Jump to content
Sports Interactive Community
Thorne

Newgens making the game unplayable.

Recommended Posts

1 minute ago, noikeee said:

You clearly aren't understanding us at all, and quoting a SI staff member in several different colors isn't going to change any of that. The 2nd paragraph you're quoting is completely pointless because Thorne has provided statistics obtained using 3rd party scouting tools (just like I did for FM16), and not the player search tool in game which is limited to scouting range.

I also find this a poor explanation:

"Either the AI or the human can fall short here and create a surplus/deficit of certain types of player."

Given that the human user is only ever taking over 1 club at a time, and the AI is managing all the remaining thousands of clubs, I don't understand why can it be the human's responsibility for a DB-wide issue. It's clearly either the AI failing to train the players in a realistic balanced way, and/or the generation of players at the beginning.

Whether we play the game well to properly retrain players or develop youth players properly, is a completely different thing that has absolutely no relation to this issue. The fact of the matter is if you take over a club in the first 5 seasons you've got immediately loads of ready-made IFs in the transfer market without needing to train anyone, but if you do it 20 years into the future they have become very rare. This is a balancing issue, not a feature or the users not understanding how to play the game. The option to develop players rather than buy ready-made is always there, but it shouldn't suddenly become the only option years into the future.

I'm gonna wait and see what happens at the end of this post.  I clearly understand what you guys are saying don't get me wrong.  if i can't find the fruits that i want what do you think i'm going to do ???/ yes i will make my own farm and grow them simple as that.  The laws of supply and demand.  Regardless of the DB do you expect to find players like Ronaldo every season ???

Maybe the players are already generated with other better stats only requiring you to form him in his new role.  Looking in one direction doesn't mean your looking at everything.  This is an interesting post and read and can't wait till we get more info.

As for others that want to know, i'm not bragging on hear but i'm yet to have that issue because it's history for me.  I've learnt to deal with it and get young talent where i can morph them to play into x2 positions at the very least.  Let alone when i do find some of these players they are wanted to big teams in europe for well over 5 million just after i have picked them up.  I never keep the same team for more than 2 seasons and the talent and profits coming in are bombastic.  hence why i agree with seb.

Share this post


Link to post
Share on other sites

@fmfan74

If the Dutch newgen AMR/AMLs are consistently being developed with a very high crossing ability and a mediocre off the ball ability, then there is indeed lack of realism in the game as it ignores a feature which is fundamental to Dutch football, has been for decades and probably will continue to be fundamental for the foreseeable future.

A newgen who is 27 years old in 2029 must have been born in 2002.
So he will be 14 years old when the game starts in 2016 and would have received 5-6 years of youth training with methodology that is still valid today.

The game being dominated by wide players who have a crossing rating which is circa 5-6 points higher than their off the ball movement is a paradigm change because the reverse is true today, especially at the top level.

If there is to be a paradigm change with respect to player development, as has been voiced by the 'variation from the database' argument, we shouldn't fully observe it until at least 15 years after the paradigm change occurs.  However we are observing this within 2-3 years of game start, which means the paradigm change occurred at least 5 years before the game start.  Hence the lack of realism in this regard.

 

Share this post


Link to post
Share on other sites
56 minutes ago, fmfan74 said:

I'm gonna wait and see what happens at the end of this post.  I clearly understand what you guys are saying don't get me wrong.  if i can't find the fruits that i want what do you think i'm going to do ???/ yes i will make my own farm and grow them simple as that.  The laws of supply and demand.  Regardless of the DB do you expect to find players like Ronaldo every season ???

Maybe the players are already generated with other better stats only requiring you to form him in his new role.  Looking in one direction doesn't mean your looking at everything.  This is an interesting post and read and can't wait till we get more info.

As for others that want to know, i'm not bragging on hear but i'm yet to have that issue because it's history for me.  I've learnt to deal with it and get young talent where i can morph them to play into x2 positions at the very least.  Let alone when i do find some of these players they are wanted to big teams in europe for well over 5 million just after i have picked them up.  I never keep the same team for more than 2 seasons and the talent and profits coming in are bombastic.  hence why i agree with seb.

Sigh. We already do this. I already develop players as well as I can from a young age and considering their personalities, adapt to play with different tactics if I don't have the players necessary for my regular tactics, retrain players whenever appropriate, hire players from different positions with the attributes I want in order to retrain. I already adapt as well as I can.

