canes21's Madden 19 Rosters (PC)
 

Some of you have probably seen me show displeasure in certain ratings and the lack of spread out ratings. In no way am I calling Kane bad at his job, but there are some things I personally want to see different and now that Madden is going to PC again I am going to make those changes. Since there is a chance even one other person here may be similar to me I am making this thread.


I have a few goals with this thread/Madden 19.

1) Lower ratings across the board.

No, not a mass edit. I am not setting up any formulas to make a 99 a 91 every time or anything like that. I am going through and lowering ratings for every player one position/one attribute at a time with data from the combine for things like speed, strength, etc. and with data from PFF and others to try and get some things a bit more accurate. 40 times, bench reps, etc. are not the end all for these physical traits. I am using them as a baseline and tweaking them to reflect where a player is in their career now and how they have changed.

2) Spreading the ratings out more.

Not only am I lowering ratings, I am stretching the ranges out. Out of the box an attribute like THP may have a range of 79-99, but with my lowering and stretching out my rosters may have a range of 64-95(with potential for a once in a lifetime arm getting a 99). This will make players feel more unique and allow players to fit their roles/archetypes better rather than everyone being a a jack of all trades as is a common issue with Madden's rosters at times.


3) Using the custom draft class feature so that these aren't play now rosters only.
This is a huge reason I never played as much Madden in the past as I wanted. There were some excellent rosters that spread ratings and lowered them, but draft classes ruined them after 1 season. Now I can create classes to match my roster and franchise mode will be able to work.


4) Lowering XP sliders to get a more realistic progression.
Something similar to TDawg's 30% set. I have always disliked 63 overall rookies becoming 96 overall players in their prime. That isn't how it works. In the NFL, the majority of stars were good rookies, good enough to play early. Good enough to where they need some progression to become the star they are. A lot of average players were average prospects that progressed some to become average NFL players. I like what I saw in his thread where some superstars were 84 rookies and 91 in their prime. I like a 73 overall rookie with normal development only becoming a 76 in their prime. This is a more realistic approach to progression in my eyes.


I also want to point out that my custom draft classes will not follow EA's drafts with simple edits to keep the attributes in line. I have a different vision for classes altogether. One of my biggest issues with the generated classes has always been that I feel they go too extreme in both directions of the draft, meaning the good players are sometimes too good and the bad players are often too bad.


Let's face it, the draft is a crapshoot in real life. After the first 2 rounds there is a lot of luck involved.



One major thing I am changing that will be complimented with the lowered XP and altered progression is being very stingy with development traits. This will allow me to really make a more robust busts/sleeper/gem/etc. setup. Right now a gem/sleeper is basically a random player with 1st or 2nd round talent that is programmed to fall late into the draft. You can scout them out ahead of time too easily currently. It's true for busts as well. They are just poorly players programmed to go higher in the draft.


To combat this I am going to be tightening up the classes a bit. Now the last 3-4 rounds won't be filled with such bad players they'll literally have no shot at making a team and won't just be wasted space. In EA's generated classes, unless you get a sleeper/gem type in the last few rounds you're going to be cutting the player you picked and he'll be deleted from the game within a season or two because they are so bad. By tightening the class talent this won't be the case anymore. Now guys in rounds 4-7 may be rated between 62-70, but the higher round guys will have a higher chance at having a better development trait. The same will be the case with the entire draft.


This changes a few things. First, busts are going to be a bit differently. Now instead of picking a 72 overall in the 1st round and recognizing he is a bust right away I am going to set it up to where more 1st round picks are talented and the busts are guys that have Normal development. This means your 78 overall 7th overall pick may look good as a rookie, but with his development trait combined with the lower XP/slower progression he may only progress to an 81 by his 3rd year and it may begin to look like he is a bust in a way because he never really reached the potential you thought he had as a top 10 pick. He may be an 81, still a pretty good player with the stretched ratings, but he isn't the elite player he was projected to be on draft night.


