Madden NFL 11 News Post

FBGRatings Seeks to Recalibrate Madden Ratings, Change Game

02:46 PM - January 24, 2011 by RaychelSnr

I recentely sat down for a talk with FBGRatings.com's Dan Berens to discuss his site's vision and what's going on over there today. The site is currently working on getting accurate ratings for every player using real hard data converted into the Madden ratings universe. Dan claims that when these numbers are plugged into the game, it plays much better and much closer to real life. Check out the interview below and also check out Dan's website to see what he's got going on!

Interview with Berens on the OS Radio Show on BlogTalkRadio

Game: Madden NFL 11Reader Score: 6/10 - Vote Now
Platform: PS3 / Xbox 360Votes for game: 96 - View All

Madden NFL 11 Videos

Member Comments

# 901 buckey00 @ 10/23/12 03:31 PM

Quote:

Originally Posted by DCEBB2001


	FBG Ratings does not produce rosters at this time. We are in negotiations, and have been for some time, with a 3rd party for roster production. As for an external rosters editor for M13, I am unaware of this. Someone care to elaborate?

In the rosters section of the forum, there is a thread called madden roster editor v1 and it is an external editor you can use to edit all ratings on your computer. It was made for madden 12, but works perfectly fine for 13 as I've tested it. The thread has all the directions on how to do this.

# 902 DCEBB2001 @ 10/24/12 09:11 PM

Roster production will have to wait until next year.

I am presently working on some HUGE changes to the ratings that will make the game even more realistic. I am partnering up with a local university math department and an expert in bio-mechanics to better analyze and implement FBG ratings into Madden. It will be ready for M14 at which point I hope to be able to have roster downloads available.

# 903 8mileroad @ 10/24/12 11:52 PM

Quote:

Originally Posted by DCEBB2001


	Roster production will have to wait until next year. I am presently working on some HUGE changes to the ratings that will make the game even more realistic. I am partnering up with a local university math department and an expert in bio-mechanics to better analyze and implement FBG ratings into Madden. It will be ready for M14 at which point I hope to be able to have roster downloads available.

Wow. Why doesn't EA just hire guys like you to do factory rosters?

# 904 DCEBB2001 @ 10/25/12 09:05 PM

Quote:

Originally Posted by 8mileroad


	Wow. Why doesn't EA just hire guys like you to do factory rosters?

Because EA likes the status quo. Now I say that realizing that they "add" features, but let's be real here; EA adds stuff that "revolutionizes" the game and yet removes features that also "revolutionize" the game. Their direction is nonsense.

What I offer, in merely the ratings only, is ratings based on real data...not youtube videos and stats. It is a different, yet professional, approach to rating players. If the data is sound, the methodology is logical, and the approach is possible, then Madden can move forward. What I am doing now is the start of that.

This year I demonstrated that weekly updates for the database can be easily implemented. I also demonstrated that massive changes to numerous players in one update is possible in a short period of time.

Next year I want to have players who behave more realistically to their real-life counterparts, more than they do presently, that is. Next year each player will have SPD ratings based off of their real-life maximum velocity. ACC ratings will be based off of their real-life rate of acceleration. STR ratings will be based off of the player's ability to push, pull, or affect a force in multiple tests, which I have data for. All of this will be better implemented thanks to the team I have working on the site now.

If EA would realize the potential and how far behind they are in simply creating real-life representation in the game, they would be wise to bring someone on board who has knowledge in regards to data interpolation, biomechanics, statistics, and forecasting. They appear to want to keep things in-house, however.

