1) Ratings distributions need to fall within the user’s expectations (a certain amount of stars, a certain amount of average players, a certain amount of scrubs, etc.) This is true not jsut of the league as a whole, but within age groups. Rookies should be generated with ratings you’d expect from a rookie, 2nd and 3rd year playesr should improve at rate comparable to how real players improve, players should age, get worse, and retire like real players, etc.

2) Rating distributions need to stay relatively consistent from year to year. A league full of fake players needs to produce the same kind of games that a league full of real players does. In practice, this means that you want individual rating distributions of fake players (agility, awareness, strength, and so on) to be around the same as it is in the default roster. Otherwise, you end up with a game that isn’t quite balanced because, (a made-up example), fake running backs tended to develop break tackle 8 points higher on average than the real ones.

3) Rating distributions need to be varied and interesting. When rookies are generated, there need to be some elite players at the top of the draft. There needs to be a variety of player “archetypes” created (at HB, you would have power backs, finesse backs, all-purpose backs, etc.). There need to be players who have a high potential to be stars, but who aren’t going to be ready to take a starting job (have low initial ratings) for another year or two.

Neil, thanks for that, and I agree. I just don’t think it’s a priority. No doubt that it could be done and not be a huge drain on the budget.

Brando, I think it should be slanted more toward ratings than age IF they fix the age model. If a guy is 37 and still a 96 OVR, he shouldn’t quit. Thing is, VERY few players should be a 96 OVR at the age of 37. Every year when you look at the retire list it is filled with 88+ rated players. I think that’s way off.

In fact, I think pushing the high 30s…you should start to see 5, 6, 7 point drops rather than the small ratings drops that we see in Madden.

I have simmed to 2010 now and I’ll post a “new player” rating breakdown when I’m done which should show a bit more clearly the ratings oddness, esp with incoming rookies.

I am surprised they haven’t done more with franchise over the years. The CPU trading is better than it used to be because there used to be no trading, but the logic is screwy. It amazes me that a game programmed by one man (FoF) can do a better job than a company with EA’s resources.

Have you settled on a slider set yet? I keep tweaking mine, but sliders aren’t my thing.