My worst fears confirmed.

[TMJOHNS18]

Maybe EA changed the scale to screw you over

[DCEBB2001]

Quote: Originally Posted by CHooe

I'm not going to question anything about how the ratings are specifically calculated - that's your purview - but I am confused as to where the 17000 players number came from. There can't be but a quarter of that number in Madden this year (assume 70 players a team * 32 teams + 500 free agents, being generous, falls just short of 3,000 players). Further, I'm confused as to why you don't have some sort of computer program, spreadsheet, or database to help you interpolate the raw data you have? I assume you have things like 40 times, shuttles, bench reps, etc., that you could use to quickly generate a baseline file which could be fine-tuned. Finally, this I'm obviously ignorant of, but did any player you generated really have a rating below 12 in any significant category? We don't care about a kicker's man coverage rating, for example. None of this is intended as a complaint of any sort, I merely offer this as a means to perhaps open up internal dialogue with you and your team about your player rating process to account for changes such as this in the future. It surely won't be the last time EA messes with the ratings.

I actually have scouting data for about 60000 players including college kids and some high school seniors. The 17000 consists of the top guys who have not retired and undrafted FAs from the past few seasons in other leagues (semi-pro/UFL/CFL/etc.).

I do have many spreadsheets for all of this. The problem is the interpolation between two sets of data. Let's say that speed is calculated using 40 times. Well those times go from about 4.20 to 6.15 and are scaled from 99 to 12. 99 being 4.20 and 12 being 6.15 with other points in between. Now, lets say that the Madden scale now goes from 99 to 0 That means that a 4.20 is still a 99, but a 6.15 is now a 1 instead of a 12. That then affects every number in that range. Get the point?

Yes, many players did have ratings below 12 in certain attributes, that is why it matters. Chalk it up to attention to detail and the strive for numeric perfection. Like I said before, this issue has already been resolved. Can we get this thread closed yet?

[DCEBB2001]

Quote: Originally Posted by sportsgamefanman

Yea, I noticed when entering all the new numbers that the overalls did not match. Can I ask, when you re rate a player do you do the whole player 1 player at a time and do every attribute for that 1 player or do you do speed and accel edits first for everyone and then move to agility, strength, etc? Do you have all the speed, accel, agility, jumping, and strength done for everyone or are you still working on this? I thought I might just use your edits in those areas to start before moving on if you are done with those edits. I think the thing is for me is I love the speed edits and I really want to get the offensive and defensive line interactions right as well as the WR and corners playing realisticly.

I have a particular method to the madness when re-rating the players. I prefer to work smarter, not harder so using several algorithms to help me out is in order.

[Nukleopatra]

Quote: Originally Posted by DCEBB2001

Can we get this thread closed yet?

Kerry Meier approves this message.

[sportsgamefanman]

Quote: Originally Posted by DCEBB2001

The whole point in them contacting me was because they felt I did great work, for free, nonetheless and they felt that I could help make their system better. The phone interview lasted over an hour and we exchanged a TON of ideas on who we rate players. Got to have a great chat with Donny about all this. I didn't get the feeling that he used any legitimate data outside of opinion. The stuff I post is SOLELY based on real NFL scouting data. I am simply interpolating from the sources I have. If you disagree, feel free to contact NFLDS and take it up with Frank Cooney. If they weren't good at what they do, they would not provide all the data for Lindy's and CBS. Just saying.

I'm curious about this. I like the fact of having real data being used for ratings. BUT, we have never got any explanation from EA about what each attribute does. So, I wonder if they know the effect some of their ratings have on gameplay. Example being super linebackers with jumping ability in the 90's. If they read these forums they would see that this has been talked about for 2 years with no change in this area.

Also, what determines who wins a battle on the line? Is it strength, footwork, blocking, shed blockers or all of the above? I think they need to rate the game and players realisticly but also to the game and how it is programmed. I think EA would be surprised at how different their game could, would, and does play with having proper ratings for not only each individual but position specific as well.

Do you factor in your re ratings how it all translates into gameplay or do you just go off of hard data?

[DCEBB2001]

Quote: Originally Posted by Nukleopatra

Kerry Meier approves this message.

Very nice. I'm not a big fan of how EA uses position to rate player's skills. Just because a guy is a LT doesn't mean his KPW should only be a 12. I highly doubt that most football players have never attempted in a practice to kick a ball at some point in their lives. I think this post from a while back sums up the satire:

http://humor.gunaxin.com/john-madden...ht-ranking/633

[DCEBB2001]

Quote: Originally Posted by sportsgamefanman

I'm curious about this. I like the fact of having real data being used for ratings. BUT, we have never got any explanation from EA about what each attribute does. So, I wonder if they know the effect some of their ratings have on gameplay. Example being super linebackers with jumping ability in the 90's. If they read these forums they would see that this has been talked about for 2 years with no change in this area. Also, what determines who wins a battle on the line? Is it strength, footwork, blocking, shed blockers or all of the above? I think they need to rate the game and players realisticly but also to the game and how it is programmed. I think EA would be surprised at how different their game could, would, and does play with having proper ratings for not only each individual but position specific as well. Do you factor in your re ratings how it all translates into gameplay or do you just go off of hard data?

I use a combination of both. The problem is that Madden puts us into a box. In fact, we discussed some of the parameters in the interview. Can't say I was the most pleased with some of the answers. So much potential thwarted by lack of logic and sound football knowledge. Very surprising.

[vln13]

Why can't you keep it as is for this year knowing their imperfect? 1. The Madden rosters are just as if not more imperfect. 2. What matters is the difference between overall players, so if all players have FBG ratings, then it doesn't matter. 3. We could potentially adjust sliders to compensate for the visual elements of ratings if they are improperly calibrated.