Good idea to make the database more accurate

As many of you know, most of the (lesser-knowns) players in the database aren't fully researched. Often, besides some personal info, only the Current Ability and Potential Ability are defined, and maybe some attributes.

Upon starting a new game, the undefined attributes are generated by the game, using the CA and the position of a player as a guide (so a defender is likely to be a better tackler than dribbler).

While this is better than having random attributes, the lesser-known players still aren't very accurately portrayed in the game. I understand that it would be impossible to fully research every player, but I've come up with an idea to make the database much more realistic without too much hassle.

FM 2010 will use specific player roles, with multiple options for every position (for exampe: a midfielder can have the role of a deep-lying playmaker, a box-to-box midfielder or another role). What if the game not only uses the various options as tactical roles, but also as player types? When generating attributes for players, the game can use these types as a guide to distributing the attributes.

Researchers who can only assess a lesser-known player a little bit, might not be able to tell exactly if he should have a 9 or 10 for marking, a 7, 8 or 9 for crossing and so forth, but if he/she can tell if the CA should be 81, he can probably also tell if the player is for example a ball playing defender. The researcher can then add the player type to whatever tool they use to update the database.

These 'player types' could also be used for newgens, giving a larger variety of players in long term games. In addition, this sort of thing could also be used for more realistic personalities (possibly using the already existing personalities like 'spirited'). This will reduce the possibility of the most loyal and professional player in a squad in reality, being someone who is trying to leave and is causing conflicts all the time in the game (for example).

Sorry for the long read, I can be rather long-winded, especially in a foreign language. Thanks for you attention so far, and I look forward to reading your views on this idea.

I like this idea, I know that it is not essential but it would add a bit more realism to the game. I suppose it depends on how many players types there are if it means the players are better portrayed then that can only be a good thing.

Its also strange to see someones first post being a well thought good read instead of whens the patch out or will the Chelsea ban be in FM10.

Thanks for your quick reply AcidBurn, and for your compliment as well.

It's not essential, but I always play with my hometown team in a lower league, and the way it is now, a game like that just isn't as realistic as it could be. If I would use my FM tactic in real life, and use the best players at the club for that, then the starting team would be (sometimes very) different than the starting team I would use in the game.

I know that with my idea, the starting striker in my tactic might have a 15 for jumping instead of the 17 he may deserve in real life (for example), but with the way his attributes are distributed, he can at least play as a target man. With the way it is now, he may have the attributes of a fast striker, and would possibly be far less effective in my tactic.

Good idea. I guess it'll be too late for FM2010 research though.

To a certain extend this is where the PPM's (preferred player moves) are useful but as of yet there's no real definitive way to creating a 'player type' as you suggest.

BTW- whenz da demo out?????????

At first glance I thought this seemed like a good idea but after a little consideration I think to do this would be more trouble that its worth and also actually quite unrealistic and illogical.

For a manager to choose to play a player in a particular role, as is being implemented directly in the new edition of the game is realistic but for a player to be permanantly defined as being a particular type of player is not. The role a player plays, as we all know - and is a major part of the game, depends greatly upon the team he is playing in, in what league, what quality and style of play as well as the manager's own perception. A player's main role may change significantly through his career. When Ryan Giggs was much younger, he was pretty much an out and out winger but he certainly isn't that now despite being of signifant use to his team. Steven Gerrard on the face of it is a box-to-box midfielder but Liverpool had the best season for years last year playing him in a much more advanced role. I bring up these high profile names because surely for player descriptions like the ones you suggest to be added to the database they should be added to every single player. And implementing this would surely necessitate complicated ways which player roles adjust in-game over seasons. Also, if researchers knew a particular lower league player was a deep-lying playmaker for example he must know he has good passing and creativity and would put these attributes in accordingly. And good passing in BSN is not the same as good passing in the EPL...

Sorry this response isn't very well structured but there are so many reasons why the OP idea is in my opinion flawed I couldn't think how to put them in any order.

A decent post but not a good idea for me.

The OP is not saying it should be used for every player and it doesn't need to be. If a lower league player has a CA of 80 and it listed as a box to box midfielder then his attributes would be spread out evenly dependant on his ability. Obviously a deep lying playmaker in the BSP would have their attributes distributed to give them less for the key stats than Riquelme would due to a low CA.

You example of Giggs is flawed as Giggs moved to the middle as he has lost some of his pace, his ability to cross the ball is still as good as it was. Just because someone has the stats to play a certain role doesn't mean they can't play another role.

Using a high profile player as an example Henry would be best as a fast striker stats wise but can still perform as a cutting in AML.

