Strikers score too much? Too many penalties etc? Some analysis

[Mitja]

all the issues he describe above are existing in 10.1, there's no reason why should 10.2 be even worse. as I said before, the long shot bug is probably not as bad as most people on these forums think it is.

if anyone's interested to read some more analisys: http://community.sigames.com/showthread.php?t=168395

[DP]

But the question was what progress has been made in 10.2.

[Loversleaper]

what's wromg with how CA works?

A counter attack in FM is launched when a team gets the ball and there are less than X opposing players between the ball and their goal.

Ticking the counter attack box means that X is a higher value.

Once on counter attack, a team will play as if on highest mentality until the counter attack phase of play ends.

If you have examples where you dont feel this is working as it should, then let me know and I will be happy to look at a pkm example.

You see, the issue here is if this doesn't create an exploitation of the Attacking side of the gameplay and I think this is why Counter Attacking is ticked in the more Attacking Frameworks. I think it could be one of the reasons that Strikers Scoring Ratio is a little high, as it could be the effect of the 'x' number of opposing players between ball and goal causing them to go to the highest Mentality. Counter Attacking in Real Life is a different type of gameplay than represented in FM10 and looks more like an exploitive option than the intended ideology...

[Mitja]

But the question was what progress has been made in 10.2.

there are probably much more things that are getting addresed than those he mentioned in that post. from what I've been reading, I'd say that defensive behaviour is far up the priority of fixes for 10.2.

personaly I'd be more than happy if the goalscoring share was sorted out.

[son_of_sosidge]

It's always nice to see thorough, correct analysis of the game.

While some people get in a tizzy about how the game is "unrealistic", a few points are clear to me from this data...

• The game engine is tweaked to give a realistic EPL result (no surprise there)
  
• The high proportion of goals scored by strikers can be simply explained by the game engine not generating enough goals for wide midfielders and defenders
  

Once you accept that FM is just a game, then the way scores are generated is acceptable. You just have to adapt your team selections and tactics to suit.

Taking an example from real life, Cristiano Ronaldo's scoring record from the wing is freakishly good. So much so that if you tweaked the game to let any winger be the club's leading scorer, you would have a long list of complaints about "super-wingers"!

[Mitja]

You see, the issue here is if this doesn't create an exploitation of the Attacking side of the gameplay and I think this is why Counter Attacking is ticked in the more Attacking Frameworks. I think it could be one of the reasons that Strikers Scoring Ratio is a little high, as it could be the effect of the 'x' number of opposing players between ball and goal causing them to go to the highest Mentality. Counter Attacking in Real Life is a different type of gameplay than represented in FM10 and looks more like an exploitive option than the intended ideology...

I would say the striker are scoring way too much. I don't know maybe you're right, maybe that is one of the reasons why they score too much. imo one of the reasons why they score too many is because wingers (especially AMLs, AMRs) score way too few. if this was balanced and with couple of more odd goals from full backs, the Goalscoring Share would look much more realistic.

also I think that CA as a tactical feature works fine. I don't have problems with whether the CA is ticked in more attacking Strategies but I agree with you that CA as a gameplan, a strategy looks odd, I wrote about it in tactical forum.

[Super Striker]

Hah typical, in the edition where strikers are massively over scoring in everyones games, mine currently stands at 3 goals in 19 matches. FUUUUUUUUU-

But I tend to knock long shots down to rarely so maybe thats why?

[PaulC]

That's not good news at all. Strikers better on through balls? Even more one-on-ones you mean? Looks like I'll have to pass 10.2 completely then.

Did I say more one on ones? No. I said that they are being made a little more efficient, at least for the better players.

[PaulC]

The counter attacking thing is interesting. Can you provide any clear examples?

There are a few in this thread who we should have been involved in beta testing in an ideal world.

[PaulC]

I know you guys are doing a great job there and although I only had a few issues with the ME in 10.1 would you seriously consider releasing a new ME version with the issues you've described above?

I'm in no way belittling what has been tweaked in what is a very difficult coding job but surely it would be counter productive releasing a version that could be worse than 10.1???

That's how I'm interpreting your message Paul.

I trust our beta testers to raise that kind of flag if needed. I'm just trying to be upfront about how the work is progressing.

