Player Progression In Franchise Mode.

[PuddleMonkey]

I think it was 08 I played 23 seasons of franchise. Could have been 07, not sure. Anyway, after about 13-15 seasons all the offenses were rated upper 80's and 90's overall, while all the defenses were mid 60's and low 70's. It was completely broken. Last year showed tremendous improvement from years past, hopefully they continue to build on this. Many tweaks still need to be made, unfortunately I get the impression these tweaks didn't happen in 11.

All the attention seems to be going to on-line play, which blows for guys like me who have never played a game on-line before

[jamesduke42]

i agree completely with post.... if i draft someone in 6th rd and he has a 70 overall rating and makes the pro bowl with 25 sacks or more. his rating should be atleast a 94 or higher the next season.

[Deelron]

Just some comments, hope I don't butcher the quotes to much to keep the size down

Quote: Originally Posted by Boregard

1. Have to hide it - must hide it cuz knowing your 7th rd pick is A POT ruins the replay-ability of franchise I would love to see the game "randomize" POT for existing players at the start of each franchise -

I agree except I'm pretty sure that randomizing the players at a start of franchise would go over well with most of the franchise player base, but there'd be nothing wrong for it being an "on/off" option.

Quote:

2. Coaches need to be rated, so we can see it. Numbers or letters or something - graded in groups (DL, LBs, CBs, Ss, QB, RBs, WR,TEs, OL and K/P) at the minimum.

I agree they should certainly be graded but I'm not a big fan of them effecting progression, coaches are even harder to measure in terms of performance and generally speaking I think it'd be a drag to play to play a newer head coach who would drag down progression. I think the current version with the boosts to specific stats in combination with a varying ability to assess the potential value of players would both increase value to older, established coaches while not overly penalizing players who choose to play a lower rated one.

Quote:

3. User needs to have option to tell a player to "bulk up/slim down" in the off-season and it should have a potential impact on their ACC,AGI, SPD, STR and so on again no more than a point or two

This has a big potential for abuse, it'd have to be fairly heavily restricted and include some sort of risk, particularly if it's repeatable. If I did something like that for a long term player I could do something like increase my a players SPD by 5, which would have a rather significant effect on certain players.

Quote:

5. Players should have random "oh my gosh I get it now" points in their career. Not just early

For sure, there definitely should be more career paths (for players who improve) then "reliably improve->level off at peak->slowly decline, or the ones in there that are different should be more prevalent and visible.

Quote:

6. Progression should be completely and exclusively in POS specific (non "physical" attributes) and AWR.

I'd like to see a few outliers, but they really should be few.

Quote:

7. AWR, PRC, Route Running, PUR should never regress based on age...

I'm not so sure, I can see regression in those areas being possible for some RL reasons (a loss of desire to play for example), but I'm not really against what you're saying in general.

Quote:

8. Regression should be almost exclusively "physical" ratings...

I doubt I could agree with anything more.

[Deelron]

Quote: Originally Posted by jamesduke42

i agree completely with post.... if i draft someone in 6th rd and he has a 70 overall rating and makes the pro bowl with 25 sacks or more. his rating should be atleast a 94 or higher the next season.

A computer player with a 70 OVR rating would be exceedingly unlikely to get 25 sacks while not being on the human team, consequently the 25 sacks would likely come about by the human player causing a situation the AI doesn't handle properly to rack up the sacks. A system that rewards that would inevitably result in a completely unbalanced franchise, as a human player could just play longer games and get higher stats then the CPU sim engine.

The fundamental question is what actually causes the stats to occur, does the player get good stats because he's rated well (in real life he'd have better football skills, the core of what the ratings are supposed to be for) or does he get good stats because the human player can manipulate the game at will (change settings/manipulate inconsistent AI) to get whatever result he or she desires? In actual football players don't play better because they had good stats the year before, they play better because they have a higher skill level then comparable players.

On a side note, if that player had 25 sacks in a HUM played season, went to a 94 but then managed 1 sack the next season that was completely simulated (no injures), should he then drop back to a 70?

[MikaV]

Bulking up / slimming down shouldn't change any physical attributes - it should change weight, which is a coefficient of these attributes. Just becouse Kris Jenkins is cutting down this offseason, his strength wont go anywhere, his body control wont change, but he will probably be a bit more agile since he's lighter. He might get pushed back a little more since he's lighter. But his physical attributes have not changed at all. Im not sure if in madden the weight rating is there just for cosmetic purposes, but im quite sure i've read that locomotion takes weight into account - not sure if that covers linemen though.

This also covers another topic altogether.

A lot of people want a MMO-style progression, meaning the more stuff you do the better you get. A lot of people seem not to grasp the concept that a couple of people have outlined already: Absolute ratings and perceived ratings.

Now, take a guy like Miles Austin.

Coming into last year, people perceived him to be a mid-of-the pack WR - say, an 78. That was the perception of his skills. After he played a lights-out season, people look at him again, and say, man, this guys good. This year, he'll be, what, 90 OVR? Again this is the PERCEPTION of his skills.

The reality of the matter is that he always was what you would call a 90 OVR player in real life, and played to that ability. Our perception was wrong.

This does not work in a videogame such as Madden, since in the game, we are dealing with ABSOLUTE skills. Not perception of these skills.

Now, in real life, Austin played like a 90 OVR WR for a couple of years. Our perception of his skills went from 70 to 90 as we saw more of him.

In madden, when he had 70 OVR, he played like 70 OVR. There's no error, there's no wiggle room, that was his absolute skill level in Madden.

Now, just another example. In real life, if an unknown DE has a 20 sack season, he will be better rated in the next Madden. Not because he got better, but because our PERCEPTION of his ability changed. In a videogame, the system is different. If a 70 OVR DE has 25 sacks, he had them with a 70 OVR.