The OP is not saying it should be used for every player and it doesn't need to be. If a lower league player has a CA of 80 and it listed as a box to box midfielder then his attributes would be spread out evenly dependant on his ability. Obviously a deep lying playmaker in the BSP would have their attributes distributed to give them less for the key stats than Riquelme would due to a low CA.

You example of Giggs is flawed as Giggs moved to the middle as he has lost some of his pace, his ability to cross the ball is still as good as it was. Just because someone has the stats to play a certain role doesn't mean they can't play another role.

Using a high profile player as an example Henry would be best as a fast striker stats wise but can still perform as a cutting in AML.

I just think it would be illogical to have this drop-down selection in the database which for so many players is used to select a role and for so many others is not. Can you imagine how this would be for the researchers?

"Do we know enough about this player to give him specific attributes?"

"No but I have seen that he is suited to being a deep-lying playmaker."

"How so?"

"Well he has good passing and creativity."

"Well why don't we just put his passing and creativity attributes in?"

"Because we don't know enough about him..."

Surely that would be a recurring problem if this system was implemented.

I use examples such as Giggs not to say that particular player types can't be played in other ways but to illustrate how the way players play changes - which would become valid for the players with player roles selected in the database. For example after so many seasons you will have a player who might not have played a particular role for years but this is still his defined role in the database. How would things like Genie Scout and FMRTE handle this? And Assistant Managers in game perhaps? And player descriptions in the personal screens which do now often relate to player roles, would they remain the same as in the original database or change with the change in attributes over game time? And how will player attributes change? If a player has a particular role in the database will this mean that certain attributes will remain high as the player goes past his peak while other drop unrealistically?

Just confusing and unnecessary in my opinion.

One thing I do not care about is how FMRTE or Genie scout would handle it, that does not matter one tiny bit. I guess we will just have to agree to disagree.

I don't think this is needed. What is needed is more researchers (apparently). ruudmeul, why don't you contact the head researcher from the relevant country and ask him if he needs some help researching your particular club?

Surely this already exists to a degree in the player descriptions? Examples include "Midfield playmaker", "Explosive striker", "Commanding centre-back" etc. I think if these were to be more formalised so that every player had one, it would be important that they were dynamic, and that the role be determined by player attributes and positions, rather than the other way round. The one exception being that the assignment of random stats at the start of the game could be based on a defined player role, as suggested in the original post.

Going back to the point about the roles being dynamic. As pointed out by Rejected Bid, some player's roles change significantly throughout their career, and it's important that this would also happen in FM, based on both the natural development/decline of player's stats, and also any specific training you might give the players. So the likes of a Ryan Giggs could have their role change from winger to midfield playmaker for example.

It would also be a nice touch if these changes cold be hinted at in coach/ass man reports. maybe something like "Due to a decrease in player X's stamina levels, they are not as effective a box-to-box midfielder as they once were and could be better used as a holding midfield player."

I would see this as an initial way to determine attributes for players who just have zeros at positions, and then progression would happen naturally, its just better than a total randomness of attributes.

I just agree with the OP that random stats could be generated by the player role. Although I guess that depends mostly on if the researchers know enough about the player. As Rejected Bid said I suppose if you know enough to suggest that someone is an advanced playermaker then you should know enough to give them their key stats.

Hi Rejected Bid,

Thanks for considering my idea and giving feedback. When a player isn't being researched thoroughly, it's not easy to give certain ratings for specific attributes. With attributes that aren't 'being used' at the moment the player is being assessed, that's even harder.

But when you see enough of a player to be able to say what his CA is, I would image that you have seen enough to have a general idea of what type of player it is (regardless of what role the manager tells him to have at that time, if that isn't his ideal role)

And I'm not saying that every player must have this in the database, but when there are (a lot of) undefined attributes, this system would be able to fill in the blanks, according to a more specific 'guideline'.

Also, I know that some players become a different type of player, because of getting older and slower for example, but what I am suggesting isn't something that is fixed, and in the game itself for that matter. It is merely a tool that will help with a more life-like distribution of attributes at the start of a game.

And lastly, since Genie Scout and FMRTE aren't official SI tools, there is no need for SI to consider that tools when making their game.

One thing I do not care about is how FMRTE or Genie scout would handle it, that does not matter one tiny bit. I guess we will just have to agree to disagree.

I'm not trying to have a go or anything, I'm just trying illustrate the flaws in the idea I've listed several potential problems and you've only responded to/rejected one. Admittedly FMRTE and Genie aren't the most important of these but I think they represent a point which would need to be addressed - that of consistency. Having this piece of information in the database for some players but not others is inconsistent. And again surely if a researcher knows enough to label a player role they should know enough to set an attribute instead.

