prof101

A bad signing? Yes. But don't all managers make mistakes?

ps: I'm not sure what relevance this has to the topic at all.

The point of this thread is to [b] Outline possible improvements to the youth model, by looking at real world examples. I here cited Wenger, however I am sure this process stretches further than Arsenal.

In FM the youngsters are given to us. Their natural position is fixed (to a large degree). In real life youngsters are not "born" into positions. In FM they are.

In FM, like real life, players have certain attributes. In real life, say in Ashley Coles case he has the attributes to be a leftback, Wenger has the ability to train and mould such a youngster by looking at his abilities and then giving specific training to put the player in the position suited to his role. In FM, I feel the Youth module should be more flexible, asking the manager to be more involved in developing the youngster by judging what position he feels he would be best suited to. This would increase the immersion and in judging the player the user will have to think about what position the FRED is best suited to. :cool:

retired_sue_ishy

Chris Sutton, Paul Warhurst all started in defence and moved up front. Scholes started up front and moved into midfield. John Barnes was a flying winger and became in his twilight years a central playmaker.

Players change position all through their careers, but especially when they are still forming. So it would be good to get away from the "natural position", that is fixed at the start and never changes through their career, to reflect their changing physical attributes, training shcedules, desire to improve etc.

And if a manager decides he wants to make the most of his youngsters and build up this area of the club, he can invest time and money into developing them, rather than buying in players. At the moment, the manager simply doesnt seem to me to have enough levers to make an impact at the moment - a players path seems pretty fixed the moment they are produced. Sure they might not all reach their PA, but if you give them gametime and get good coaches in, they normally will. Which might explain why potential costs so much in this game, because its not a risk.

I would love to have a chance to identify potentially good players, give them specialised training and encouragement that suits them and really make a difference, like Wenger.

prof101

I agree 100% Baggio! :thup:

I agree that we don't have enough say currently in player development, all suggestions thus far would greatly aid the creation and development of our own youngsters.

Another issue we are raising here is the issue of development over a timescale.

Some players peak early, some take it steady and some peak late in their carers. I am not talking about positional ability (eg: when they have their best years) but more as *how* they develop.

In FM players will develop in the same way, to their potential bar catastrophic injury or lack of facilities.

In real life players with high potential whilst young sometimes do not make it, peaking early and fizzling out. (Jeffers?)

Some youngsters come "out of nowhere" and continue to improve at a fast rate. (Rooney?)

Some development is steady and gradual. (Milner?)

Some players develop "late". (Ian Wright)

Example

Say we had named these "types" A,B,C,D. Lets call them "Developmental Rates".

Each FRED would be assigned a similar developmental rate which would influence how fast hist attributes "improved."

This would add greater variation to FRED development, making it more unpredictable and thus less of a game of hide and seek where you send out your scouts and sign those with high star potential.

A crude representation of the "growth curves".

retired_xcallibr

Yep your idea is great and I'm sure it's doable.

I reckon youth improvement is gonna be FM09's big thing.

prof101

Lets hope so too eh :thup:

prof101

Anyone got anything else to add on this subject?

leftfoot

I like your idea in theory but I think in the game players have to be generated in such a way that the database remains balanced. For example as it stands say the game generates x wingers and y forwards at the season change over point. If I understand your idea (if I am getting the wrong end of the stick my apologies) then using your system the game might generate x+y players with the potential attributes to be wingers or strikers. What happens in the hypothetical extreme where every manager, AI and human, decides to train these players to be strikers instead of wingers every year. This would result in a shortage of wingers and an excess of strikers over the long term.

I also wonder what the impact would be for processing time with every in game manager analysing attributes, deciding on a position and then setting training to maximise the attributes to align with each position. Essentially every youth player would have to be assigned an individualised training scheme and I don’t believe the current AI is capable of critical analysis of player attributes. I think the way AI managers in 07 would pay huge transfer fees for players and select players in their first team based on CA/PA rather than attributes is evidence of this. One that springs to mind is on one 07 save game I played Conor McCormack had atrocious technical attributes, a 9 for passing, but had maxed out mental and physical attributes and was a first team player for Man Utd in the Premiership and Champions League.

Like I said I think the need for a balanced database would still require players being generated with attributes that give them the ‘potential’ to play in a certain position. So what would be the difference between

1. Your system:-
a player being generated with attributes that give him the potential to be a centre back and you deciding to train him as a centre back based on his attributes

or

2. Current System:-
A player being generated with the attributes that give him the potential to be a centre back and being assigned centre back as his natural position based on his attributes

Would 1 not just give the illusion of control?

Or have I just misinterpreted your idea completely?

retired_taustin

Not necessarily youth players, but i have noticed in the South African league that many players "naturally" play out of position, so my purchase of a rubbish striker winds up doing a job at central midfield over time.

retired_sue_ishy

Alurny - great ideas :thup: The idea of different development rates that allow the game to generate late bloomers like Ian Wright, Jamie Carragher and Luca Toni would be excellent. Fingers crossed for FM09!

isuckatfm - I agree you would have to be careful not to unbalance the database. But I think this could be done. For one thing, the game should still generate players with attribute profiles similar to now, and it shouldnt be hard (I dont really know to be honest!) for the AI managers to identify most players best positions using the exisitng attribute/position scouting rating.

On the illision of control thing, the difference would be that you would have more options to mould the small percentage of FREDS that have attributes that dont match their natural positions. At the moment you can retrain, but never to natural over time!

Oh well, its been fun guys!

leftfoot

That brings up something I have thought about posting on, namely what exactly is the effect in the match engine of player's position ratings. I understand the concept of a player's comfort in a given position in real life but how does the game replicate this?

Does the positional rating modify attributes in the engine e.g. does a defensive midfielder with positioning 20 when played at centre half have his positioning attribute reduced by some value related to his positional rating? The furthest I got a player to was 19/20 for a non natural position. If attributes are modified then how wide the scale is would impact on this effect.

TBH I don't see SI disclosing exactly how it works but the way researchers talk about it in various threads in the Data Issues forum I reckon the 'attribute modification' as described above might be how it works. It is probably applied to mental attributes only as a player who can dribble can do so regardless of where he plays. That's my theory on it anyway.

From looking at FM modifier it is possible to get to 19/20 which is close to 20 = natural. It just depends on how this relates to attributes as to how it affects their match engine performance. It might be a % so if you have an attacking midfielder with off the ball 20 but his positional rating as a striker is 15/20 then when you play him as a striker the match engine treats him as a striker with off the ball 15?

To Alurny's growth curve suggestion :thup:. Did you study maths or engineering at college .

At the moment it kind of implements this but in a simpler way. This might be affected by which leagues are active but in my games with the English leagues only active some regens appear at older ages than 16 in non active leagues and it tends to be these that don't reach their potential ability.

Also players often appear with different CAs. I've used Genie Scout (yes I cheat like a son of a bitch) to track player development to try to figure it out and I've seen players appear with similar PA but large differences in CA. So maybe it is in there but is too subtle to be obvious on the surface. I have also seen players gain CA points at a much greater speed than others, but again it isn't as obvious just from looking at attributes in the game.

All of the above was in 07 so whether things have changed in 08 I can't say.