Player Development
Mon Nov 08, 2004 6:50 am
Something disturbing to share or discuss, I guess - it seems player development is completely random again.
I'm in year 2 of my Knicks dynasty (I simmed the whole first year, but am playing all games now), and after some changes, my roster looks like this:
C Eddy Curry (traded for Kurt Thomas in a multi-player trade)
PF Mike Sweetney
SF Wally Szczerbiak (traded for Tim Thomas in a multi-player trade)
SG Jamal Crawford
PG Stephon Marbury
---------------Bench----------------
SG Allan Houston
SF Penny Hardaway
PF Jerome Williams
PG Troy Hudson (he came over along with Wally)
C Nazr Mohammed
SF Trevor Ariza
C Mark Becker (rookie)
----------------IR------------------
Vin Baker
Shandon Anderson
Now, I'm thru 15 games of my 29-game-season, and Curry is a 67 overall averaging 20PPG and 12RPG; Sweetney as a 62 overall has 12PPG and 10RPG atm.
For the heck of it, I decided to sim the rest of the season (I still intend to play it, this was just a test) to see if the ratings would change. Well, Curry's averages dropped slightly through the simmed games, but nothing dramatic; same for Sweetney. But at the start of the next season, Curry had dropped to 66 overall and not improved in a single category he was good at (inside scoring, offensive rebounding etc.). Sweetney went up by 1, my rookie center gained 4 overall despite playing only occassionally.
So judging by this, it looks like player development again is not based on actual performance. Curry should prolly have gone up 4 or 5 points overall, with the main improvements coming in the categories mentioned above.
I know these are "just" ratings, but they do have an influence on the game, so I was hoping that we'd finally see realistic results here... I mean, they have a big influence on simmed games for example.
Has anyone else experienced this yet?
I'm in year 2 of my Knicks dynasty (I simmed the whole first year, but am playing all games now), and after some changes, my roster looks like this:
C Eddy Curry (traded for Kurt Thomas in a multi-player trade)
PF Mike Sweetney
SF Wally Szczerbiak (traded for Tim Thomas in a multi-player trade)
SG Jamal Crawford
PG Stephon Marbury
---------------Bench----------------
SG Allan Houston
SF Penny Hardaway
PF Jerome Williams
PG Troy Hudson (he came over along with Wally)
C Nazr Mohammed
SF Trevor Ariza
C Mark Becker (rookie)
----------------IR------------------
Vin Baker
Shandon Anderson
Now, I'm thru 15 games of my 29-game-season, and Curry is a 67 overall averaging 20PPG and 12RPG; Sweetney as a 62 overall has 12PPG and 10RPG atm.
For the heck of it, I decided to sim the rest of the season (I still intend to play it, this was just a test) to see if the ratings would change. Well, Curry's averages dropped slightly through the simmed games, but nothing dramatic; same for Sweetney. But at the start of the next season, Curry had dropped to 66 overall and not improved in a single category he was good at (inside scoring, offensive rebounding etc.). Sweetney went up by 1, my rookie center gained 4 overall despite playing only occassionally.
So judging by this, it looks like player development again is not based on actual performance. Curry should prolly have gone up 4 or 5 points overall, with the main improvements coming in the categories mentioned above.
I know these are "just" ratings, but they do have an influence on the game, so I was hoping that we'd finally see realistic results here... I mean, they have a big influence on simmed games for example.
Has anyone else experienced this yet?
Mon Nov 08, 2004 7:45 am
I don't know if anyone has finished a season yet.
Mon Nov 08, 2004 8:28 am
I've finished a season with the clippers (in fact - two seasons) and I've seen pretty good improvement in players. I'm not disappointed at all. Everything has turned out the way I expected them to.
Mon Nov 08, 2004 11:32 pm
The player development can't be all stats... That would just make it too easy for a player to exploit the system... It would make it possible for the player to play 12 minute quarters and get really high averages for everyone and get huge raises in ratings for all.