On the flip side you may get a QB in the 4th round who is a 73, but has Superstar development so you groom him behind your veteran for 4 years, focus on him in training, and he is an 81 overall by his 4th year, he's taking over as a starter and he's definitely a sleeper in hindsight. Now busts/gems are taking years to show that is what they are and you aren't instantly knowing draft night.


Now the draft won't be super easy like it is now. Now you and the AI are going to be getting a lot of players in all rounds that are actually going to make you make decisions. Since rounds 4 and on are not going to be filled with guys that are basically fillers and will get cut, you'll have more decisions to make. Do you keep the quicker developing 72 overall rookie in the hopes that he becomes an above average player at least or do you keep the 78 overall veteran who has normal progression and likely won't progress much more, but he's a more for sure thing being more polished. It's a low floor, high ceiling vs high floor, low ceiling situation that teams find themselves in in real life.


Imagine the experience of 4 of your last 5 draft picks all being Normal developing. They're not bad players, but you've "wasted" nearly 4 picks not getting even 1 potentially elite player. You're rebuild is behind and you have decisions to make now. Or imagine the opposite happens, you get lucky two drafts in a row and have a QB you got in the 4th and a WR in the 5th that both are star developers. You can now groom them, build a team around them, and get way ahead of schedule.


Simply changing the way the draft ratings are done and using the development traits as a ceiling level tool with a slower progression setup opens up the draft to being more Fog of War-esque and gives you so many more scenarios you otherwise would not get by default. Having the class be not so extreme on both ends is going to make it that much harder to scout it and figure out who you want. There will be more projecting and hoping. I aim to rid of it being a simple scout a rating or two and move on system where you can pick out the busts/gems so easily. With everyone being closer together and the backend being more talented, but slower developing it is going to be much much harder figuring out who really is going to be able to progress some and become good and who is just going to stay nearly the same as they are as a prospect and never really become true NFL material. I'll still have some players that really shouldn't gone pro like real life and I'll still have my top picks be talented. It won't be the same each class. Some classes will be weak, some strong, but minimizing the extreme ratings each way will really make the draft so much more dramatic compared to the simple way it is now.

Re: canes21's Madden 19 Rosters (PC)
 

I'll be following this thread closely throughtout the yr, even though I wont be playing on PC. I think Madden rates the players the way they do is because of certain threshold points in the game (Dice rolls) ( which im sure your aware of) & not the way a player would be rated in real life at times. Ive thought about doing ratings from scratch, but in my mind it was never worth it enough to take on the full project. Didnt really see the desired results I was hoping for that would warrant many hrs editing
in the end game. I do edit on a way smaller scale, but its more for better player movement & weight.

Best of luck!

Re: canes21's Madden 19 Rosters (PC)
 

Decided to buy on ps4 and PC so I’m going to create a class for both.


Sent from my iPhone using Tapatalk

Re: canes21's Madden 19 Rosters (PC)
 

After doing some thinking I have decided that I will be doing variants of each draft class. The reasons for doing this are a) making the classes myself, the more variants I have the less likely I am to remember any busts/sleepers in each variant, b) so that if I run multiple CFMs with friends or alone I can have the same players in each year's class, but have different stories each time, c) if any other users actually use these classes they can download one version to check out the ratings, traits, etc. and then download another to use so they won't know the exact ratings, busts/gems, etc. in the class they actually use.



The biggest changes is that each class will have totally different sleepers. If I decide to have 2 players in rounds 5 through 7 that have superstar development, Player A and B will have it in Class 1 Version A, but in the other versions they will never be the sleepers again together. Player A and Player Q may be the sleepers in Version C. Player B may never be a sleeper in another class, but A. Also, I will do minor changes at the top of classes as well. Say I have a class that is headlined by two top QBs projected to go 1 and 2. In 4 of the 6 versions these two QBs may be 1 and 2 interchangeably and both have top potential with their development trait. In 1 version of the 6 QB2 may end up being rated well, but be a bust potential/development trait-wise. In the final version both QBs may not be complete busts, but neither may have Superstar development so they may never be the two greats they were expected to be, but they'll still be two good QBs more than likely.


