Tactics page flaws

The forwards are selected based on "Fast Striker" and "Target Striker". There are other strikers, like those who lack pace or height, but have exceptional finishing. Like an Adam Le Fondre. Do they get missed out in the top 5 formations page, because they are not a fast striker or a target man?

How is positional rating determined? It seems this is what decides who gets in the team in the top 5 formations. I've noticed that a player can have 20 in DC and 15 in CM in positions, yet his position rating is higher for CM.

What does advanced generation of team squads do?

Thanks for the great tool. If I had more money I would have donated more. In future versions, I will donate more.

Bump.

The rating is based on few skills and calculated using different weights for each skill (you can adjust it in the way you like it in the Player Rating options).

Advanced generation of team squads works longer but in some cases can suggest better squad than regular algorithm in Club Tactics screen.

And thanks for your donation!

2013-12-20 11:56#150286 Genie : The rating is based on few skills and calculated using different weights for each skill (you can adjust it in the way you like it in the Player Rating options).

Advanced generation of team squads works longer but in some cases can suggest better squad than regular algorithm in Club Tactics screen.

And thanks for your donation!

Grr I lost a very long post. I'll try and remember it

What is the difference between positional rating and general rating? Does current ability affect positional rating? I have players in my Rotherham side who can't get in the top 5 formation, despite much higher CA, because they have lower positional rating.

Shouldn't the player rating coefficients be firstly based around the FM 14 key attributes for a role ( with key attributes 50 or higher)? I assume the key attributes for a role are what the FM 14 programmers see as most useful for a position. For instance, your DM weighting has technique, anticipation, composure, and creativity of 40. Of those four, only concentration is in the key attributes for a DM, so shouldn't it get a higher rating then the others?

Should I sign players based on CA/PA or current/potential position rating? An example would be Connor Hunte. He has high PA but very low potential position rating. Plus, there are players with low CA but high position rating.

2013-12-20 15:58#150340 kief :

2013-12-20 11:56#150286 Genie : The rating is based on few skills and calculated using different weights for each skill (you can adjust it in the way you like it in the Player Rating options).

Advanced generation of team squads works longer but in some cases can suggest better squad than regular algorithm in Club Tactics screen.

And thanks for your donation!

Grr I lost a very long post. I'll try and remember it

What is the difference between positional rating and general rating? Does current ability affect positional rating? I have players in my Rotherham side who can't get in the top 5 formation, despite much higher CA, because they have lower positional rating.

Shouldn't the player rating coefficients be firstly based around the FM 14 key attributes for a role ( with key attributes 50 or higher)? I assume the key attributes for a role are what the FM 14 programmers see as most useful for a position. For instance, your DM, in the player rating options, has technique, anticipation, composure, concentration and creativity of 40. Of those five, only concentration is in the key attributes for a DM, so shouldn't it get a higher rating then the others?

Should I sign players based on CA/PA or current/potential position rating? An example would be connor hunte. He has high PA but very low best potential position rating. Plus, there are players with low CA but high position rating.

For future versions, I suggest adding more striker options then Target Striker or fast striker. There are good strikers who are neither fast or a target men. Like a excellent finisher, or a technical striker with good first touch and long shots.

I think you should be looking at the role rating of a player rather than CA and PA. Asfaik genie scout shows the % based on the highest role % for that position.

For example, you may have a player rated 70% as a target man, but 65% as a Advance Forward. Genie will always use the higher number.

So if you are choosing your best formation according to the tactics screen, you have to remember that it doesn't take into account whether it chooses your two strikers as target men, or poachers etc, so it may not work that well in practice.

Your best bet is to use the role rating screen of each individual to see how they play in each position, looking through your team should give you a good idea of what kind of formation you want to be going for.

[/color][color=#AD7FA8]

2013-12-21 00:12#150411 Dodgexander :
I think you should be looking at the role rating of a player rather than CA and PA. Asfaik genie scout shows the % based on the highest role % for that position.

For example, you may have a player rated 70% as a target man, but 65% as a Advance Forward. Genie will always use the higher number.

Your best bet is to use the role rating screen of each individual to see how they play in each position, looking through your team should give you a good idea of what kind of formation you want to be going for.

I think positional rating is better then role rating for judging whether to pick a player in your formation, and whether to sign a player. Asfaik Positional rating is not influenced by role rating, only the players attributes. Genie's formations are based on position rating, not role ratings. Also, there is no column for best role rating. So it would be very time consuming to go through each players role rating for each position.

Sorry, I got Positional rating mixed up with role rating.

But what I said still stands, you're best to go and see each players best "role" rating, then decide your formation on that.

My guess is the calculation for positional rating is based on a set calculation, perhaps unrelated to role rating.

But then, if role rating was taken into consideration, then it may pick one or two players in a position in the same position. For example, two central midfielders as playmakers and so on.

Not necessarily better than positional rating that way. Still comes down to just learning what players play best as and remembering that when it comes to choosing your formation and tactics.Its not difficult or a long process.

I set my formation, then look at the role ratings of each player in Genie, then I edit the player instructions for each player in that position. That way I know when I swap players out, or sub players, they are using my specific role AND special instructions.

Ignore PA and CA. They are not realistic as to what a player can reach, because they could just be a defender with world class striker stats and they would be rated as 50% soley because they are just unsuited to their position.

Also, don't trust PA, without first using Genie chance of development, or proffesionalism/determination.

I can't count how many players don't reach their potential of being world class because of this.

I feel current ability does influence how well a player performs. The hidden attribute consistency strongly indicates this. It is based on how often a player plays at his current ability. If so, would it be worth adding CA to the weighting system for position rating.