This is quite hard to explain in easy terms, but i wish guys on the other side of the fence would take a minute and think about this side of the argument.

I'm not saying you should agree with me, i'm just saying that for the sake of this conversation, we have to make a real line between percieved ability and absolute ability, and we cant talk about these two things in the same sentence, IF we want a realistic progression system.

And also, if you would like an RPG-type progression system, thats absolutely fine, but my point is it's not realistic.

[oneamongthefence]

Quote: Originally Posted by Deelron

There's only 15 guys to ever go over 1800 also btw. The point I was trying to make is that in the Madden sim engine (as it is in real life) it's fairly rare for a non-high rated player to get that many yards, the typical culprit is a human player racking them up against an inferior computer then releasing the player. I'd argue it's actually more effective for the CPU AI to avoid that player then to respond to those stats, particularly given that Human players rarely sign based off of in-game stats over ratings.

Exactly, we look at abilities and overall pretty much when making signings so why should the cpu be forced to look at stats. especially when we as humans can artificially inflate stats due to sliders and bad ai. seems fair the way we have it now. besides having potential ratings helps keep the league balanced in terms of talent. In madden 05 within 5 years the number of 95+ overall players quadrupled and it was so easy to build a team of superstars. completely unrealistic imo.

[Deelron]

Quote: Originally Posted by MikaV

Coming into last year, people perceived him to be a mid-of-the pack WR - say, an 78. That was the perception of his skills. After he played a lights-out season, people look at him again, and say, man, this guys good. This year, he'll be, what, 90 OVR? Again this is the PERCEPTION of his skills. The reality of the matter is that he always was what you would call a 90 OVR player in real life, and played to that ability. Our perception was wrong. This does not work in a videogame such as Madden, since in the game, we are dealing with ABSOLUTE skills. Not perception of these skills. Now, in real life, Austin played like a 90 OVR WR for a couple of years. Our perception of his skills went from 70 to 90 as we saw more of him. In madden, when he had 70 OVR, he played like 70 OVR. There's no error, there's no wiggle room, that was his absolute skill level in Madden.

Yep, really dead on, the ratings for Madden for franchise have to accomplish two very different goals, one to represent a player as we know him to based off of actual accomplishments and observation, and the other is to simulate how a player is going to develop in the future.

The first (initial) ratings are reasonably debatable, potential is really a crapshoot. Take two QBs, and we'll use QB rating (a unreliable stat, but it's a fairly simple one for comparison).
First two full seasons starting
QB A - QB ratings 61.9, 79.4
QB B - 76.9, 67.5

One would argue based on those numbers alone that those two QBs should be rated fairly similarly in the next Madden game. QB A is the first 2 years of Kerry Collins career, QB B is the rating for the 2nd and 3rd year (he sat the vast majority of the first year) of this years cover athlete Drew Brees. If you're trying to be unbiased in making ratings it'd be very hard to say QB A should have a higher potential then QB B, there's just little to no evidence to back that up, except one turns into a solid lower tier starter and the other has 4 HoF level years in a row. There's just no way to tell.

(Btw Collins was also drafted earlier, 5th pick in the draft vs. pick 1, Rnd 2 for Brees).

[PGaither84]

Quote: Originally Posted by Deelron

Yep, really dead on, the ratings for Madden for franchise have to accomplish two very different goals, one to represent a player as we know him to based off of actual accomplishments and observation, and the other is to simulate how a player is going to develop in the future. The first (initial) ratings are reasonably debatable, potential is really a crapshoot. Take two QBs, and we'll use QB rating (a unreliable stat, but it's a fairly simple one for comparison). First two full seasons starting QB A - QB ratings 61.9, 79.4 QB B - 76.9, 67.5 One would argue based on those numbers alone that those two QBs should be rated fairly similarly in the next Madden game. QB A is the first 2 years of Kerry Collins career, QB B is the rating for the 2nd and 3rd year (he sat the vast majority of the first year) of this years cover athlete Drew Brees. If you're trying to be unbiased in making ratings it'd be very hard to say QB A should have a higher potential then QB B, there's just little to no evidence to back that up, except one turns into a solid lower tier starter and the other has 4 HoF level years in a row. There's just no way to tell. (Btw Collins was also drafted earlier, 5th pick in the draft vs. pick 1, Rnd 2 for Brees).

I just wanted to say that when it comes to your example and hand crafted rookies, I think I drafted a Drew Brees type QB. At the end of the 2ed round of the draft I picked up a QB who is a 78OVR with B potential. He has 90THP and 87THA which translates into around 87 accuracy all around. My understanding is that if he were, say, an A potential instead of a B potential, he could "ride the bench" and raise his awareness and throwing accuracy ratings much like Drew Brees. At this stage in his career, Drew is thought of as the most accurate QB in the NFL. this wasn't the case after his first three years in the league. Instead, Philip Fivers was drafted [or more to the point Eli manning was drafted and traded to the Giants.] I am looking forward to seeing how my new QB pans out after riding the bench for one or two years.

When it comes to hand crafted rookie it isn't a problem because the only opinion that matters is Josh Looman's. When it comes to real life players that is a different story. I like the system we have in place for progression, and it gets better, imo, with hand crafted draft classes who's potential is lessup to debate compared to a real life person.

Think of it this way, in Madden you set in stone a player's potential. As franchise palys out you are playing out an "alternate history." A video game history. Don't worry about trying to compare that to real life. Where would Wes Welker be if the Patriots hadn't of traded for him? It's likely he would have stayed on the Phins roster as the 3rd or 2ed stringer. they may have even chosen not to resign him, we would have never really heard of him. His skills along with a great QB and team that took advantage of those skills combine together to turn Wes into the Superstar he is today.