What should attributes be based off of?

[DCEBB2001]

I know this has been brought up before but I would like the community to give me some opinions as to what attributes should be based off of. I am attempting to create more realistic ratings for the game on my website FBGratings.com/members for the benefit of the entire Madden gaming community. In doing so I need some ideas on how to determine some attributes. I have hard combine data for just about every player since 1998 which helps, but I can't necessarily use 40 times to determine stiff arm ratings...get it?

I also have a wide variety of scouting data to rely on as well, but this data does not always show particular attributes like stiff arm, trucking, etc.

So what I would like to know is how should certain attribute values be calculated? Should the stiff arm rating be dependent upon strength? Should the juke move be based on one's agility and weight?

Try to consider the physics behind some of these terms. Something like Hitpower being a function of FORCE x VELOCITY where FORCE = MASS x ACCELERATION. Now we know the mass of many players by their weight at 1 earth gravity and we can measure how fast they accelerate from combine data. Throw in a velocity and bingo! You just found a hitpower!

I am looking for ideas like this. How would you, in a perfect Madden world, want attributes to be calculated. PLEASE BE SPECIFIC if you can.

Thanks again to the community and check out the site!

[guaps]

Re: What should attributes be based off of?

How should your proposing the ratings system look? How many individual ratings and what ratings would you create?

Are your suggested ratings system based on the current Madden system, or do you envisage a system with, say, proper physical interactions, a route-based passing system etc??

Also how detailed is your combine data? Have you got the 10 and 20 yard split times for the 40 yard dash times or just the 40 yard dash times?

[DCEBB2001]

Re: What should attributes be based off of?

How should your proposing the ratings system look? How many individual ratings and what ratings would you create?

Are your suggested ratings system based on the current Madden system, or do you envisage a system with, say, proper physical interactions, a route-based passing system etc??

Also how detailed is your combine data? Have you got the 10 and 20 yard split times for the 40 yard dash times or just the 40 yard dash times?

The system I use now is within the box that EA gives us. That means that we have to use the attribute categories they use.

The combine data is very detailed. 40, 20, and 10 yard splits, cone drill, shuttle, vertical, broad, bench reps, height to the nearest 1/8th inch, and weight.

[cheapargentina]

Re: What should attributes be based off of?

only problem i see with ur suggestion is the following

u cant base a stiff arm on strength alone let me give u a reason y. brandon jacobs probably has the highest strength rating of any rb (or at least close to the top) but LT isnt as strong and has a better stiff arm than he does. u cant completely take technique of of the equation

chris jhonson is probably one of the most athletic guys in the league but he doesnt spin like marshall faulk used to so once again technique and individual skill play a part in it too its not just raw physical talent

ive posted this before but i figured it would help u out and finally i think the only way to really get prefect ratings is to watch film stats lie in football and at times they are very far from the truth

[adembroski]

Re: What should attributes be based off of?

Ratings should be based on one thing and one thing only: Player performance. I don't care if a guy is 5-2, 115 pounds, if he's knocking guys out on the field, he should have the POW rating to replicate it in Madden.

Alex Smith, great example I've watched all three weeks. He makes good decisions, gets rid of the ball quickly,takes off when he needs too, but utterly MISSES receivers high and they get tipped for interceptions. Solid but not great awareness, terrible accuracy. No reason to over think it, rate exactly as they perform.

Scouting agencies are highly useful for this sort of thing.

[DCEBB2001]

Re: What should attributes be based off of?

only problem i see with ur suggestion is the following

u cant base a stiff arm on strength alone let me give u a reason y. brandon jacobs probably has the highest strength rating of any rb (or at least close to the top) but LT isnt as strong and has a better stiff arm than he does. u cant completely take technique of of the equation

chris jhonson is probably one of the most athletic guys in the league but he doesnt spin like marshall faulk used to so once again technique and individual skill play a part in it too its not just raw physical talent

ive posted this before but i figured it would help u out and finally i think the only way to really get prefect ratings is to watch film stats lie in football and at times they are very far from the truth

I have a lot of scouting data for most of these players to help make sure that they are rated accordingly. However, the issue arises when you have guys who are rookies or are new to the league. You may not have as much data on them which will make it more difficult to rate them.

[DCEBB2001]

Re: What should attributes be based off of?

Ratings should be based on one thing and one thing only: Player performance. I don't care if a guy is 5-2, 115 pounds, if he's knocking guys out on the field, he should have the POW rating to replicate it in Madden.

Alex Smith, great example I've watched all three weeks. He makes good decisions, gets rid of the ball quickly,takes off when he needs too, but utterly MISSES receivers high and they get tipped for interceptions. Solid but not great awareness, terrible accuracy. No reason to over think it, rate exactly as they perform.

Scouting agencies are highly useful for this sort of thing.

Preaching to the choir on the scouting agencies. I work for one...so I have nearly unlimited data. In fact the boys in Florida who make this game...are all subscribers. Too bad they don't use the information effectively.

[cheapargentina]

Re: What should attributes be based off of?

I have a lot of scouting data for most of these players to help make sure that they are rated accordingly. However, the issue arises when you have guys who are rookies or are new to the league. You may not have as much data on them which will make it more difficult to rate them.

and u cant get anything from their college career? i knw its not the same but at least u can get a feel for their style of play

[DCEBB2001]

Re: What should attributes be based off of?

and u cant get anything from their college career? i knw its not the same but at least u can get a feel for their style of play

Getting stuff for their college career is easy. Getting the right stuff is hard. You don't typically find how well a guy can stiff arm for all 2500 rookies every year.

[cheapargentina]

Re: What should attributes be based off of?