[Äktsjon Männ]

Did I say more one on ones? No. I said that they are being made a little more efficient, at least for the better players.

Higher chance conversion ratio then? That's better news. There have to be less through balls that find their target though. That ties in with the defending issues but we've been through that already I think. How come midfielder runs into the box are less efficient in this version than in FM 09? Isn't that the core of the problem (the goal distribution I mean)?

[PaulC]

Without wanting to go into too much detail I think the midfielder runs are ok from the centre but less effective from wide.

Its always a work in progress, but I'm doing all I can with the aim of each update being an improvement overall.

[Mitja]

I wonder why it is so hard to replicate wingers movement in real life, all other positions' movement is ok. I think I know what's wrong with it, but can't understand why are the wingers such a pain in the ass. and the problem goes way back, probably since introduction of 2D.

the problem is even bigger when you consider that best players in recent years are all wingers or played as wingers-wide midfielders at some point (Zidane, Figo, Giggs, Pires, Ronaldinho, Kaka, Messi, CR, Ribery, Iniesta). I'm not sure how it works when programing the ME, but I think the answer is to make them roam and have more freedom in their movement..

[bigdunk]

You've basically answered your own question there. The fact that the best players in these positions (I think AMC suffers as well here) are considered the best players in the world is an indication that playing these positions well requires the highest level of skill/ability/magic/whatever. It's no surprise then that replicating this requires the most effort in coding terms, and given the regular release dates, has been the slowest to develop.

[CPS]

I'm finding the opposite - down in the 10th tier. I've created my own league and in the bottom tier clubs have reputations of 50-10 with truly poor amateur players.

My team's figures for goals in 43 matches was 0.7 for, 0.7 against. Checking the rest of the division I noticed that the top team scored less that 1 per game and the bottom team conceded less than 1. All-in-all there were loads of 0-0 draws and one-goal games. Across tier 10 and I think 9 it was a similar picture, but higher up the tiers the figures seemed normal. It see,s to be the case that at that level however rubbish the defenders are, the forwards can't create or score chances.

Following on from this here are some stats from what I think is the worst playable league in FM10. Championship 2 -the third tier in Northern Ireland.

Northern Ireland Championship 2

Total goals per season

FM10 09-10 = 358 goals in 210 matches. Average goals per game: 1.705

IRL 08-09 = 413 goals in 132 matches. Average goals per game: 3.129 (the 3rd tier NI was called the Interim League last year as part of a re-structuring of the league pyramid - it had only 12 teams hence the fewer games)

Average number of goals per game in FM10 data analysed in Eng, Fra, Ger Ita, Spain & Scot top divisions = 2.835 (5,823 goals from 2,054 games).

Goalless draws vs. high scoring games (games with 5 goals or more)

FM10 09-10 = 33 0-0 draws, 3 games with 5 goals or more

IRL 08-09 = 4 0-0 draws, 24 games with 5 goals or more

Leading scorers

FM10 09-10 (28 game season) = Players scoring 15+ goals = 1 Total number of players with 10+ goals = 1

IRL 08-09 (22 game season) = Players scoring 15+ goals = 2 Total number of players with 10+ goals = 5

I know PaulC has said that the match engine hasn't really been designed with these leagues in mind and these stats certainly bear that out - only 3 games out of 210 produced a 5 goal match. Moving up a tier to Championship 1, I noticed that Limavady Utd scored 10 goals in 26 games and still didn't get relegated.

When the announcement of the FM10 editor arrived quite a number of people were excited by he prospect of playing in small countries or below Blue Square S/N in England - unfortunately that excitement seems misplaced to me as the match engine simply doesn't replicate amatuer leagues properly. In my view, two terrible teams pitted against each other should produce the same, and in many cases more goals than Barcelona vs Manchester Utd. Go watch a Sunday League match then try to tell me that poor defenders cancel out poor attackers - it doesn't happen. In FM it appears that the higher up you go the more likely it is that defence can't match attack and I feel that's wrong. Again, I sympathise with the creators of the ME as it must be a nightmare of a balancing act.

[PaulC]

I think what is apparent is that there arent enough goals in the ME when you go down to this kind of level.

Perhaps that indicates that the players in those leagues in FM terms simply dont have the technical attributes to make goals, or perhaps that the defensive side is overpowered at those levels. CPS if you PM me I will get you added to beta and we can work through some examples over the coming months, but I'll be straight with you and say this is unlikely to be changed much for 10.2 simply due to other issues taking priority.