# 905 RAVENOUS @ 10/26/12 07:24 AM

Quote:

Originally Posted by buckey00


	In 13 most of the Star development players from the draft are in the 70s and 80s depending on schemes. That would be another variable with madden 13 if you do end up doing it. And in ccm the new development system allows us to easily progress players much more so than in years past, especially if the player user controls the games. The physical attributes do cost more which is nice. Another problem in madden is there no chance for those extremely athletic prospects or physically skilled prospects to bust or underwhelm like they do in the NFL all the time. Such as Jamarcus Russell, Aaron Curry, Vernon Gholston, Felix Jones and so on. And then in Madden there is also no way to turn the marginal athlete in a quality player because this game relies so much on physical attributes. It is much easier to use a heyward-bey type player with bad route running and catching but burning speed and agility than using someone like Welker who does not have the same speed but has the catching and route-running necesarry to be a good receiver. Almost all teams in the NFL would take Welker but in Madden I find it easier to use bay.

The Roster Editor, and I'm sure I've said this elsewhere, opens up the possibility to completely redo the Madden game to how you see fit, to the point where you'd really only need it for the gameplay. This is of course if you could write a program to work with the editor to implement your vision (I can't).

You can write your own formula for player pro/regression, pop your franchise into the editor every week, or at whatever interval you want, and have it do progression YOUR way before you put your franchise back on the console. Get yourself some first round busts going on.

You can pretty much replace any sim phase of the game, really. Save 'chise in the editor before coach changes, retirements, resignings, free agency, etc. Sim through the phase and save again. Pop first save in the editor and let it do the phase your way. Overwrite the second save.

To the Bey>Welker problem, I can't agree, though I'm not and I haven't played 13 other than the demo. I remember playing a franchise as the Dolphins on 11 and I tore up with Brandon Marshall despite his high 80s speed. That's what (somewhat) got me off my speed fix. Release was the key, his was 97 or 98, and he'd punish press coverage every time. I'd always count on him to be open on almost any route I wanted against the blitz.

Now when I play 12, I see press and I'm looking Boldin first and Torrey second every time. It really depends on how you play with sliders and difficulty, too. I can't wait for Torrey and his 95 speed to get open somewhere downfield if I'm facing a decent pass rush, or throw to a 96 speed 70 CIT player on a slant route because he WILL drop the ball when he gets hit.

To that end, I've been in the process of testing ratings individually to see what they do by using the editor to set everything to 50 and upping stuff one at a time. Stumbling across these FBGRatings though, WOW! I couldn't ever hope to be this precise in fine-tuning anything. Do you know how each rating impacts gameplay and take that into account, or do you focus only on matching players' RL skill sets to their corresponding in-game ratings?

One thing the editor would allow you to do, if you have your own formula for player overalls, is to put those in the game. They'd show in Player Management, but the Depth Chart would still show default overalls.

These are amazing.

# 906 DCEBB2001 @ 10/26/12 12:21 PM

Quote:

Originally Posted by RAVENOUS


	The Roster Editor, and I'm sure I've said this elsewhere, opens up the possibility to completely redo the Madden game to how you see fit, to the point where you'd really only need it for the gameplay. This is of course if you could write a program to work with the editor to implement your vision (I can't). You can write your own formula for player pro/regression, pop your franchise into the editor every week, or at whatever interval you want, and have it do progression YOUR way before you put your franchise back on the console. Get yourself some first round busts going on. You can pretty much replace any sim phase of the game, really. Save 'chise in the editor before coach changes, retirements, resignings, free agency, etc. Sim through the phase and save again. Pop first save in the editor and let it do the phase your way. Overwrite the second save. To the Bey>Welker problem, I can't agree, though I'm not and I haven't played 13 other than the demo. I remember playing a franchise as the Dolphins on 11 and I tore up with Brandon Marshall despite his high 80s speed. That's what (somewhat) got me off my speed fix. Release was the key, his was 97 or 98, and he'd punish press coverage every time. I'd always count on him to be open on almost any route I wanted against the blitz. Now when I play 12, I see press and I'm looking Boldin first and Torrey second every time. It really depends on how you play with sliders and difficulty, too. I can't wait for Torrey and his 95 speed to get open somewhere downfield if I'm facing a decent pass rush, or throw to a 96 speed 70 CIT player on a slant route because he WILL drop the ball when he gets hit. To that end, I've been in the process of testing ratings individually to see what they do by using the editor to set everything to 50 and upping stuff one at a time. Stumbling across these FBGRatings though, WOW! I couldn't ever hope to be this precise in fine-tuning anything. Do you know how each rating impacts gameplay and take that into account, or do you focus only on matching players' RL skill sets to their corresponding in-game ratings? One thing the editor would allow you to do, if you have your own formula for player overalls, is to put those in the game. They'd show in Player Management, but the Depth Chart would still show default overalls. These are amazing.

