Madden ratings: the underlying problem

[Argooos]

1) Player ratings need to be based on actual measurable things like 40 times and based on professional scouting data rather than pure opinion. It needs to be scaled the same for every position, as described by DCEBB2001 in the FBGRatings thread.

2) Progression needs to be done in a way where you end up with a realistic spread of Overalls, but that has already been taken care of:

Quote: Originally Posted by adembroski

I really focused on balance and making sure it made sense. The scope of the cycle was such we couldn't overhaul progression, we had to figure out a way to better manage what we had. I think we came up with slight tweaks that will result in the progression and regression you're seeing make more sense. Progression is basically split into two parts- there's an evaluation part which is a huge spider web of code we couldn't do much with (though we did make it a bit smarter), and then there's a smaller section that uses the evaluation to assign progression. This is what we rewrote, because it was small and manageable. That said, it's not vastly removed from what was there before, unfortunately. With the addition of cut days, larger rosters, scouting, dynamic player performance (very labor intensive feature!), etc.., the franchise team was rather swamped. 11 is about the absolute maximum you'll see anyone improve in a single season. 12 is theoretically possible, but I've never seen it occur. Progression is a big target going forward, as Josh and I are both intensely dislike how it's done now. I think we've squeezed absolutely everything we can out of the existing code and this will be the best progression Madden's ever had if you evaluate it by balance (i.e. there'll be roughly the same number of 90+, 80-90, 70-80, etc. players each season from 1 to 30). Because that was achievable under the existing system, that's what I shot for. Had I made it more dynamic, more able to reflect those Foster/Brady edge cases, I'd have had to do the same in the other direction, which would have made the league nothing but Superstars and Scrubs by about year 4. I absolutely wanted to avoid upsetting the roster balance.

3) Potential needs to be randomized to some extent every time you start a franchise mode, so that every time you start a franchise you end up with a different range of OVRs.

[Big FN Deal]

Quote: Originally Posted by Argooos

1) Player ratings need to be based on actual measurable things like 40 times and based on professional scouting data rather than pure opinion. It needs to be scaled the same for every position, as described by DCEBB2001 in the FBGRatings thread. 2) Progression needs to be done in a way where you end up with a realistic spread of Overalls, but that has already been taken care of: 3) Potential needs to be randomized to some extent every time you start a franchise mode, so that every time you start a franchise you end up with a different range of OVRs.

I agree and potential should be hidden as well.

[pjmoore 28]

It's because of this issue that I honestly believe that DPP was invented. The "madden guru" makes the ratings which I too believe are sketchy, but players overalls will change to ur playstyle(I'm believing that's how it works at least). If u r a pass happy person, then ur qb and WR will see a larger progression than ur rb who only had 500 yards. Vice versa if u pound the rock with a two headed monster attack. So maybe that will fix the rating problem and possibly tie into a players potential rating in future games. Let's say u get two productive years out of a running back, I would assume his potential would increase as he starts moving towards his best years, and then watch his potential start going down once he hits the other side of 30. But then again, we won't fully know the ability of DPP until we get our hands on it, which won't be for another few weeks....


Sent from my iPhone using Tapatalk

[jyoung]

It gets even more ridiculous when you're in an online franchise with 32 intelligent human owners.

After two or three seasons everyone's teams are full of super players rated in the 80s - 90s.

[Argooos]

If what adembroski said is true, then this will not be as much of a problem in Madden 12.

@Big FN Deal: I think it there should be an option to have it hidden. It shouldn't necessarily be hidden by default - especially if they make scouting matter for determining potential.

[pjmoore 28]

Quote: Originally Posted by wEEman33

It gets even more ridiculous when you're in an online franchise with 32 intelligent human owners. After two or three seasons everyone's teams are full of super players rated in the 80s - 90s.

Don't even get me started on this. Eric Berry was inhuman in 3 years.


Sent from my iPhone using Tapatalk

[bucky60]

Quote: Originally Posted by Big FN Deal

"Progression Tuning Another one of AJ’s tasks was focusing on progression and making it more realistic this year. Last year, players would go up automatically regardless of their season stats. This year’s tuning is much more predictable. Players will go up or down based on their performance and age.

This is highly disappointing and completely unrealistic. Realistic progression is never based on stats. There are a myriad of problems with this. The bench player that you are trying to groom doesn't improve because he doesn't get on the field much. Typically in madden the LE gets all the sacks and stats and the RE doesn't. The LE improves but the RE goes down. Stats never drives progression in the real world, progression drives stats. Stats based progression is completely backwards and unrealistic.

[Argooos]

Quote: Originally Posted by bucky60

This is highly disappointing and completely unrealistic. Realistic progression is never based on stats. There are a myriad of problems with this. The bench player that you are trying to groom doesn't improve because he doesn't get on the field much. Typically in madden the LE gets all the sacks and stats and the RE doesn't. The LE improves but the RE goes down. Stats never drives progression in the real world, progression drives stats. Stats based progression is completely backwards and unrealistic.

Until they put in my coaching suggestions or some variation thereof, it doesn't make too much sense to not base progression on stats at all. Right now, the only non-stats thing they can base it on is age and potential. They need to introduce more variables (coaching, systems, work ethic) before that can even consider moving towards a completely non-stats-based progression system.
It's not at all that I disagree with you though.