[Mitja]

really good analisys again CPS, there's nothing more to add. it's quite usual for my local teams to play games where scores are more like 6-3 than 0-0.

[Bromley Raven]

Perhaps that indicates that the players in those leagues in FM terms simply dont have the technical attributes to make goals, or perhaps that the defensive side is overpowered at those levels.

My personal opinion is that it's to do with the attacker/defender dynamic. IRL a defender will contain a similar ability striker most of the time. To break past the defence takes one of:

Lucky defelection (rare)

A great piece of skill (occasional)

Defensive error (more often)

This means that a solid defence will give away very few chances even against a great strike force. Therefore the higher up the pyramid you go the less goals there are.

In FM10 defenders frequently let the strikers through. It's par for the course to position themselves badly to defend against through balls. This means that a striker will always get chances regardless of the quality of defence and it's the quality of the striker that most determines the amount of goals. Therefore the higher up the pyramid you go the more goals there are.

I think to rectify this defenders need to have a greater influence on the game.

[PaulC]

Ok, so you make defenders better at higher levels. We have around 2.7 goals per game in 10.1 at those levels which is fairly realistic, so where do you make up the inevitable loss of goals per game?

[Bromley Raven]

Ok, so you make defenders better at higher levels. We have around 2.7 goals per game in 10.1 at those levels which is fairly realistic, so where do you make up the inevitable loss of goals per game?

Watching football in real life, the spread of goals is different. There are far more goalmouth scrambles in real life, and I think this is where FM10 can make up the difference (also probably more goals from corners and crossed free kicks - can't remember ever seeing a goal from a wide free kick crossed in).

IRL you will often see blocked shots running free to a player in the box, or a tackle causing a loose ball. In FM10 a blocked shot will either bounce back out a long way, or deflect out for a corner.

Watch this from 4:50 to 5:00:

the second goal is a free kick which hits the wall, bounces loose in the box to the right midfielder, who hits it in. This is an example of what doesn't happen in FM10.

To summarise, there are very few scrambles and instances of scrappy play in FM10. Without these the majority of goals seem to be from through balls, meaning a higher than realistic proportion of chances going to strikers.

[Bromley Raven]

In fact - if you look at the clip in my post above, the first goal (2:25 - 2:35) was a pass that bounced off a defender and fell to another player. Both goals in the game from deflections causing the ball to fall to someone who then puts it in.

So what I think FM10 needs is a remodelling of the ball physics when:

1. The ball bounces off a player

2. A player gets his foot/head to the ball but can't control it.

[PaulC]

Interesting idea, cheers.

[watsonsclarets]

Paul i was wondering who test the game ie seasoned fm/cm players or do you get some people in who are shall we say not regular players because at first most of the points i agree with on here until i really got into the game and yes strikers do score a tad to much but the goals per game ratio are about correct

I think we need a broader band of testers not just seasoned players but also new players so people pick up different areas as seasoned players tend to be better at the game. Maybe you already do this but i have found as usual the more you play each version the better you get as expected but if you put testers who are not regular fm players maybe they can spot things that maybe testers who are good at the game will not easier i know said than done

[phnompenhandy]

Top marks CPS for picking up my concern with excellent analysis, and many thanks to PaulC for taking the concerns on board seriously. In fact this thread is absolute quality - proper mature analysis of ME problems free of negative rants, and a great example of SI listening to player concerns and responding both honestly and positively.

[pares101]

Campeonato Brasileiro

Total Goals Per Season

FM10 2009 = 811 goals Final Table

2008(irl) = 1,035

2007(irl) = 1,047

2006(irl) = 1,030

Leading scorers

FM10 2009 = Players scoring 25+ goals = 0 Total number of players with 15+ goals = 3

2008(irl) = Players scoring 25+ goals = 0 Total number of players with 15+ goals = 6

2007(irl) = Players scoring 25+ goals = 0 Total number of players with 15+ goals = 4

2006(irl) = Players scoring 25+ goals = 0 Total number of players with 15+ goals = 1

Well, here's a thing. Completely bucking the trend the Brazilian league has significantly less goals than the real life equivalent. pares101 was good enough to show me that the Campeonato regularly just breaks the 1,000 goal barrier - so I thought maybe this FM season with over 200 fewer goals was just an aberration. I was initially puzzled by this then noticed that not one team in the league uses a formation with wide midfielders or wingers. The three stock formations appear to be the now traditional Brazilian box midfield, a 3-5-2 using wing backs and a 4-1-3-2 with one DMC and three CM's.