The basic system for development has to have some sort of guidelines in order for it to work. It's not like Eddy Curry has become the huge star he was supposed to be in real life... Curry and Chandler were supposed to be the next David & Tim combo out there. It hasn't happened. Not all the young guys get to be good. Curry is a good guy already, he hasn't shown the improvement in real life to become a superstar...
I'd like to know how the training works... I'm prolly getting the game on wednesday. Damn sweet time it takes for the game to get on the shelves here in Finland... Just get a sad feeling looking at it for the consoles... It's so WRONG...
The basic system for development has to have some sort of guidelines in order for it to work. It's not like Eddy Curry has become the huge star he was supposed to be in real life... Curry and Chandler were supposed to be the next David & Tim combo out there. It hasn't happened. Not all the young guys get to be good. Curry is a good guy already, he hasn't shown the improvement in real life to become a superstar...
I'd like to know how the training works... I'm prolly getting the game on wednesday. Damn sweet time it takes for the game to get on the shelves here in Finland... Just get a sad feeling looking at it for the consoles... It's so WRONG...
Tue Nov 09, 2004 2:24 am
Metsis wrote:The player development can't be all stats... That would just make it too easy for a player to exploit the system... It would make it possible for the player to play 12 minute quarters and get really high averages for everyone and get huge raises in ratings for all.
The basic system for development has to have some sort of guidelines in order for it to work. It's not like Eddy Curry has become the huge star he was supposed to be in real life... Curry and Chandler were supposed to be the next David & Tim combo out there. It hasn't happened. Not all the young guys get to be good. Curry is a good guy already, he hasn't shown the improvement in real life to become a superstar...
I'd like to know how the training works... I'm prolly getting the game on wednesday. Damn sweet time it takes for the game to get on the shelves here in Finland... Just get a sad feeling looking at it for the consoles... It's so WRONG...
The training works. No deductions but only additions. However, sometimes the player will have no improvements.
Tue Nov 09, 2004 4:35 am
Metsis wrote:The player development can't be all stats... That would just make it too easy for a player to exploit the system... It would make it possible for the player to play 12 minute quarters and get really high averages for everyone and get huge raises in ratings for all.
Totally agree that it should not be based on stats alone. However, if a guy basically doubles his career averages, there should be some improvement, and I don't see that at all.
Tue Nov 09, 2004 10:37 pm
JohnnyKnoxville wrote:Totally agree that it should not be based on stats alone. However, if a guy basically doubles his career averages, there should be some improvement, and I don't see that at all.
Career averages are one thing, but as you said Johnny... The stats dropped off only a little bit after simulation, which basically means that you played him at his current level when it comes to how the CPU sees it. A little over, so actually it isn't that much higher than you would first think. A lot of players in the NBA are not playing at their full potential... And with limited playing time... This is where the problems rise when it comes to simulation etc. When you take one guy out of their real team and put him into another, how would he play then??? That's a big question and something that you will collide into almost immediately in Dynasty mode or atleast after the first season...
They have projected that Curry would play as you have played him so he's THAT good already... He doesn't need an upgrade to get to that level as he's already there.
Remember that the really high overall guys are good at almost everything and I see some great holes on Currys game like the defense, rebounding, and ball handling... Players like KG, Timmy D and TMac have them all down pretty well. They can score, they can rebound, they can pass, they can put the ball to the floor with confidence and they can hold their own in the defensive end as well. So a player without a good all around game could score like 20 ppg, but have an overall of 70... Overall does not correlate into scoring directly... Ben Wallace probably has a high overall and he scores like 10 points per game... If he would pick up his scoring his overall would be like 5-10 points higher...
If he scores well, it doesn't matter what his overall is... He might go up after the next season...
Wed Nov 10, 2004 9:10 am
First of all, all good points, but IMHO not exactly on topic for this discussion 
His averages dropped by 2.5 (scoring) and 1.8 (rebounding) during the simmed games. Seeing that I had played nearly 2/3 of the season (error in my first post, I am actually at game 18, not 15), he scored significantly less in those simulated games. So I played him at a higher level than the ratings put him at.