So much of this post reminds me of the great work done by NZA and Gommo back on M08 for the PC. Using editors to truly customize your franchise experience.

I haven't looked into the editor yet, but it seems like a cool idea if it can be implemented properly. That stuff is way above me, however, as I am not a programmer by any means.

What I do is use real life NFL scouting data to determine Madden ratings. I have data for over 63000 players including those in college and some high school seniors. The primary goal is to rate players according to the real life data distribution via the scouting information. Right now, however, we are looking at better linking the ratings to the animations in the game.

Our first phase is determining a players real life maximum velocity to determine their SPD ratings. This is done in conjunction with their rate of change in their velocity, or acceleration. Once these are figured out and made compatible in the game, you will be able to have two players with equal 40 times and differing acceleration ratings run the same 40s but differentiate after that point. In Madden 13, ALL PLAYERS stop the acceleration phase at 45 yards, regardless of ACC or SPD rating. I suppose that EA assumes that all players stop accelerating by this time. These two ratings are also joined with the STA rating to determine how long it takes until their acceleration phase ends and the deceleration phase begins. This is why I have employed an expert in biomechanics. From there, all other raw attributes (STR, AGI, JMP) are determined. This would signify the end of the first phase.

The second phase is the implementation of the dynamic attributes (position specific attributes) from the scouting data into the Madden ratings, utilizing the distribution of the data.

The third and final phase is altering all of the data into Madden ratings via the OVR calculations. This is when the new, revolutionizing, ratings will be published online.

So as you can see, there are big changes coming to the site and how we rate players. I decided to go straight to the experts on this via the academic realm. A few lucky graduate students are even considering this project for their stats/applied mathematics graduate theses. If anyone else is interested in joining the project feel free to send me a PM. It is not for pay, but if you are a student and want to get some nice research papers out of a topic that allows you to analyze your favorite video game, then get in touch with me.

# 907 splff3000 @ 10/26/12 02:59 PM

Quote:

Originally Posted by DCEBB2001


	So much of this post reminds me of the great work done by NZA and Gommo back on M08 for the PC. Using editors to truly customize your franchise experience. I haven't looked into the editor yet, but it seems like a cool idea if it can be implemented properly. That stuff is way above me, however, as I am not a programmer by any means. What I do is use real life NFL scouting data to determine Madden ratings. I have data for over 63000 players including those in college and some high school seniors. The primary goal is to rate players according to the real life data distribution via the scouting information. Right now, however, we are looking at better linking the ratings to the animations in the game. Our first phase is determining a players real life maximum velocity to determine their SPD ratings. This is done in conjunction with their rate of change in their velocity, or acceleration. Once these are figured out and made compatible in the game, you will be able to have two players with equal 40 times and differing acceleration ratings run the same 40s but differentiate after that point. In Madden 13, ALL PLAYERS stop the acceleration phase at 45 yards, regardless of ACC or SPD rating. I suppose that EA assumes that all players stop accelerating by this time. These two ratings are also joined with the STA rating to determine how long it takes until their acceleration phase ends and the deceleration phase begins. This is why I have employed an expert in biomechanics. From there, all other raw attributes (STR, AGI, JMP) are determined. This would signify the end of the first phase. The second phase is the implementation of the dynamic attributes (position specific attributes) from the scouting data into the Madden ratings, utilizing the distribution of the data. The third and final phase is altering all of the data into Madden ratings via the OVR calculations. This is when the new, revolutionizing, ratings will be published online. So as you can see, there are big changes coming to the site and how we rate players. I decided to go straight to the experts on this via the academic realm. A few lucky graduate students are even considering this project for their stats/applied mathematics graduate theses. If anyone else is interested in joining the project feel free to send me a PM. It is not for pay, but if you are a student and want to get some nice research papers out of a topic that allows you to analyze your favorite video game, then get in touch with me.