None of this negates that there's a fundamental balancing issue with the regens in the database. This is about realism, not about a human manager not having the players we want. Plenty of times I've managed sides that didn't have the players I want, or the resources to get them. I just adapted and accepted the challenge and that's fine. But this is a simulation, it's supposed to be realistic and this is not a realistic challenge. I'm speaking almost from a beta tester perspective, not from a manager perspective, merely reporting an issue with the simulation. I'm not looking for excuses as to why I'm not winning or anything like that.

Share this post


Link to post
Share on other sites

There's something related to retraining of youngsters which is "tutoring".

I haven't seen in my save any AI teams across the world doing tutoring for their players and I'm now approaching the end of first season!

Is this a bug or am I missing something?

Share this post


Link to post
Share on other sites

Maybe the newgen creation is linked to the manager/clubs preffered formations when the youth intake is generated?

I didn't bother to check Inside forwards as i never used one for my 5-2-2-1 formation but I notice in my 2037 save that some top teams use either 4-1-4-1 or plain 4-4-2 formations in which there is no place for an inside forward. 

If the AI manager doesn't have a preffered formation that includes inside forwards they will never train one to that position.

Share this post


Link to post
Share on other sites

@Champion Please keep this on topic and you already made that post in a different thread.  If you think you have found a bug please raise it in the bugs forum, or if you want to discuss it further start a thread.

Share this post


Link to post
Share on other sites
On 3/30/2017 at 09:58, aderow said:

Yes. That is very much is an opinion. Agree to disagree and all that jazz.

I'm not the one who's going to investigate so I don't see the words of a random stranger on the internet like me is going to impact what SI choose and not choose to investigate. I don't see how giving an opinion on something is all that detrimental in the grand scheme of things. Just like you've given yours.

And btw, in a game where every save progresses uniquely, one example from one save does not amount to fact. At least not outside that one save. I am not dismissing what you've stated. I'm not that rude. Most of the time at least. All I'm saying is that it could be an indicator of an issue or it could be an outlier. The only ones who could know are SI.  And you've already brought it to their attention in the bugs forum.

Once again, if there an issue, I hope it is addressed. Every little improvement to the game, I am all for.

Also it seems you missed my last question in my previous response.

A left footed player will not pass or shoot the ball with his right foot as effectively as a right footed player and that's a fact. But maybe we've misunderstood each other.

It's detrimental because SI might not address the issue despite being aware of it if they believe people don't know or care about it. A similar issue was raised for FM16 by noikeee but it appears nothing changed.

The issue has now been confirmed by an actual researcher so hopefully we can stop discussing whether it exists but rather how to fix it (not you specifically but other people who might want to give their opinion).

And I did miss your last question, apologies. I look at his attributes not what AM tells me, although he does usually agree with me. (not always though!)

Share this post


Link to post
Share on other sites

@perpetua 
Finally, yours was the post I was waiting for. Thanks for your confirmation (and explanation) of the problem. (Also thanks to @noikeee)

If you read my bug report you probably know I suspected player generation over training but your youngster got me thinking.
Was he trained as a winger because his key and recommended attributes for a winger were higher on average than those for an inside forward? Since winger has fewer such attributes this is the case with almost all youngsters and if AI trains to get the best out of a player it could very well explain why the issue exists. I'm also wondering why the AI manager didn't want to retrain him to the other side but the assistant did.

Another thing that I've noticed is when you take over from an AI manager the individual training that he set up does not reset. I've created a few new managers to see how clubs are training and this is what I found out:
- AI will train a player as an inside forward if he fits its tactic
- AI will retrain an inside forward to ML (instead of MR) if he doesn't fit (442)
- AI will retrain a player to a totally new position
- most of the first team is training for a specific role
- NO youngers (u18) are being trained for a specific role, unless they are being retrained to a new position (so when does the AI actually start 'role' training the youngsters?)
- AI will retrain a youngster to a totally new position as an inside forward!

 

I plan to do some more testing. I'll create a few youngsters with different sets of attributes to see which ones get trained as an inside forward. Hopefully that gives some more insight about what the problem could be.

Edited by Thorne

Share this post


Link to post
Share on other sites

One other question on this, has there been any movement in tactical changes during the time of the 20+ years of the test save?