The real-life stats show that the league is not low scoring (it consistently out scores the Premier League for example) - what I think it does show is that the AI simply doesn't have the variability or unpredictibility to make a variety of formations work. I would suggest that there almost certainly is a rigidity/uniformity to these formations. Having experimented a bit with 'wingless' formations I noticed that the midfielders got in each other's way quite often and showed little positional sense. Would be interesting to know if anyone has played in Brazil using an orhodox 4-4-2 with two wingers.

Doesn't prove everything, but if you want goals and prolific strikers you might struggle to manage it unless the have ML/MR's or AML/AMR's incorporated into your formation.

EDIT: Just noticed that in the player average ratings in Brazil the highest striker was 66th in the top 100 players with an average of 7.03. Only 3 strikers got an average rating of 7+ in contrast 49 defenders had an average of 7+.

Also, as an example of the difference between matches in full detail and no detail - Adriano scored 11 goals in 32 league games with an average of 6.99 (full detail), however he managed 10 goals in 10 cup games with an average of 7.66 (no detail).

Great Job you are correct on all the formations showed... So what youre saying is FM has trouble with the 4-4-2 formation without wingers ? and to score a high percentage you would use the 3-5-2 formation ? Oh and BTW in your league finals for this year ehehhe!! palmeiras was leading for 18 - 20 rounds , and now Flamengo is 1st on the top 4 and with Sport being on the bottom hehehe!! SPFC ( my team is in the correct postion except there is a 3 way tie for 2nd place or if Flamengo loses the championship is open game for 3 teams ( today and tomorrow are the most important days for the brasileirao !! )

[Helvete]

Great Job you are correct on all the formations showed... So what youre saying is FM has trouble with the 4-4-2 formation without wingers ? and to score a high percentage you would use the 3-5-2 formation ? Oh and BTW in your league finals for this year ehehhe!! palmeiras was leading for 18 - 20 rounds , and now Flamengo is 1st on the top 4 and with Sport being on the bottom hehehe!! SPFC ( my team is in the correct postion except there is a 3 way tie for 2nd place or if Flamengo loses the championship is open game for 3 teams ( today and tomorrow are the most important days for the brasileirao !! )

No, the problem is that the wingless tactics, such as 3-5-2, box 4-4-2 etc, don't score enough goals, but in the formations who use wingers, the wingers don't score enough goals but the strikers do score to many.

Hence, with ex a normal 4-4-2, the opposition defenders defend very well against the wingers, they prevent them from scoring goals so good that they forget about the strikers, while with no wingers, the defense just have the forwards to conentrate on.*

*= This is not a fact, just something that I thought about when typing the first answer.

[CPS]

No, the problem is that the wingless tactics, such as 3-5-2, box 4-4-2 etc, don't score enough goals, but in the formations who use wingers, the wingers don't score enough goals but the strikers do score to many.

Hence, with ex a normal 4-4-2, the opposition defenders defend very well against the wingers, they prevent them from scoring goals so good that they forget about the strikers, while with no wingers, the defense just have the forwards to conentrate on.*

*= This is not a fact, just something that I thought about when typing the first answer.

That's pretty accurate, an orthodox 4-4-2 will generally produce two very prolific strikers in a successful team but the wingers will rarely get into a scoring position. I'll guess that 4-4-2 is the most popular formation for users in FM10 so that would explain the goal ratios for strikers that are doing the rounds. However, I'll reiterate the point that it is possible to organise a formation that shares the goals around - I had a game a day ago where my LB scored twice and hit the woodwork twice in the same game. I've also got over 20 players getting on the scoresheet this season specifically trying to avoid producing a 'super' striker. If the AI showed a little more invention in its tactics/formation I think that might go some way to avoiding these 15+ strikers that many teams are producing.

[Mitja]

what did you do with tactics CPS?