When I search for players for Rotherham, most of the players with the highest positional rating are very old. Is there a way of telling how much these players stat's will decline? Like a reverse of potential rating. Otherwise, I see no reason not to sign a 39 year old, who has good hidden attributes like consistency and prone to injury.

Shouldn't the position rating weighting give highest weighting to the FM 14 "key attributes for role" ( with key attributes 50 or higher)? I assume the "key attributes for role" are what the FM programmers have programmed to be most useful for a position.

You can edit the player rating rating coefficients as Eugene said above, but I think that's for the general rating (as you pointed out is shown in the best formation section). So those ratings are based on attributes you chose yourself. (Not sure if this is a Donation feature only btw)

Role rating I imagine is a set of pre defined stat weights based upon each role in a given position. So I think what you are asking for already exists.- Just not yet taken into consideration when it comes to calculating your clubs favoured formation.

Anyway, Current Ability is separate from Player Rating.

You should be looking at player ratings and THEN consistency. Current ability asfaik is just a number of all the players stats added up. You could have a central defender with 200 CA and 20 of those are spent in Long Shots and 20 in Long Throws. Meaning only 160 of those points are actually useful.

Its no good having a consistent performer with the wrong stats. This just won't work. A player who has the wrong stats in a certain position is more likely to play worse even if they have a high Current Ability and Consistency. Than they are if they have a high rating and consistency. In fact I think that's the reason Genie Scout bought in player ratings to begin with. I think I can remember using one of the older versions when only CA and PA were available.

There is no set way to tell how a players stats will decline, but generally there won't be too much in one season, so try to sign them if they are very old on just a years deal and then re-assess them next summer.

The player rating coefficients are for the position rating, which is the rating used in the formation section. General rating is what the position rating would be if the player had a 20 in that position. It is useful for deciding if you want to convert a player to a new position.

I've found out role rating is the same as football manager, so it is the most accurate. I have found position rating to be quite accurate though. A CM with a higher position rating mostly will have a higher role rating in the CM roles. I've been through every position testing. The least reliable is the forwards, followed by DC, WB and AML/R.

Ideally top 5 formations would be based on role ratings, but then you might end up with two target men, and two deep lying playmakers which wouldn't work. Position rating gives you a player who is rounded and accomplished at all roles of his position.

Yup, thats exactly what I meant!

I've spent many hours on FM15, using various editors, to try and find out more about some of the things brought up in this thread. My timing is dreadful but this might help genie in further refining his excellent formulas for position ratings. Here is what I found:

Fascinatingly, CA (current ability) isn't actually based on all the attributes added up. It's based on an algorithm that gives strongest weighting to all the "key attributes" for a role. It is discussed in this thread - http://community.sigames.com/showthread.php/381779-Current-ability-cost-of-of-attributes-position-breakdown. There are some attributes that have been given bizarrely high or low weighting for the players position, like heading for goalies. The ability to use the weak foot well is crucial in almost all positions. I think this more accurately reflects what SI think the attributes mean, which has always been loosely described.

The main problem with CA is that if a player has more then one natural position, like a DC/DM, the CA is split between the two positions. While Positional rating is based purely on how good that player is in one position, which is better for precise scouting.

I decided to create my own positional rating for genie scout, based on the algorithms used for CA. I used the methods from the quoted thread to work out the weighting cost for each attribute in FM15. I then compared my positional rating with Genies. Again, the findings were interesting...

My ratings closely matched CA, when a player is only natural in one position. For instance, Godin and John Terry have similar CA. Both are natural CB, who can't play in other positions. Yet, genie's positional rating has Terry at 82.41% and Godin 77.21%. When I change to my positional rating they have positional ratings of 72.4 and 72.78. Which is much closer. Another example is Garay and Carvalho. Again, both are only natural in one position. Garay has a much higher CA of 152 to Carvalho's 140. Yet, genie's positional rating has Carvalho higher with 74.6 and Garay at 70.96. When I changed to my positional ratings, it was again in line with CA - Garay had 70.04 and Carvalho 66.84.

I also noticed that if you hire top scouts and coaches, the players in your squad that they recommend in the squad depth report, tend to be the players with the highest CA, or my positional rating. The scout were less reliable. Even a top scout, with high Judging CA, was recommending good players above other players who were excellent for CA, or my positional rating.

Whether the key attributes weighted for the CA, or the positional ratings that Genie's has carefully refined based on real life footie, have more influence on player performance is a big question. I might send a tweet to Miles Jacobson, or someone at SI, to see how important the weighted attributes are for CA to how a player performs in match. Although, I don'd see why these attributes would be put as "key attributes", or weighted this way if they didn't impact the match engine.

Also, FM doesn't have a CA weighting for Target men, only one rating for CF. So I did the Target Man position rating based on role rating. This didn't work at all. I found that most attacking players had target man as the best positional rating. Even players like Luis Suarez! If you sorted Genie Scout by best positional rating, the list was dominated by Target Men. I think if you have an advanced tactic and know exactly what you want from a player, it may be best to pick by role rating, included in the G edition. To allay this issue with target men, i set the weighting of Target Men to 0 on genie scout and If I want a target man, I'll bump it back up to 100, or just search by role rating in the G edition.

Hope this doesn't come across as criticism, Genie. I think your scout tool is amazing. I wouldn't find the time to play FM without it and appreciate the long hours that must go into programming the scout each year. I was just sharing my findings. I'm not sure whether to use my positional rating or yours, which is based on real footie and has been refined over many versions.

I can share my positional rating file for Genie Scout, or the document with a list of the attribute coefficients, if anyone is interested in having a tinker.