My take on the "Injustice" of Madden 13 Ratings

[Big FN Deal]

Quote: Originally Posted by tazdevil20

Not going to happen bud, ever. Stop torturing yourself with the idea that it might Also, I believe I saw in one of the old interviews that although the overall was a team value thing, it didn't change how the player played. So in other words, it means nothing. Just because his overall may be 85 on one team, and 74 on another, all of the other specific ratings are not affected. What's the point then? Just don't look at the overall rating since it's just for show, right? Hopefully I am wrong about this, because that would seem pretty stupid.

I think the OVR does actually matter in CC because even though it doesn't directly effect the individual ratings, it will effect their XP potential and therefore their progression. So Wes Welker in NE should presumably progress differently than if he were in Miami. Also, it's an improvement that OVR is finally a unique team value rating based on scheme and philosophy.

My standing issue is that since the individual ratings are not calculated using real life verifiable data in some scientific method, the foundation for OVR, which is individual ratings, is not balanced.

In NCAA Football they legally can't use real players so real player data is understandably not an option. However, in Madden, the are replicating real life players so, imo, the best way to achieve realistic balance with ratings is to create a universal formula(s) to translate their real life verifiable data into ratings.

[mpeterso]

Quote: Originally Posted by Jdoug312

Even with spread out ratings, you can't realistically use the full scale for 0-100. Think about it, are you really going to keep an 8 ovr player on your team? Even with spread out ratings, I wouldn't look at anyone under a 60 ovr, this is after all the NFL. The problem with EA however, is that (1) the ratings are not spread out, and (2) as DCEBB said, EA is top-heavy.

I agree that keeping an 8 ovr player on your roster is absurd. However, I'm not sure where this notion that ratings are not spread out comes from. I took a very close look at the M12 final roster update and I'm pretty sure that at least 70% of the players in the game were rated less than 80. In fact, I'm sure there were as many players under 60 as there were over 90.

Off the top of my head, I think ovr's for the entire roster breakdown were something like this:

90+ approx 10% (95+ less than 5%)
80-89 approx 20%
70-79 approx 25%
60-69 approx 30%
50-59 approx 10%
<50 approx 5%

SMH... that seems pretty stretched to me.

[ZACKATTACK52]

Yea I agree that these ratings are an injustice...it's insane!

[mpeterso]

Quote: Originally Posted by KBLover

Maybe not OVR, but that's just a formula made up of whatever the other ratings are. However, if I had a player who was a physical freak, but had 8 in AWR - I probably would. Say a WR, I probably use him as a KR/PR. A CB with 32 in CTH? Probably if he was good at coverage. Won't get picks, but won't allow completions. I'll take that. A QB with 16 ACC? As long as he wasn't supposed to be a mobile QB - sure, why not? I'd imagine Dan Marino would have a pretty bad ACC and he did alright as a pocket passer. I would imagine a pretty fast and quick WR with little awareness or much other football skill other than being decent at carrying the ball would be pretty close to that 40-50 area, especially if his CTH/CIT/SPC is low - but I see that as a special teams player, a potential KR weapon. Same for a guy who can do nothing but run pretty fast for his position and hit hard. Coverage, ball skills, overall awareness, all of that would be low. But plus speed and hits like a ton of bricks? Perfect guy to send flying down the field like a mad man at the enemy returner, imo. Yet his OVR would probably be 45-50. Heck, even in my M12 franchise, I turned Taylor Mays into a LB because he played like one and I had drafted a well-rounded SS - never got higher than 60 OVR, but I kept him around because he had some zone coverage from his SS experience, serviceable AWR (wasn't good, but wasn't Noel Devine bad), and hit like a ton of bricks. He retired on my squad as a special teams and a LB I formation subbed into the lineup with a 53 OVR. To me, it's about the skills to determine a player's worth and if he can do a couple things well, there could be a spot for him on a team, even if his OVR is pretty weak.

Not trying to be too nit picky, but how would it be possible for a NFL player to have 8 for AWR or 16 for ACC? I mean, to put it in terms of another rating that is relatively easy to quantify like SP, a 16 SP would literally be 6 times slower than the fastest players. That means over a measurable distance like 40 yards, a 16 SP player would not even have run 10 yards by the time a 96+ SP player had run 40!

Even if you scale ratings where you set a 4.2 40 to 99 and increment -1 for every .03 of a second more in the 40 (i.e. 4.23 = 98, 4.26 = 97, 4.29 = 96 etc) a 16 would equate to something in the 6.5-7.0 ball park - which is much worse than the worst 40 times I've ever seen recorded. I mean the worst 40 times I've ever seen are in 5.7-5.8 range and that was for the odd offensive lineman. Basically, no player in the NFL runs more than a 5.8 40, or less than 30-35 SP. Therefore for some ratings there surely are limits to how low they can be realistically set.

[Skyboxer]

Ratings need to go away. I mean they really need to be hidden and let the user/coach use pre-season and the season to see how each player is. Maybe even have an area to enter what we think each player should be rated.
Just give the normal 40 times, bench etc...
It's time we move on from this system.

[DCEBB2001]

Quote: Originally Posted by mpeterso

Not trying to be too nit picky, but how would it be possible for a NFL player to have 8 for AWR or 16 for ACC? I mean, to put it in terms of another rating that is relatively easy to quantify like SP, a 16 SP would literally be 6 times slower than the fastest players. That means over a measurable distance like 40 yards, a 16 SP player would not even have run 10 yards by the time a 96+ SP player had run 40! Even if you scale ratings where you set a 4.2 40 to 99 and increment -1 for every .03 of a second more in the 40 (i.e. 4.23 = 98, 4.26 = 97, 4.29 = 96 etc) a 16 would equate to something in the 6.5-7.0 ball park - which is much worse than the worst 40 times I've ever seen recorded. I mean the worst 40 times I've ever seen are in 5.7-5.8 range and that was for the odd offensive lineman. Basically, no player in the NFL runs more than a 5.8 40, or less than 30-35 SP. Therefore for some ratings there surely are limits to how low they can be realistically set.