[bieritarier]

Without wanting to go into too much detail I think the midfielder runs are ok from the centre but less effective from wide.

Its always a work in progress, but I'm doing all I can with the aim of each update being an improvement overall.

From my perspective, wingers too often stick to the byline when the ball is on the other side. I also can't get them to line up on the corner of the box for a run and quick shot from inside the 18 yard line (parallel to the penalty spot for example). A lot of wingers do exactly that, especially when offensive full backs or wing backs are covering the flank. This is especially disturbing when you play a winger you ask to cut inside. From all I've seen, cut inside makes players reluctant to cross (well, as I put Robben on the right flank I wouldn't want to see him cross with his right foot anyway), but the only times I can actually see them cut inside is with the ball at their feet, behind the penalty box and then they will mostly take a shot.

In short, I wingers (and wide midfielders) should support the attack in the box a lot more. This would porbably also solve the goal distribution problem.

Ok, so you make defenders better at higher levels. We have around 2.7 goals per game in 10.1 at those levels which is fairly realistic, so where do you make up the inevitable loss of goals per game?

The point is the way goals are scored now - mainly two options: Long shots, mostly by the central or attacking mids, occasionally by wingers asked to cut inside (Ribery is a beast in that department) or after through balls (either a long hoof from the middle or a diagonal pass from the wings).

What I rarely see is effective wing play (crossing). While crosses seem to connect, they hardly ever lead to goals, at least for me. And if they do, it is of course the striker scoring. That brings me back to the point above. Wingers and wide midfielders need to support in the box a lot more. Central mids could move into the box a little more as well, although not much.

I believe if this is done correctly you could reduce the number of long shots and through balls (talking about the long through balls here).

What this game is lacking is the deadly pass from 20-25 yards away from goal to a team mate making a deep run into the box for either a quick low cross from inside the penalty area or a quick shot.

Less one on ones would also mean that you could allow less long shots and more "mid range efforts" leading to more 2nd chances as more offensive players would be there to run after the "rebound" (currently you rarely see 2nd chance efforts, or at least I don't).

I think the potential solution is simple - get more players from all positions into the box depending on the situation, and especially when playing with a lone striker.

I appreciate this is a hard act of balancing, but by doing the above I feel you could fix winger efficiency, goal distribution and frustration - at least I can accept that my players miss out on an interception on two quick passes around the area. What I hate is when my defenders are constantly caught out by long hoofs over the top when you try to defend a lead and play deep .

Cheers

[bieritarier]

Should've read the below posts earlier. This ties in with what I said as well. More players in the box equals a higher chance of any offensive players hitting a 2nd chance effort.

And yes, this will also require some work on the ball physics on deflections, blocks and saves. Ball speed usually reduces enormously on deflections (despite minor deflections) and especially saves and blocks. In FM, the ball somtimes bounces back over the half-way line after a keeper saves a shot (or after a shot hitting the advertisement boards). A lot of the energy the ball has however goes into deformation energy (see here ) along with friction, and that does not seem to be taken into account...

Watching football in real life, the spread of goals is different. There are far more goalmouth scrambles in real life, and I think this is where FM10 can make up the difference (also probably more goals from corners and crossed free kicks - can't remember ever seeing a goal from a wide free kick crossed in).

IRL you will often see blocked shots running free to a player in the box, or a tackle causing a loose ball. In FM10 a blocked shot will either bounce back out a long way, or deflect out for a corner.

Watch this from 4:50 to 5:00:

the second goal is a free kick which hits the wall, bounces loose in the box to the right midfielder, who hits it in. This is an example of what doesn't happen in FM10.

To summarise, there are very few scrambles and instances of scrappy play in FM10. Without these the majority of goals seem to be from through balls, meaning a higher than realistic proportion of chances going to strikers.

In fact - if you look at the clip in my post above, the first goal (2:25 - 2:35) was a pass that bounced off a defender and fell to another player. Both goals in the game from deflections causing the ball to fall to someone who then puts it in.

So what I think FM10 needs is a remodelling of the ball physics when:

1. The ball bounces off a player

2. A player gets his foot/head to the ball but can't control it.

[Robioto]

I don't think this is enough of a problem to make a big deal about.

[Mitja]

I don't think this is enough of a problem to make a big deal about.

but it was enough of a problem that you posted it isn't?