Basically, I will be making sure the sleepers are never the same so that the later rounds are fresh, sort of dynamic between each version. Then I will do minor tweaks each version so that the tops of the classes aren't always the same, but I am not going to go crazy and make in unrealistically dynamic for the sake of being dynamic. The changes will all be to keep it fresh, but maintain consistency and realism in the ratings that also go hand in hand with the base roster I am creating.


Another thing I am interested in doing is also having different versions of my base roster. After I do testing with my initial roster/ratings I may end up saving the base version, then going in and tweaking different players slightly to make different versions dynamic in a way as well. In my base roster Baker Mayfield may have quick development. In another version he may have SS. Another he may get Star. Lamar Jackson may be normal development in all of my versions except one. These are just examples, these aren't what their traits really are right now. This will also make sure that if myself or others use the roster in multiple CFMs they can use different versions so that the same rookies/young players don't always get good while the same players also are buss in the end.

Re: canes21's Madden 19 Rosters (PC)
 

Cane, another level of depth you can add with your draft classes can take advantage of the "shallow" scouting system Madden currently has. For a bust, since we can only see the top 3 attributes you could pretty easily bump up 3 specific attributes to make a bust look like he might be an elite player but outside of those 3 good attributes the rest of his attributes might fall off dramatically. Then we could have two players who scout the same but has vastly different ratings/ceilings in game.

Another thought with the late round gems is because Madden has removed the ability to change traits outside of editing a player yourself you can make that late round pass rusher who may only be a 70 overall have clutch, strip ball, and the full complement of pass rushing moves. This way he'll likely play better than maybe a 76 overall who has better ratings but really has no traits except for bull rush.

Additionally, this one is more personal but it'd be nice to see late round guys with very average physical skills have one of their scouted ratings be awareness or play rec. A lot of late round gems really shine on their smarts and work ethic rather than physicality and this is a possible way to simulate that.

Anyway, just a couple thoughts I had while reading your write up that could add additional layers. Loved the write-up and looking forward to seeing and using these rosters!

Re: canes21's Madden 19 Rosters (PC)
 

Great ideas. I've definitely given plenty of thought on how to incorporate that stud player that is average physically, but is a student of the game. A MLB that falls to the 4th because he doesn't measure well, but comes in with great play recognition as a rookie and has his career last because of that is something I definitely want to create.

Re: canes21's Madden 19 Rosters (PC)
 

In case anyone is following this I have decided after some thought that I will not be rerating the entire roster. I simply do not have the time to do it in the timely manner I'd like. It would take me a month or longer at least with what I have coming up in life and I'd rather spend that time with gameplay and draft classes. If a tool ever surfaces that allows for easier roster editing then I will rethink my position, but for now I regret to say I will not be doing my rerate.


I did rerate the Bengals and Bears QBs just to see what they ended up OVR-wise and how the gameplay would go and I was very surprised to see the ratings drop as much as they did. Dalton turns into a 63, Mitch is a 59. Tyler Bray is a 37 overall. I then edited Brady and Rodgers and each turned into an 83 overall. It seems the formula for calculating OVR is very sensitive this year. It would take a bit to get used to seeing guys as 45 overall and not completely dismissing them, but I do like the idea of having guys rated 37 to 83 at the start. It may be a bit on the far end, but that is more of the type of separation I would like to see in the ratings.



Editing the players wasn't a huge pain, but it wasn't quick by any means. I also have not even halfway rerated half of the entire NFL's offensive players in Excel yet. It was very time consuming combing data and using that to create ratings in Excel. I simply have too much in life approaching these next few months for me to want to take any more time away from playing games to edit a roster. It does seem in the gameplay that there is a bigger difference in ratings these years so that softens the blow.


I still plan to mess with progression and look at what others find here so I can get a system in my franchise where players grow 3-5 points throughout their career on average while stars grow much more. And with that I'll also create draft classes. Those are both things I for sure plan to do, but for now I simply cannot do an entire roster rerating. Maybe sometime in the future I can, but I would rather spend what time I do have playing my CFM and creating classes.

Re: canes21's Madden 19 Rosters (PC)
 

Thanks for the update.


Sent from my iPhone using Tapatalk