Widening the range of player ratings

Temporary forum for the discussion of the Tuesday Morning project, to enhance the Madden 08 off-the-field experience.
Posts: 202
Joined: Wed Apr 18, 2012 7:32 pm

Widening the range of player ratings

Postby Drizzt_13 » Sun Jun 03, 2012 4:50 pm

This has been brought up quite a bit in the progression thread (because it's strongly linked to progression) but I think it needs a thread of it's own by now. I'm going to start with a brief list of the advantages of doing this and then talk about all the potential disadvantages or obstacles involved.

Pro's: - If we reduce the ratings of players across the board then players who excel in one area but are terrible in others become more valuable because the difference is more noticeable. There's no incentive to shift from pass coverage linebackers to run stoppers on certain downs if the difference is only 3-5 points of tackle and strength, but if it's 10 or so then we have a much stronger incentive.
- It makes superstars more noticeable, those few players who are truly deserving of high ratings will dominate their opponents more
- it make management more interesting as acquiring depth becomes more of a challenge now that the "basement" is not as high

That said there are some really difficult issues involved in reducing ratings across the board because the game was designed with a really small spectrum of players in mind.

1. Getting it right is a lot of work, and there are a lot of little things that could go wrong. Dropping ratings could make individual player too dominant and lead to ridiculous records being set. We would need to redo all the sliders because the ones most people use are probably going to generate ridiculous numbers if we use very different rosters. Even if we get the simmed sliders right it could have weird consequences for games that are actually played. The AI isn't smart enough to compensate for issues that actual NFL coaches do. If you have a receiver that is simply dominant they won't double team him enough and things could get really easy for the player. The AI doesn't use sub packages so they won't shift out good run stopping lineman for pass rushers in dime and nickel situations and so their teams could become really one dimensional. We would have to test through a lot of issues that could arise because we are trying to play the game at a different level then it was meant to be played at.

2. Not all abilities are comparative, we can reduce strength across the board because it only matters how strong you are compared to the guy across from you. However if we were to drop awareness across the board then when we played games we would just see some frustratingly stupid play. If we keep it high and drop everything else then the overall rating becomes a less effective indicator of ability. We have the same problem with throw accuracy and throw power. I'm sure if we messed with the sliders enough we could get normal simulated stats but played games would just be a pain to watch because the level of QB play would be so low.
Possible solutions to this would be to simply let QB's have high ratings compared to everybody else because the only 3 stats that matter for most of them are AWR, THA and THP and reducing those is problematic. If we do end up dramatically reducing awareness we could fix this on defense by just messing with the defensive awareness sliders so that everyone doesn't play like an idiot even though their numbers are lower then the game intended when the AI was written. But we don't have a similar slider for offense so we could see some truly idiotic lineman (who miss blocks enough already) and skill players.

3. We could loose a lot of tester support by doing this. We are at some point going to need people to help us with testing the app and if they have to do a whole bunch of little fixes to change their game they will be less likely to do so. If we calibrate the progression and other things around a lower rating system then people who don't want to use lower ratings won't use the manager app. If we finish the whole app and people don't use it very much that would be not only discouraging, it would also mean we would go longer before encountering bugs or potential issues.

4. Writing a whole bunch of rosters by hand is really time consuming. If we realize we've set everything too low and we need to change stuff then that means it's a lot of time that's been wasted. All the global editors I've seen let you reduce a statistic by a set amount across the board but that isn't what we need. We need something that will broaden the range not just shift it. If we had a program that could change the range, average value and standard deviation of a statistic that would be really handy cause then we could just shift stuff around until we got something that played well.

So basically what I'm saying is that lowering player ratings across the board is a good idea with some real benefits but we also don't know how it will change the game and if it does so negatively and we have to fix it we've lost a lot of time. If we can get more people who are willing to go rewrite a roster and fix issues that may arise then it's worth doing. But we also keep adding new features onto the app and never finishing the old ones and starting another big projectt now might not be a good idea unless we either have more help or don't have to do all the rosters by hand.

Widening the range of player ratings



Return to “Tuesday Morning Manager”

Who is online

Users browsing this forum: No registered users and 1 guest