Over time it's quite possible for different tactical trends to be adopted, so for example perhaps a 5-3-2 (and variants) has been successful and therefore other teams begin to adopt it.  Clearly in that kind of situation Inside Forwards would be needed less, which may impact both the generation of new "inside forwards" and their training.

No idea if this has had impact here or not, just exploring another possible variable.

Share this post


Link to post
Share on other sites
1 hour ago, herne79 said:

One other question on this, has there been any movement in tactical changes during the time of the 20+ years of the test save?

Over time it's quite possible for different tactical trends to be adopted, so for example perhaps a 5-3-2 (and variants) has been successful and therefore other teams begin to adopt it.  Clearly in that kind of situation Inside Forwards would be needed less, which may impact both the generation of new "inside forwards" and their training.

No idea if this has had impact here or not, just exploring another possible variable.

I'm sorry but this is a nonsense argument

Share this post


Link to post
Share on other sites
17 minutes ago, Trugdish said:

I'm sorry but this is a nonsense argument

Why? @herne79 isn't making an argument he is simply exploring a possible contributing reason as to why there is a shortage of ready made IF's after a while. 

Please note that he is not saying that this is the reason. In fact he is specifically pointing out that he has no idea if it have any effect or not. He is just saying that a change like that may contribute to the problem.

If it was as simple as "After a while there is a shortage of ready made IF's because of X" this would not be an continuous problem because it would be easy to fix, but this is not that. This is "After a while there is a shortage of ready made IF's and it may be because of x,y and Z is causing knock on effects that contribute to the issue.".

Edited by dieu

Share this post


Link to post
Share on other sites
2 minutes ago, Trugdish said:

I'm sorry but this is a nonsense argument

Who's arguing?  I'm exploring possible variables.

To be clear, the generation of new young players is influenced by several factors, one of which is the defined preferred formation that all staff members have.

Over time those preferred formations can change as tactical trends develop in game.  They can also change as staff leave clubs and are then replaced by staff who may have different preferred formations.  Therefore the generation of new players can be impacted.  The training of those young players may also be influenced due to changes of tactical trends over time.  You might not train an inside forward if you didn't need one, to continue the example.

Note that I am using the words "can" and "may" here, not "will".  Possible variables, not definite variables and certainly not deliberately argumentative.  The OP has posted a set of results, I'm interested to know how such variables may (or may not) have had an influence, or even if it's been considered.

Share this post


Link to post
Share on other sites
4 hours ago, Thorne said:

 

If you read my bug report you probably know I suspected player generation over training but your youngster got me thinking.
Was he trained as a winger because his key and recommended attributes for a winger were higher on average than those for an inside forward? Since winger has fewer such attributes this is the case with almost all youngsters and if AI trains to get the best out of a player it could very well explain why the issue exists.

If I am reading this correctly then it makes 100% sense.

I gave up playing with IFs on FM17 a few moths ago as I was sick of searching for them once the regens take over. FM16 and previous versions like you say dont provide the quantity of players in that role, which i could accept as the AI teams didnt really use them till FM16. FM17 it seems that most AI teams play a 451 and require an IF or winger, yet they are still rarely produced. I cant see why its not been picked up by SI as it very obvious. and the fact that people are shooting down @Thorne in this thread  is crazy. Regens have always been crap, and very much and after thought. 

A bit off topic but surely the youth team squad that you get to pick from on regen day should cover all the positions, and be a match squad. 2 gk 2 rb 2 lb 4 cb etc... as they have been playing virtual football since the u9's team for example. 

 

Share this post


Link to post
Share on other sites
5 hours ago, herne79 said:

One other question on this, has there been any movement in tactical changes during the time of the 20+ years of the test save?

Over time it's quite possible for different tactical trends to be adopted, so for example perhaps a 5-3-2 (and variants) has been successful and therefore other teams begin to adopt it.  Clearly in that kind of situation Inside Forwards would be needed less, which may impact both the generation of new "inside forwards" and their training.

No idea if this has had impact here or not, just exploring another possible variable.

From what I have observed, the norm is still 4 man defenses with most teams using 4411, 4231 Wide or 4231 DM Wide, 4141DM or 4123DM.