The overall rating was just an example. As I said, none of his ratings went up, even though his inside scoring should have gone up significantly judging by the way that I played him. Same goes for offensive rebounding (league leader until I started simming) and even passing since he had about 4.5 apg in the games that I played. This is what bugs me, not the overall rating - I just posted that to give you guys a quick overview.
Metsis wrote:Career averages are one thing, but as you said Johnny... The stats dropped off only a little bit after simulation, which basically means that you played him at his current level when it comes to how the CPU sees it. A little over, so actually it isn't that much higher than you would first think.
His averages dropped by 2.5 (scoring) and 1.8 (rebounding) during the simmed games. Seeing that I had played nearly 2/3 of the season (error in my first post, I am actually at game 18, not 15), he scored significantly less in those simulated games. So I played him at a higher level than the ratings put him at.
Overall does not correlate into scoring directly...
The overall rating was just an example. As I said, none of his ratings went up, even though his inside scoring should have gone up significantly judging by the way that I played him. Same goes for offensive rebounding (league leader until I started simming) and even passing since he had about 4.5 apg in the games that I played. This is what bugs me, not the overall rating - I just posted that to give you guys a quick overview.
Wed Nov 10, 2004 6:41 pm
All I can say is that this is something we have had to deal with basically since the start of the multiseason mode... It just sucks...
With your team, I would say that about the 13 points for Curry would be nice, with Wally, Jamal and Stephon taking most of the shots and Curry getting the ball only occationally... You have guys like Penny and Houston on your roster as well that are big perimeter threats... In reality Curry wouldn't score as much as you have him scoring...
I'm getting the game today... I'll see what sort of a system it has and I'll get back to you... You can try this as well... You have Stephon and Jamal as starters... Try trading your guards for like Jason Kidd and someone more defensive shooting guard and see how he plays then... If he then still scores 13 points per game, then it's the same gosh darn system we've had before, but if his stats go up as he gets more shots and more responsibility then the system is actually working... You could actually try this to a more extreme... Get someone like Brevin Knight to be your point and see how the stats evolve. He can't score one lick and this should be seen as a raise in scoring for other players... It's with things like these you can test the simulation system... They say they've made some fixes...
Create like two saves from a singe dynasty and with the other one trade the back court to players that don't score as much, even if you get a much worse team and simulate it to see how the stats are and then compare them to your "real" teams simulated stats... Write them down on a piece of paper or something...
This is just me guessing, but I would think that they have lowered the offseason player upgrades as the training now only produces positive results, so you'll have to train your guys to get the same results as before... This is just guessing though, but it would seem a better system so you wouldn't get everyone up way too fast and have a team full of superstars in a couple of years, no matter who they are...
So the new Q is... Have you trained your team much???
With your team, I would say that about the 13 points for Curry would be nice, with Wally, Jamal and Stephon taking most of the shots and Curry getting the ball only occationally... You have guys like Penny and Houston on your roster as well that are big perimeter threats... In reality Curry wouldn't score as much as you have him scoring...
I'm getting the game today... I'll see what sort of a system it has and I'll get back to you... You can try this as well... You have Stephon and Jamal as starters... Try trading your guards for like Jason Kidd and someone more defensive shooting guard and see how he plays then... If he then still scores 13 points per game, then it's the same gosh darn system we've had before, but if his stats go up as he gets more shots and more responsibility then the system is actually working... You could actually try this to a more extreme... Get someone like Brevin Knight to be your point and see how the stats evolve. He can't score one lick and this should be seen as a raise in scoring for other players... It's with things like these you can test the simulation system... They say they've made some fixes...
Create like two saves from a singe dynasty and with the other one trade the back court to players that don't score as much, even if you get a much worse team and simulate it to see how the stats are and then compare them to your "real" teams simulated stats... Write them down on a piece of paper or something...