WTF? How in the hell do you not work for EA at this point? Your way of doing things sounds a lot of better than "oh he dropped 2 passes last week, let's lower his catch rating by 10 pts".

# 908 DCEBB2001 @ 10/26/12 06:20 PM

Quote:

Originally Posted by splff3000


	WTF? How in the hell do you not work for EA at this point? Your way of doing things sounds a lot of better than "oh he dropped 2 passes last week, let's lower his catch rating by 10 pts".

That is a very good question. Perhaps you should write them and ask them this question.

# 909 RogueHominid @ 10/27/12 09:59 AM

Quote:

Originally Posted by Big FN Deal


	Honestly, at this point it makes perfect sense why DCEBB or someone of his ilk are not working at EA Tiburon, because EA Tiburon doesn't value optimal realism. I don't mean that as some kind of smart arse comment either, that is just evident at this point, imo. DCEBB has even posted before about how nonsensical it is, with all the exclusive access EA Tiburon has to the NFL, they haven't brought in a real pro football scouting agency to rate players for Madden. They clearly are not interested in realistic ratings because they don't think that's "fun".

I think another very important reason why this sort of data isn't included in Madden is because it's an animation-driven game in which animations are triggered by arbitrary values set by the developers.

If this were a true real-time physics game, then animations would trigger when guys had realistic attributes to unlock them. As it stands, many animations are only available when you figure out the Madden rating for them, which is usually way to high or oddly arbitrary. So in order the get players to do unique skill-based things, you have to figure out the Madden calculus, and that bears almost no relation to real-life data.

No judgment there, it's just the difference in the way the game's built vs. the difference in ratings generated by a physics-based, empirical model based on research. If the two every meet, that'd be great.

# 910 jwin122 @ 10/31/12 05:53 PM

I find it hilarious that Madden makes such a big deal out of the ratings, even going as far as "who should get a two-point increase this week?" Really??? A player has one good game and they want to boost his ratings to whatever odd-ball number they draw from a hat lol!

All in all, they can fuss all they want over them, but the bottom line is they don't mean nothing once you're in game...you have to spend weeks messing with sliders to even begin to get them to matter, and even then, lower-rated players dominate higher-rated players in some cases...

I have found that FBG ratings do provide a higher level of realism as far as player performance in-game goes...hopefully we get a roster file with their ratings to use in-game in the future...that would make my Madden-time a happy time!

# 911 DCEBB2001 @ 11/02/12 03:13 PM

Quote:

Originally Posted by jwin122


	I find it hilarious that Madden makes such a big deal out of the ratings, even going as far as "who should get a two-point increase this week?" Really??? A player has one good game and they want to boost his ratings to whatever odd-ball number they draw from a hat lol! All in all, they can fuss all they want over them, but the bottom line is they don't mean nothing once you're in game...you have to spend weeks messing with sliders to even begin to get them to matter, and even then, lower-rated players dominate higher-rated players in some cases... I have found that FBG ratings do provide a higher level of realism as far as player performance in-game goes...hopefully we get a roster file with their ratings to use in-game in the future...that would make my Madden-time a happy time!

As I said previously, rosters will hopefully come around next year. The site in itself is such a massive load of data that I really haven't had much time to create my own rosters. It used to be a lot easier when Madden was on the PC due to the great external editors. If it is true that you can externally edit rosters now, that would expedite the process greatly. Originally, I had planned on getting a team of roster editors to create them for download on the different platforms, but the negotiations have been slow. I know I keep saying "next year" for a lot of this, but I would find it to be Tom-Foolery to simply do something half of the way.

Glad you like the gameplay in-game, though. Please feel free to elaborate or write a review on the gameplay using FBG ratings so others may have an idea on the effects.

