have a place in this set. The animations are great. At 35 I feel things would just go back to speedsters > skilled WRs. But you also have to do what's best for your league. Lower speed threshold = will work a bit better for less skilled franchises, easier reads because separation is so much greater. I don't say that condescendingly, the first step is a commish team has to know their league & be honest about what's going to fit both their goals AND league skill to execute on this. Our passing stats with the set I posted are actually a tiny bit *too* good at the moment, which is a great and refreshing thing the past 4 weeks. The last thing I'd want to do is lower SPD threshold to make them even bigger. Scoring is in a good place. If you want to see a very big passing game on these sliders (23-26, 88% comp, 339 yds) that did not even lead to much scoring, here you go... note this represents a huge passing game, not league avg. But just showing what it looks like and how it's still VERY possible: https://go.twitch.tv/videos/182791216 Big downfield pass plays at: 4:00 1:03 1:04

So we tried the 35 threshold, for a few weeks, but now, I’m slowly increasing, until we get closer to that 70 range. Personally, I prefer the higher threshold, at the same time, I feel like lowering it, has lowered to frustration level of the overall league, letting guys score a little more than earlier in the season. Guys are starting to adjust/learn how to play madden 18, it’s just taking almost a full season to do, but it’s happening. Speed hasn’t become the dominant attribute yet, but no doubt, guys will figure it out, then turn to speedsters. So the quicker I raise threshold, the better.

Are there any plans to lower QB acc? I feel like I don’t get many stray passes with a rookie qb, but we also get greater separation with lower threshold, so it’s possible we would need it at 60, if we were at 70 threshold.


Sent from my iPhone using Operation Sports

[Aestis]

Quote: Originally Posted by Cajungodfather

So we tried the 35 threshold, for a few weeks, but now, I’m slowly increasing, until we get closer to that 70 range. Personally, I prefer the higher threshold, at the same time, I feel like lowering it, has lowered to frustration level of the overall league, letting guys score a little more than earlier in the season. Guys are starting to adjust/learn how to play madden 18, it’s just taking almost a full season to do, but it’s happening. Speed hasn’t become the dominant attribute yet, but no doubt, guys will figure it out, then turn to speedsters. So the quicker I raise threshold, the better. Are there any plans to lower QB acc? I feel like I don’t get many stray passes with a rookie qb, but we also get greater separation with lower threshold, so it’s possible we would need it at 60, if we were at 70 threshold. Sent from my iPhone using Operation Sports

Slowly moving from 35 to 70 is a gigantic difference in terms of gameplay--if that's what works for you guys though, more power to ya. Maybe other leagues who are struggling with AP being too easy and AM being too brutal can use that tactic.

I would like to see QB Acc come down a bit in theory, I do think default feels best. But coverage is tighter on AM, and DB reactions to the ball are so fast that if accuracy is also too low, then the passing game gets pretty rough. Accuracy will certainly be one of the things that gets adjusted a bit if the passing game needs a nerf.

[Solofunk]

The cpu pass blocking up from 50 to 75 seemed to help us out a lot. We are still getting more pics (and opportunities for pics which leads to many drops) than we would like and the separation that receivers get seems to be a lot less than in an all pro league I am in. What effect does increasing or decreasing DPI, Defensive Holding, and OPI have on this? I know it makes coverage better or worse. Also, we had a problem with receivers and safeties getting an insane amount of award xp, 3x more than other positions. 32k for a probowl! We are trying to keep xp sliders for them a bit higher now for training but will lower it down week 16 to lower award xp. Will this work? Or will the award xp lock week 1 based on the sliders, or can it change like this?