Is EA/Tiburon Committed To Making Madden 13's Player Ratings "True To The NFL"?

[DUDERMAN]

Quote: Originally Posted by mpeterso

Finally, what MJD did this season, on that wretched Jags offense, is truly worthy of recognition. I mean, he WAS their offense this season. Surely that is worthy of 95ish ovr rating.

I would be fine with a 92-95 rating for jones-drew... but he's a 98. That may sound like nitpicking but when you're that high into the 90's, you'd think he broke 2,000 yards.

[DCEBB2001]

Quote: Originally Posted by Big FN Deal

But they don't have to be and they shouldn't be, imo. Every NFL player has hard unbiased numbers/data that can be used, in a standard formula, to create player ratings in Madden. If that was done, the only thing subjective would be the parameters of the standard formula, not the ratings themselves, which is potentially less controversial and easier to acquire a consensus around. For example, if according to the standard formula, each player's top three 40 times are averaged and converted into a Speed rating on a universal scale of 0-100, with 0 being a 30.00/sec 40 and 100 being a 3.00/sec 40, Speed in Madden would no longer be subjective, it would be universally scaled.

That is exactly what I did with FBG ratings. I set 100 at a value that was considered to be nearly impossible to reach. The big thing that helped was I have all of the combine and pro day data for every player since 1997. I was able to use this hard data to determine the averages for any player of any size at any position. However, I do NOT base the ratings off of the positional values for the raw attributes that are universal (SPD, STR, AGI, ACC, and JMP). One thing I have always hated with how EA rates players is they base them off of POSITION. If an ILB like Spikes runs a 5.10 40 time and an OT runs a 5.10 40 time they should have the same speed. Why? Because no matter their size or position, the AVERAGE VELOCITY reached over the same distance would have been the same. MASS or any other variable does not factor into the time, outside of distance and time of course. Universal scaling is the trick.

[DCEBB2001]

Quote: Originally Posted by DUDERMAN

To be fair there are some ratings that are truly subjective... AWR, TRK, BCV, SFA, SPM, JKM, RTE, SPC, POW, PMV, FMV, PUR, PRC, RBS, RBF, PBS, PBF, RET, STA, INJ, TGH. The only ones I can think of that you can use direct data for would be... Strength (bench press) Agility (cone drill) Speed (40 time) Acceleration (40 time data) Catching (percentage of caught passes) Carrying (fumbles/touches) Jumping (jump drill) Tackling (Tackles/Attempts) Block Shedding (successful/attempts) Man Coverage Zone Coverage Catch in Traffic Press Even then, there still would be problems... 1. it would be hard to accurately rate rookies in some of these categories (tackling, carrying, etc) 2. Players speed on the field is very different from a 40 time without pads 3. This wouldn't solve a big issue of players' ratings not declining. If a player ran a 4.3 9 years ago, chances are he can't anymore. 4. Not all players have combine numbers to go off of. Using data should be a good baseline but it should not be the be all end all for deciding the ratings. Having them be subjective is unavoidable.

If you have loads of scouting data for some 60,000 NFL and college players who are rated on the same scale, you can do it. This would alleviate the issues with problem #1.

The 40 times taken by scouts are to measure the POTENTIAL for MAXIMUM speed. They all run in nearly the same conditions with as minimal variables as possible. So if all the players run without pads on an even track for an even distance, you can simply measure the maximum potential for speed over the course of 40 yards. So in a sense, the 40 times are a great way to compare players to one another.

Many teams still run 40s in training camps. I know the Packers still do every year and are timed. The times are posted inside the Hutson Center for tourists during training camp. If you are able to get this data or use the data from the On-Field 40 Project, you can measure the decline from season to season. The average, amongst all players regardless of position over the course of 10 seasons is .005s lost on a 40 time for each completed season in the NFL according to my calculations.

True that not all players have combine data, but the vast majority of those who want any shot of making it into an NFL camp have at least one or two pro days or an individual workout for scouts. This vastly increases the number of players who we can get data for. I have pro day/combine data for 2089 players for 2011 alone, which is far more players than who were actually invited to a camp.