# 912 buckey00 @ 11/04/12 03:05 AM

What is the scale players are rated on speed, acc, agi and strength? I assume speed is based on 40 time, acc on shuttle, agi on cone, and strength on bench? I am totally fine if you are not comfortable saying, but I am just planing on editing the fbg rosters into Madden and then I want to add some college players in just for fun. I want to make their physical attribuits realistic though. Again I am perfectly fine if that conversion is something that you do not want to be known.

# 913 SloeyEZ @ 11/04/12 04:00 AM

DC, have you actually gotten your hands on the forms/charts/spreadsheets scouts use? I'd be incredibly interested in seeing that. I used to have a great-uncle who was an offensive line coach for many many years but unfortunately, I never really got to know him, nor see any of his personal paperwork. *yeah before computers lol, but you'd know his name if I said it probably, he was pretty big and well respected in the 70s and early 80s. Okay, done name-dropping (or not name dropping on purpose), but I would love to see any official forms/data you have gleaned from the NFL especially or major collegiate schools. Of course your answer on here may have to be "No, of course I have nothing proprietary from any football organization!!" and then PM me

# 914 DCEBB2001 @ 11/04/12 10:09 AM

Quote:

Originally Posted by buckey00


	What is the scale players are rated on speed, acc, agi and strength? I assume speed is based on 40 time, acc on shuttle, agi on cone, and strength on bench? I am totally fine if you are not comfortable saying, but I am just planing on editing the fbg rosters into Madden and then I want to add some college players in just for fun. I want to make their physical attribuits realistic though. Again I am perfectly fine if that conversion is something that you do not want to be known.

There is a scale that correlates particular data to a particular ratings set. However, the one that is on the website now is in the process of being revised with the help of a local university mathematics department and a professional in bio-mechanics.

Once rosters are made for download next year there will be a guide that can help you edit rookies in Madden so they fit your FBG Ratings rosters.

Other than that I cannot get into specifics for obvious reasons.

# 915 DCEBB2001 @ 11/04/12 10:13 AM

Quote:

Originally Posted by SloeyEZ


	DC, have you actually gotten your hands on the forms/charts/spreadsheets scouts use? I'd be incredibly interested in seeing that. I used to have a great-uncle who was an offensive line coach for many many years but unfortunately, I never really got to know him, nor see any of his personal paperwork. *yeah before computers lol, but you'd know his name if I said it probably, he was pretty big and well respected in the 70s and early 80s. Okay, done name-dropping (or not name dropping on purpose), but I would love to see any official forms/data you have gleaned from the NFL especially or major collegiate schools. Of course your answer on here may have to be "No, of course I have nothing proprietary from any football organization!!" and then PM me

Let me preface this answer by stating that the site only utilizes real NFL scouting data. That data is provided to me by a close friend in the business. I have signed a Non-Disclosure Agreement to not publish this data directly nor share it with any party outside of those brought on to the FBG Ratings project. In order to get the help I needed for the site I have had various entities also sign a NDA. That data is not going to be released to any party not directly involved with the project as a result. I don't want to risk killing the Golden Goose here.

# 916 SloeyEZ @ 11/04/12 12:53 PM

Quote:

Originally Posted by DCEBB2001


	Let me preface this answer by stating that the site only utilizes real NFL scouting data. That data is provided to me by a close friend in the business. I have signed a Non-Disclosure Agreement to not publish this data directly nor share it with any party outside of those brought on to the FBG Ratings project. In order to get the help I needed for the site I have had various entities also sign a NDA. That data is not going to be released to any party not directly involved with the project as a result. I don't want to risk killing the Golden Goose here.