The scale doesn't dictate how many times slower/faster a player is than another though. Look at the speed threshold. If you put it where a 99 and 69 are very close to one another, then the rate is compounded. As for the 5.7-5.8 40 time range, there are guys who ran at the NFL scouting combine or a pro day who were over 6 seconds in the 40.

Regis Crawford 6.05
Charron Dorsey 6.03
Isaiah Thompson 6.00
Roderick Gladney 6.33
Daron Rose 6.17
Scott Mitchell 6.14
Terreal Williams 6.04
Scott Ferguson 6.02

Should we not count these guys? To me if you want an accurate measure that is universal for any player to be compared, we need the entirety of the scale. I am not saying that these players on the low end will have enough talent to make it in the NFL, but in theory, they could be evaluated and many have been. So just because a guy is on the lower end of the scale doesn't mean we should discount him.

I believe that including as many players as possible and as much data as possible is a GOOD thing!

Too often I think EAs methodology and practice is the opposite.

In addition, what if we scale it where .02 is the difference between each point for SPD? Or .0212? Then everyone from 4.21 to 6.33 fits in the scale! Then you adjust the speed slider to where the difference between a 99 SPD and 0 SPD isnt 99x but say 10x.

[mpeterso]

Quote: Originally Posted by DCEBB2001

The scale doesn't dictate how many times slower/faster a player is than another though. Look at the speed threshold. If you put it where a 99 and 69 are very close to one another, then the rate is compounded. As for the 5.7-5.8 40 time range, there are guys who ran at the NFL scouting combine or a pro day who were over 6 seconds in the 40. Regis Crawford 6.05 Charron Dorsey 6.03 Isaiah Thompson 6.00 Roderick Gladney 6.33 Daron Rose 6.17 Scott Mitchell 6.14 Terreal Williams 6.04 Scott Ferguson 6.02 Should we not count these guys? To me if you want an accurate measure that is universal for any player to be compared, we need the entirety of the scale. I am not saying that these players on the low end will have enough talent to make it in the NFL, but in theory, they could be evaluated and many have been. So just because a guy is on the lower end of the scale doesn't mean we should discount him. I believe that including as many players as possible and as much data as possible is a GOOD thing! Too often I think EAs methodology and practice is the opposite. In addition, what if we scale it where .02 is the difference between each point for SPD? Or .0212? Then everyone from 4.21 to 6.33 fits in the scale! Then you adjust the speed slider to where the difference between a 99 SPD and 0 SPD isnt 99x but say 10x.

I didn't write that any players should be discounted; I'm actually arguing the opposite - every player should be counted. I am arguing in effect, when talking about SP, that a .02 scale doesn't work with a game engine that uses a scale some where around .03 or .04.

In other words, you could use a 4.21-6.33 scale where you increment 1 for every .02 difference, where a 6.33 player would have 1 SP. However, if the game uses a .03 or .04 scale instead of your suggested .02 scale, which is what I suspect, then there is going to be huge variance in the results. Which means the .02 scale likely won't work well with the game engine. Therefore any ratings scale, as I'm certain you're painfully aware, is going to be constrained by the games engine and what scales it uses for animations and sim results.

Regardless, I get the impression that other posters are actually arguing that players are literally 10 times faster than others - which is nonsense as you point out above (i.e. a 4.21 40 is actually fairly close to a 4.6 40 on the field - much closer then many people seem to realise). Therefore, I think we are on the same page. A .02 SP scale would be fine with me if I thought the game used that scale. However, I don't think it does, and until then we are stuck with EA's scale which appears to closer to .03 or .04 than .02.

[DCEBB2001]

Quote: Originally Posted by mpeterso

I didn't write that any players should be discounted; I'm actually arguing the opposite - every player should be counted. I am arguing in effect, when talking about SP, that a .02 scale doesn't work with a game engine that uses a scale some where around .03 or .04. In other words, you could use a 4.21-6.33 scale where you increment 1 for every .02 difference, where a 6.33 player would have 1 SP. However, if the game uses a .03 or .04 scale instead of your suggested .02 scale, which is what I suspect, then there is going to be huge variance in the results. Which means the .02 scale likely won't work well with the game engine. Therefore any ratings scale, as I'm certain you're painfully aware, is going to be constrained by the games engine and what scales it uses for animations and sim results. Regardless, I get the impression that other posters are actually arguing that players are literally 10 times faster than others - which is nonsense as you point out above (i.e. a 4.21 40 is actually fairly close to a 4.6 40 on the field - much closer then many people seem to realise). Therefore, I think we are on the same page. A .02 SP scale would be fine with me if I thought the game used that scale. However, I don't think it does, and until then we are stuck with EA's scale which appears to closer to .03 or .04 than .02.

What parts of the game engine do you think are scaled this way and why would it matter? Are you referring to how the generated rookies come in whereas if we were able to edit the players this could be changed? Or...are you referring to the animations directly. In my experience with the Madden game engine, if I have a player with a SPD of 1 and another with a SPD of 99 and adjust the threshold you can get the results you like. It is too bad we cannot have more adjustability in the game itself for these types of issues.