[DCEBB2001]

Quote: Originally Posted by mpeterso

While I agree certain physical attributes like height, weight, 40 speed, vertical jumping ability are quantifiable, and in the case of height and weight pretty much indisputable (though teams / players have been known to fudge these), many other attributes related to things like skills or "playing" abilities are harder to quantify, and certainly are subjective. Take speed; while some players have (or had) great 40 times, their "playing" speed never matches(d) those times. Other players, Jerry Rice being a prime example, played much faster than their 40 time ever indicated they could. Therefore, while I use 40 times as a guide to how fast a player might be, I consider on field performance almost more. Based on this, I don't think it's as simple as taking 3 40 times for each player and setting their speed rating based on those measurements alone - and that's why ratings are so subjective.

Jerry Rice was the best route runner in history. I would not attribute his ability to maintain a cushion and separate from a defender on his speed. I would attribute it to his ability to run great routes. Watch the vid of him wasting a 20 year old kid at the age of 45 on youtube. He didn't get open and run away from the CB because of being faster...he did it because he ran the route better than the ability of the defender's coverage skills.

[DCEBB2001]

Quote: Originally Posted by Big FN Deal

So if Donny Moore and the ratings "team" decide that AWR should be based on the Wonderlic Test and a players number of years in the NFL, converted into a 0-100 scale rating for every single player, AWR is no longer subjective.

If only they would have hired someone with all the information needed to make the ratings subjective from a massive, single, source.

[PGaither84]

Quote: Originally Posted by DUDERMAN

I would be fine with a 92-95 rating for jones-drew... but he's a 98. That may sound like nitpicking but when you're that high into the 90's, you'd think he broke 2,000 yards.

While I think you make a good pont and I agree with the idea behind it... you also can't focus on OVR... or at least you shouldn't. Look at the positional ratings. What is it about MJD that you think is overrated? Then we can talk about that.

For example, in Madden 11, Chris Johnson was WAY overrated just to please people with a high OVR. He had 80 or 81 trucking and a 78 stiff arm or something. Now he has like mid 50's or something in those ratings and is now only like an 88 or so OVR. He was never someone who trucked you or threw you down with a stiff arm. If he had more accurate ratings in Madden 11, people would be so upset] because of his OVR. Just look at the player's individual ratings and tell me what you think should change. I don't mean for every player, but give me some real example to look at please.

[pegout]

Quote: Originally Posted by PGaither84

While I think you make a good pont and I agree with the idea behind it... you also can't focus on OVR... or at least you shouldn't. Look at the positional ratings. What is it about MJD that you think is overrated? Then we can talk about that. For example, in Madden 11, Chris Johnson was WAY overrated just to please people with a high OVR. He had 80 or 81 trucking and a 78 stiff arm or something. Now he has like mid 50's or something in those ratings and is now only like an 88 or so OVR. He was never someone who trucked you or threw you down with a stiff arm. If he had more accurate ratings in Madden 11, people would be so upset] because of his OVR. Just look at the player's individual ratings and tell me what you think should change. I don't mean for every player, but give me some real example to look at please.

sorry to go off cource but the reason why cj was rated so high in trucking was because it would up is overall but also he did run thru tackles that year not in a willie or AP type of way but in a im going 500 miles a hour you cant wrap me up so i bounch off type of way that was the only was to get him close to hes style of play

[DCEBB2001]

Quote: Originally Posted by pegout

sorry to go off cource but the reason why cj was rated so high in trucking was because it would up is overall but also he did run thru tackles that year not in a willie or AP type of way but in a im going 500 miles a hour you cant wrap me up so i bounch off type of way that was the only was to get him close to hes style of play

Is trucking the same thing as bouncing? I don't think so. If he exemplified bouncing, up his ELU. If he exemplified trucking, up his TRK. But what escapes logic is why you would give a player more credit in an attribute that he did not deserve credit in. The mindless attribute adjustments based on how a player plays one week, even without considering consistency, is absurd. I could see them increasing CJs TRK attribute if he actually ran through and over defenders, but he didn't. Bouncing is not trucking. If he trucked over players, add to his trucking. Don't add to an attribute that is not worthy of actually being increased.

This type of stuff without any objective explanation is maddening IMO.