Getting stuff for their college career is easy. Getting the right stuff is hard. You don't typically find how well a guy can stiff arm for all 2500 rookies every year.

very true but i assume the bigger the school and the higher the guy got drafted the more info u might have which is probably better then no info at all

[DCEBB2001]

Re: What should attributes be based off of?

very true but i assume the bigger the school and the higher the guy got drafted the more info u might have which is probably better then no info at all

Not necessarily. This year alone we had pro day data for some 2000 players! That's unreal! I have all the hard data, but I do not want this to be lop-sided in favor of guys just because they are drafted higher than others. In actuality, it shouldn't make a difference. Perhaps you can raise some ideas for solutions instead of arguing the point?

[guaps]

Re: What should attributes be based off of?

Not necessarily. This year alone we had pro day data for some 2000 players! That's unreal! I have all the hard data, but I do not want this to be lop-sided in favor of guys just because they are drafted higher than others. In actuality, it shouldn't make a difference. Perhaps you can raise some ideas for solutions instead of arguing the point?

First, I don't like the current rating system as I find it to be flawed and completely lacking transparency. Second, I've advocated for a revamped system almost all over OS by now and I really want to see a new philosophy from Tiburon on ratings. They should just look at the Fifa series for an example of a really great rating system.

Regarding making top draft picks lop-sided (a natural bias) has much to do with the fact that people expect more from a first round pick compared to a seventh round pick. I believe JaMarcus Russell was 82 overall in Madden 08, because everybody expected a lot from him at that time. As he didn't live up to expectations his ratings plummet, regardless of pro day and combine data. Had he been a 5th round pick, his ratings would have been lower from the beginning, because so were the expectations.

So basically what I am trying to say is, that the current ratings system in Madden awards overall and individual ratings according to expectations, which is exactly what we see when Donny Moore releases the weekly roster updates. If a player preforms above expectations EA rewards him with better ratings and if he preforms worse then his ratings go down. This has nothing to do with pro day or combine data, because the player's measurable abilities hasn't changed, yet we see a weekly roller coaster ride from EA Tiburon.

As long as EA Tiburon sticks to that philosophy, I don't see any point presenting solutions on how to make more accurate ratings. A revamped ratings system could do wonders for the game, while any tweaks to the current system will not change the flawed nature of the current system.

[DCEBB2001]

Re: What should attributes be based off of?

First, I don't like the current rating system as I find it to be flawed and completely lacking transparency. Second, I've advocated for a revamped system almost all over OS by now and I really want to see a new philosophy from Tiburon on ratings. They should just look at the Fifa series for an example of a really great rating system.

Regarding making top draft picks lop-sided (a natural bias) has much to do with the fact that people expect more from a first round pick compared to a seventh round pick. I believe JaMarcus Russell was 82 overall in Madden 08, because everybody expected a lot from him at that time. As he didn't live up to expectations his ratings plummet, regardless of pro day and combine data. Had he been a 5th round pick, his ratings would have been lower from the beginning, because so were the expectations.

So basically what I am trying to say is, that the current ratings system in Madden awards overall and individual ratings according to expectations, which is exactly what we see when Donny Moore releases the weekly roster updates. If a player preforms above expectations EA rewards him with better ratings and if he preforms worse then his ratings go down. This has nothing to do with pro day or combine data, because the player's measurable abilities hasn't changed, yet we see a weekly roller coaster ride from EA Tiburon.

As long as EA Tiburon sticks to that philosophy, I don't see any point presenting solutions on how to make more accurate ratings. A revamped ratings system could do wonders for the game, while any tweaks to the current system will not change the flawed nature of the current system.

This is where you are wrong. What we CAN, in fact, do is alter how these players are rated within the current system. It has been done before and can be done again for those of you who were around during the football-gamers.com (now FBG ratings) era. All you have to do is stick to the hard data and refuse to rate players based upon potential, which is what I intend to do.

I have more data than those guys at EA and more to pull ratings from. I also know for a fact that they are subscribers to my site (irony?) as they have used FBG in the past and their accounts are active...they still pay for the draft site too BTW.

The bottom line is given the box they put us in, I firmly believe we can do a better job given the amount of information we have to pull from and the reactions to ratings in particular by the community. So to get back on topic and once again ask...how do we correlate real life data to ratings in the game outside of guessing or going off of spotty and often incomplete scouting data?

[Maelstrom-XIII]

Re: What should attributes be based off of?

I think the issue comes from the intangible qualities that EA tries to quantify, or are necessary to be quantified for the game engine to properly run a simulation of the sport. How can you quantify a QBs accuracy? "Terrible" isn't a quantity. How terrible is a 65 rating? It gets kinda iffy...

[steelcurtain311]

Re: What should attributes be based off of?

They should just look at the Fifa series for an example of a really great rating system.

Or what NBA 2K11 did this year. I was looking at players ratings, and I was freaking astounded by the tendencies ratings alone. I immediately thought "Madden SO needs this." It would do amazing things on the defensive side of the ball, if players were actually rated correctly. Not simple ratings, but ratings that affect the little things, especially help with offensive line/pass rusher play. One of the major gripes of Madden is no difference in the players, and I agree, because there's nothing to set them apart. You can do the same things with some third tier QB that you can do with Peyton Manning. Or if you're DeMarcus Ware, you can get blocked out of this world by a RB when you're about to hit the QB. These types of things are gamekillers.

Of course, ratings are just ratings. We all saw how the "roles" system worked in Madden. It changed nothing. Why? Because the game itself is horribly broken. Until there's actually legitimate player momentum, where strength, size, weight, momentum, all are in effect, there will never be legit line play. As long as your QB continuously throws to the DB instead of his WR, since the spacing and ball movement are so horrendously screwed up, and you can't even complete basic routes because of it, then what's ratings matter? They make no difference until the actual gameplay is fixed, and they've made it pretty clear that won't happen.