While formations with MR/ML positions are used quite commonly and probably more commonly than in real life today, that shouldn't mean IFs need to go extinct since the WM role is essentially the MR/ML version of the IF role (given that the WM cuts inside and is more of a passer than a crosser).
What I see instead are Wingers with very high ratings for pace and crossing and relatively low ratings for passing and movement.

In 2036:
I can identify 13 wide players with 15+ ratings in both passing and off the ball.
On the other hand, there are 215 wide players with 15+ ratings for both crossing and pace.

So the wide players available 20 years into the future are those who are more suitable to kick, rush, cross type of football than the football which attempts to control the game by meaningful possession which we see today.

They go hand in hand with the plethora of playmakers available in midfield with industrious central midfielders or B2Bs which balance technical ability and physical ability having gone almost extinct (even though, I would argue, they are the norm today).

Share this post


Link to post
Share on other sites

Thanks @perpetua.  I rarely (if ever) go that deep into saves myself, so have no conclusions of my own to draw.

It's interesting to see how databases develop over time and shifts in player abilities/roles along with the impact (if any) potential variables can have.  On the face of it results presented do seem to be at odds with SI's own tests - exactly why that might be is probably now up to SI to look into further.  Good to discuss variables here though.

Obviously trying to assess exactly how staff preferred formations may alter over time would be incredibly difficult and laborious to assess, and even then we wouldn't empirically know the magnitude of how that may impact new player generation without SI sharing their algorithms.  Still, something to keep in mind I guess.

Share this post


Link to post
Share on other sites
14 hours ago, herne79 said:

One other question on this, has there been any movement in tactical changes during the time of the 20+ years of the test save?

I checked the top 100 currently employed managers (by World Reputation) in 2040 and these are their top 5 preferred formations:

4231 Wide (19), 422 (18), 4123 DM Wide (15), 4231 DM Wide (8), 4222 DM Narrow (8)

Looks like the issue has nothing to do with tactics.

Interestingly enough, only 11 of those 100 managers are newgens and all 11 are ex-players.

Share this post


Link to post
Share on other sites

Having read this & the other threads on the subject I've not seen anything to suggest that there is a clear flaw in the way that new players are generated, what has become clear though is that the way players are trained is more likely at fault & to identify any development patterns will require more analysis than a two point comparison.

Share this post


Link to post
Share on other sites

If there is such a lack of inside forwards in long term games an ingame editor can be used to edit the wingers position so they can be effective in either flanks.

Now it will up to the AI if they can use the inside forwards properly. 

it's the only workaround i can think of.

Share this post


Link to post
Share on other sites
9 hours ago, Barside said:

Having read this & the other threads on the subject I've not seen anything to suggest that there is a clear flaw in the way that new players are generated, what has become clear though is that the way players are trained is more likely at fault & to identify any development patterns will require more analysis than a two point comparison.

Yup. Looks like the most likely conclusion from Thorne's much more detailed analysis in the bugs forum isn't that the game generates hardly any players capable of playing on the opposite flank from their footedness (unlike Noikee's thread from last year when only 6% of newgens accomplished or better at AMR were left footed), but that they tend to develop with striker/winger templates so finding one that can pass is relatively rare (the AI not retraining its AMCs to play there doesn't help). So you get Martials, but virtually no Matas and an AI with a young Mata-type in their squad is very unlikely to try to teach them to play wide

How and why the AI is training players in particular roles (and whether actually uses the IF role that much when playing formations that suit it) is something SI need to look at in detail because we can't really evaluate it ourselves

Edited by enigmatic

Share this post


Link to post
Share on other sites
On 3/31/2017 at 05:00, Thorne said:

A left footed player will not pass or shoot the ball with his right foot as effectively as a right footed player and that's a fact. But maybe we've misunderstood each other.

It's detrimental because SI might not address the issue despite being aware of it if they believe people don't know or care about it. A similar issue was raised for FM16 by noikeee but it appears nothing changed.

The issue has now been confirmed by an actual researcher so hopefully we can stop discussing whether it exists but rather how to fix it (not you specifically but other people who might want to give their opinion).

And I did miss your last question, apologies. I look at his attributes not what AM tells me, although he does usually agree with me. (not always though!)

No. No it's not. You think it's a fact. I think it isn't. That's how opinions work  :brock:

But seriously, if the issue's been confirmed, then great. Hopefully it gets sorted.