This is just me guessing, but I would think that they have lowered the offseason player upgrades as the training now only produces positive results, so you'll have to train your guys to get the same results as before... This is just guessing though, but it would seem a better system so you wouldn't get everyone up way too fast and have a team full of superstars in a couple of years, no matter who they are...
So the new Q is... Have you trained your team much???
Thu Nov 11, 2004 12:15 am
Metsis wrote:With your team, I would say that about the 13 points for Curry would be nice, with Wally, Jamal and Stephon taking most of the shots and Curry getting the ball only occationally... You have guys like Penny and Houston on your roster as well that are big perimeter threats... In reality Curry wouldn't score as much as you have him scoring...
That's all nice and dandy, but IMHO not relevant to my dynasty. If I understand you correctly, you are taking real life into account here, but I doubt EA has a built-in stats collector that takes RL stats into account
JFTR though: Curry is third on the team with 225 shots attempted. Marbury is first with a little over 300, Wally second with 240. Crawford has taken less shots than Eddy, he's at 210.
The fact is that Curry is averaging 20 and 12, with loads of inside scoring and a bunch of offensive rebounds (I'm using pared's sliders, btw), so I would expect those ratings to go up after the season.
So the new Q is... Have you trained your team much???
Not really, I lack the dynasty points for that, and I use a lot of them for scouting. As the team is pretty balanced, I rarely get a statistical outburst from a player, so I usually only get the standard points.
I did do one individual training session for Sweetney though. 70 shooting - 10 each for the rest; his inside scoring went up by 1 point, that's it.
Thu Nov 11, 2004 3:35 am
Going away from the topic but iam just glad that the training camp bug has been fixed, i wanted to send Ariza though training but was worried that he would get worse but it turnout nicely so thanks EA
Fri Nov 19, 2004 5:50 am
I actually prefer the current system of improving rookies according to their potential, and not their PT. Their PT, however, DOES affect their production. I don't think it would be a good idea to base a player's improvements according to how they are played because it can easily be abused, and the player progression with be totally predictable.
I believe the training option is your best bet to improve players you want to improve the most. That is what I use. In my current dynasty, I'm focusing on making Chandler stronger. He's 72 str now, up from 62. And I don't do this randomly either. I'm strengthening Chandler because he gets boxed out and banged up too much in the post. He needs the boost.
I believe the training option is your best bet to improve players you want to improve the most. That is what I use. In my current dynasty, I'm focusing on making Chandler stronger. He's 72 str now, up from 62. And I don't do this randomly either. I'm strengthening Chandler because he gets boxed out and banged up too much in the post. He needs the boost.
Fri Nov 19, 2004 7:52 am
i love how player progress works in madden. would love to see it in nba live. the top 5 things imo that should be taken into account with player progression are....
1.playing time
2.training sessions
3.stats
4.injuries
5.age
now 1 3 and 5 should be able to work both ways. 2 should be progress only and 4 be only for a decrease in ratings. now none of the 5 should be the main reason. for instance if a player gets alot of PT but does not shoot very much and shows hustle he should improve. now if same player gets the pt and shooting 21% it should not help him much. another example would be if a player got injured for along time at the start of the season but by the end was healthy and you gave him several training sessions and plenty of playtime the injury should not lower his ratings. same goes for age, if a player still plays alot of minuts and you train him propper then just cause he turned 37 does not mean he needs a 4 point drop overall.
1.playing time
2.training sessions
3.stats
4.injuries
5.age
now 1 3 and 5 should be able to work both ways. 2 should be progress only and 4 be only for a decrease in ratings. now none of the 5 should be the main reason. for instance if a player gets alot of PT but does not shoot very much and shows hustle he should improve. now if same player gets the pt and shooting 21% it should not help him much. another example would be if a player got injured for along time at the start of the season but by the end was healthy and you gave him several training sessions and plenty of playtime the injury should not lower his ratings. same goes for age, if a player still plays alot of minuts and you train him propper then just cause he turned 37 does not mean he needs a 4 point drop overall.