[CPS]

what did you do with tactics CPS?

Mitja, was playing 4-2-3-1. The 3 behind the striker all had maximum creative freedom and were swapping position with each other. Full-backs were playing on their 'wrong' side, cutting in and crossing rarely.

[Robioto]

but it was enough of a problem that you posted it isn't?

That doesn't make sense. I looked at the thread and dismissed it as being a minor issue and something not to be too worked up about.

[Stampler]

Yes, a projected total is useful for 09-10 as, let's face it, that is the season we start with in the simulation.

German Bundesliga

this league has only 18 teams playing 34 matches each, in comparison to 20 teams with 38 matches in the two leagues above

Total goals per season

FM10 09-10 = 892 goals

2008-09 (irl) = 855

2007-08 (irl) = 856

(couldn't find accurate stats for seasons prior to this)

Goalless draws vs. high scoring games (games with 5 goals or more)

FM10 09-10 = 25 0-0 draws, 57 games with 5 goals or more

2008-09 (irl) = 14 0-0 draws, 50 games with 5 goals or more

2007-08 (irl) = 20 0-0 draws, 48 games with 5 goals or more

Penalties

FM10 09-10 = 78

2008-09 (irl) = 32

2007-08 (irl) = 24

Leading scorers

FM10 09-10 = Players scoring 25+ goals = 1 Total number of players with 15+ goals = 8

2008-09 (irl) = Players scoring 25+ goals = 2 Total number of players with 15+ goals = 7

2007-08 (irl) = Players scoring 25+ goals = 0 Total number of players with 15+ goals = 3

Goal trends in FM10 seem to suit the high scoring Bundesliga perfectly. Its remarkable to think that the German League produces almost exacly the same amount of goals as the French IRL despite the big gap in matches played. So far the FM goal trends are similar across the Leagues - could FM perhaps take these leagues cultural and historic differences into account somehow? Twice as many 0-0's in France (40) than Germany in 08-09 (14) in many more games is a major difference. Could be very tough to implement no doubt.

Still don't know what to make of these penalty figures...

it was 77 penalties in the 2008 season in german bundesliga.

http://t-online.sport-dienst.de/bundesliga/2009/statistik/

[Barkermush]

Well these are the penatlys in games involving Burnley in real life.

Stoke City Away : N/A
Man Utd Home : Penatly missed by Man Utd
Everton Home : Penatly missed by Everton
Chelsea Away : N/A
Liverpool Away : N/A
Sunderland Home : Penatly scored by Burnley
Spurs Away : Penatly scored by Spurs
Birmingham Home : N/A
B***kburn Away : N/A
Wigan Home : N/A
Hull City Home : Penatly scored by Burnley
Man City Away : Penatly scored by Burnley
Aston Villa Home : N/A
West Ham Away : 2 Penatlys scored by West Ham
Portsmoth Away : Penatly missed by Pompey 

That's 9 penatlys in 15 games. Don't know what that works out at? A penatly every other game, Or 1 pentatly every 3 games? No idea but i see less penatlys on my FM2010 game lol.

[PaulC]

You cant really judge any of these stats over less than 300 or so matches IMHO.

[Jhemo]

I think it's a big problem. Will it be fixed in 10.2.0?

[Loversleaper]

The counter attacking thing is interesting. Can you provide any clear examples?

The only thing I can do here is make a suggestion: I think that Counter Attacking should be triggered in conjuntion with Closing Down and Defensive Line (perferably 14 & under -Own Half/Normal- on slider) where Tempo is increased (with maybe added more Direct Passing to get the ball further up the pitch into space). Mentalities could be increased on all players that don't implement Defensive Mentalities from the get-go, for example, or what ever is viable, of course. This would still enable one to have an alternative Play Style if the Counter Attack has not been triggered...

Just a thought...

[PaulC]

I think it's a big problem. Will it be fixed in 10.2.0?

Strikers scoring ratios or penalties?

[Stampler]

both alas..

[DP]

Strikers scoring ratios or penalties?

So what one is fixed then?

I believe you guys have acknowledged the striker issue but it seems to me that slightly too many penalties are converted.

HOWEVER this shouldn't be tweaked so that too many are missed!!!

[PaulC]

Trying to get both as close to RL as possible, without upsetting the feel of the game.