Surely this already exists to a degree in the player descriptions? Examples include "Midfield playmaker", "Explosive striker", "Commanding centre-back" etc. I think if these were to be more formalised so that every player had one, it would be important that they were dynamic, and that the role be determined by player attributes and positions, rather than the other way round. The one exception being that the assignment of random stats at the start of the game could be based on a defined player role, as suggested in the original post.

Going back to the point about the roles being dynamic. As pointed out by Rejected Bid, some player's roles change significantly throughout their career, and it's important that this would also happen in FM, based on both the natural development/decline of player's stats, and also any specific training you might give the players. So the likes of a Ryan Giggs could have their role change from winger to midfield playmaker for example.

Thanks, John. I think this illustrated the two sides of this idea perfectly. Basically to implement the use of player roles in the database would require a lot more work with regards to making them dynamic and effective within the game. That's why I don't think this idea would be particularly useful.

Hi AB-forever,

SI already has a vast network of researchers, and still only a small percentage of all the attributes for all the players are defined in the database. It would be unrealistic to assume that even with a huge increase in the number of researchers, most of those undefined attributes would be defined.

In closing, I'll refer to my second post: the way it works now, when you play as a lower league team which you know well in real life, it's highly unlikely that you can use the tactic and formation you would in real life in the game. That's a pity for a lot of lower league FM players like me. In a quite simple way, my 'system' would overcome this problem for the most part.

Apologies to the OP and AcidBurn I think I lost track as to who started the thread and who had said what.

Just a slightly alternative idea: Would positional attributes like those currently in the database perhaps offer a better solution?

Instead of DM - 16, MR - 15, MC 20, AMR - 12

Perhaps: Holding Midfielder - 12, Deep-lying playmaker - 20, Wide Midfielder - 15 etc

or a combination of the two.

This does actually get me thinking about the flaws in that current system though. To repeat an example I think in previous editions Giggs was Natural AML whereas now he is Accomplished to reflect his change in position due to decreasing pace in recent years but in the game players' position ratings, as far as I know, don't naturally change in line with their attribute increases/decreases. Is this correct?

But when you see enough of a player to be able to say what his CA is, I would image that you have seen enough to have a general idea of what type of player it is (regardless of what role the manager tells him to have at that time, if that isn't his ideal role)

Firstly, I like the principal of the idea, but not the idea itself. I understand your worries and concerns about randomly generated attributes, but this doesn't cover the problems for me.

To come back to your quote, most players with loads of randomly generated attributes haven't been seen enough to have a CA set properly by a researcher. Their attributes are set based on their squad status, based on statistics. If a player is playing week in week out, he'll be given a CA within guidelines for a first teamer at that club. These guidelines are based on quality of the league and league position of the team.

I also think it's more likely that they'll be able to judge attributes than roles most of the time.

I think the reason I didn't answer many of your statements was that you are slowly turning me against the idea. It was when I thought about you saying that if the researcher knows the players role then they will know the main stats that the player has, no need to apologise lol.

You are correct their positions only change if you retrain them which is realistic. I think an even better example would be Alan Smith who would have been a natural striker but now he is a natural at DMC or MC. I guess it depends how you view it, was it more Alex Ferguson that changed Smith and Giggs or was it their own imput that changed their position?

Apologies to the OP and AcidBurn I think I lost track as to who started the thread and who had said what.

Just a slightly alternative idea: Would positional attributes like those currently in the database perhaps offer a better solution?

Instead of DM - 16, MR - 15, MC 20, AMR - 12

Perhaps: Holding Midfielder - 12, Deep-lying playmaker - 20, Wide Midfielder - 15 etc

or a combination of the two.

This does actually get me thinking about the flaws in that current system though. To repeat an example I think in previous editions Giggs was Natural AML whereas now he is Accomplished to reflect his change in position due to decreasing pace in recent years but in the game players' position ratings, as far as I know, don't naturally change in line with their attribute increases/decreases. Is this correct?

Yes, that is correct. But how suited a player is to a role is different to how... comfortable they are there. A role related more to a skill set, whereas a position is more of a mental thing first and a skill set later. I prefer the first idea tbh!

Yes, that is correct. But how suited a player is to a role is different to how... comfortable they are there. A role related more to a skill set, whereas a position is more of a mental thing first and a skill set later. I prefer the first idea tbh!

When you say positions are a mental thing do you mean as in mental attributes or kind of "experience/mentally happy" playing there? Sorry to wander off topic but I've just realised this is something I've been confused about for a while. Example:

Player A - Accomplished AM, Natural FC

Player B - Natural AM, Accomplished FC, but better FC attributes

Who would play better as an FC if generally similar standard?

Back on topic, my alternative idea was a bit naff but I personally still think the original idea is too lol