Nice, and understood. thanks

# 917 KBLover @ 11/04/12 03:47 PM

Quote:

Originally Posted by RAVENOUS


	To the Bey>Welker problem, I can't agree, though I'm not and I haven't played 13 other than the demo. I remember playing a franchise as the Dolphins on 11 and I tore up with Brandon Marshall despite his high 80s speed. That's what (somewhat) got me off my speed fix. Release was the key, his was 97 or 98, and he'd punish press coverage every time. I'd always count on him to be open on almost any route I wanted against the blitz. Now when I play 12, I see press and I'm looking Boldin first and Torrey second every time. It really depends on how you play with sliders and difficulty, too. I can't wait for Torrey and his 95 speed to get open somewhere downfield if I'm facing a decent pass rush, or throw to a 96 speed 70 CIT player on a slant route because he WILL drop the ball when he gets hit.

Yeah, but the fact you have to use sliders illustrates that it doesn't matter out of the box.

Sure, I feel the same thing in M12 when I put the WR catching at like 15 or something. Then, yeah, the bad RTE (and good CIT) receivers really stand out more. I really have a possession guy and a DJax type guy who's great in the clear, but I wouldn't throw him a slant between zone LBs on a key 3rd down unless I had to.

# 918 buckey00 @ 11/04/12 04:13 PM

Quote:

Originally Posted by DCEBB2001


	There is a scale that correlates particular data to a particular ratings set. However, the one that is on the website now is in the process of being revised with the help of a local university mathematics department and a professional in bio-mechanics. Once rosters are made for download next year there will be a guide that can help you edit rookies in Madden so they fit your FBG Ratings rosters. Other than that I cannot get into specifics for obvious reasons.

Understood, and thank you.

# 919 KBLover @ 11/05/12 12:08 AM

Quote:

Originally Posted by Arrow218


	Lavon Brazill, who Reggie Wayne said is the Idk Donald Brown's 40 time, but he is very fast (watch him, speed is more than a 40 time) and has 87 speed?

The 40 time might show a player's top speed.

If he 'plays faster' then that probably means he has good acceleration and ability to maintain his speed during changes of directions, quick start-stop, etc.

A player can't go faster than he's physically able to move. He can use his speed more effectively than other players of the same speed or even those of greater speed, but that's not speed, then.

A 4.7 player can not move as fast as a 4.4 player, he can get to a spot faster, though, as long as it's not straight line speed depending on other skills.

Since there are ratings for these other skills (like AGI, RTE, and ACC) then that's where those efficiencies should show up, imo.

# 920 DCEBB2001 @ 11/05/12 10:53 PM

Quote:

Originally Posted by Arrow218


	Lavon Brazill, who Reggie Wayne said is the fastest player on the team, has 88 speed. TY Hilton and his 4.34 40 has 93 speed. Idk Donald Brown's 40 time, but he is very fast (watch him, speed is more than a 40 time) and has 87 speed? Rookie TE Dwayne Allen has 68 speed? Have you watched him at all?

Brown officially posted a best 40 time of 4.46. That is good for a SPD rating of 87 in this system. Brown may look fast not because of his top SPD, but because of his near-elite acceleration rating of 88 after posting a 10 yard split of 1.50. It is the COMBINATION of all ratings that can make a guy LOOK faster than he really is.

Allen ran a 4.89, good for a 68 SPD (He didn't even match the 14 year average of 4.81 among all positions. He also chose not to run at his pro day on 3/8/12, meaning he clearly ran what he considered to be a "good time")

Brazill, a 4.44, who also chose to stand on his combine time and not run at his pro day. Good for an 88 SPD. Is Reggie Wayne a college or pro scout? The data doesn't verify Mr. Wayne's personal beliefs, clearly.

Hilton's 4.34 is good enough for a very elite SPD rating of 93.

No matter how you slice it, the data doesn't lie. How you interpret that data is up to you. I am presently revamping how SPD and ACC is calculated, but until those results are posted, I am sticking to this method. If I instead went by what every "scout", player, or strength coach said a player's speed should be then everyone would be a 99. Just saying...the data yields no bias.

« Previous 1 2…43 44 454647 48 49…76 77 Next »

More Links

FBGRatings Seeks to Recalibrate Madden Ratings, Change Game