Trust me if SI put that much stock into what players say on here, we wouldn't have collision detection and farrows and barrows would still be a thing. 

Soyou have your own criteria for IFs. You sure you're just not too strict when looking for IFs? (Just playing devil's advocate at this point)

Edited by aderow

Share this post


Link to post
Share on other sites
10 hours ago, Barside said:

Having read this & the other threads on the subject I've not seen anything to suggest that there is a clear flaw in the way that new players are generated, what has become clear though is that the way players are trained is more likely at fault & to identify any development patterns will require more analysis than a two point comparison.

That was my initial feeling too.

Share this post


Link to post
Share on other sites
On 4/1/2017 at 09:57, Barside said:

Having read this & the other threads on the subject I've not seen anything to suggest that there is a clear flaw in the way that new players are generated, what has become clear though is that the way players are trained is more likely at fault & to identify any development patterns will require more analysis than a two point comparison.

I imagine that is something that folks at SI is better positioned to investigate if they convinced by the arguments presented here that something goes awry during player development.

My conjecture is that the AI needs to be smarter when assigning training roles.

For example, players should not be trained as a defensive centre back or a defensive full back.  These roles mainly put emphasis on physical development in comparison to technical or mental development.  In my opinion, players don't get trained to be defensive centre back or defensive full back, they simply end up suitable for these roles because they don't have the potential to be able to sufficiently develop technically or mentally but have the drive and/or genetic disposition to develop physically.

In the game, when you take a 170 PA central defender and put him in a defensive centre back regime, he ends up with 17-18's in all of his physical attributes and 12-13's for relevant technical and mental attributes.  In reality, if this player was trained as a defensive centre back, he would probably cap out at 130-140 CA at a maximum because he would reach his physical capacity (individual's capacity, not maximum capacity of 20 for each attribute) before reaching his overall potential ability.  

Since the game doesn't, to my knowledge, differentiate between physical potential, technical potential and mental potential we end up with very high ability players who are most suitable for limited roles.  One way to prevent this is by ensuring the AI doesn't train high potential players in limited roles and turn them into athletes instead of footballers.  If this is done, perhaps we won't need so many high potential players, which increase the quantity of high quality players in the long term (this, in my opinion, ruins gameplay in sub-top leagues like Turkey due to the relative ease of access to 130-140 CA players and typically big teams turning into mid-table sides as a result - but that's a whole other issue).

Similar to the example with defenders, in my opinion, wide players shouldn't always be trained as wingers.  The objective should always be to develop a player in a more well rounded role like IF or WM or AP or WP.  If the player doesn't have the capacity to develop mentally or technically, then he becomes a winger who relies on his pace and his crossing.  Alternatively, he ends up as a full back/wing back.

In midfield, more rounded CM or B2B role training should be preferred.  These guys may end up as BWM's if they can't develop technically.  Only those newgens who are created with superior vision should be put on playmaker roles.  Since pretty much every young midfield player is put on a playmaker role, we get plenty of players suitable for AP or DLP roles.  However these players are acqually quite difficult to train since creativity, as proxied by vision, is not so easy to train

So, in summary, the game needs to stop training high potential players in limited roles and instead prefer roles which will develop players to be well rounded players.  With the current AI training assignments, I imagine Ronaldo would have been turned into a winger who would look to get to the byline and cross the ball with every opportunity rather than the very dangerous goal threat that he actually became.  Goals are more valuable than assists.  

If you have a player who could become a great goalscorer, why would you turn him into someone who puts his blinders on and consistently crosses balls into the box and pray someone else puts it into the net?  If he fails to become that goalscorer for which you strive, then he can still settle for being a winger.

Similarly, if you have a player with high enough potential that he can learn to play the ball from the back,  why turn him into a defensive full back or defensive centre back who can only launch long ball after long ball?  

At the end of the day, high CA implies the player is to play for a big club.  
Being a big club implies control/attacking mentality is to be used in most matches.
Attack mentality being used in most matches means a necessity for players who can pass the ball accurately.
If so many high PA players are developed in roles that struggle to pass the ball, then how will big clubs win matches?  
Physical superiority and long balls, of course.

Share this post


Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.


  • Recently Browsing   0 members

    No registered users viewing this page.

×
×
  • Create New...