SI needs to fix the mid-season collapse

[noikeee]

I haven't read anything beyond a post here and there, I just want to say that in this version of the game it feels very easy to beat attacking teams (or at least teams that give you space to play in) and quite a bit hard to beat teams that park the bus. I've posted my struggles in the tactical forum, it's definitely something I'm doing tactically, but it's pretty amusing that once my team became world class and reached worldwide reputation, I stopped overachieving and started having pretty average results for the first time in my career and stayed at that level for years - and I remain absolutely convinced that the major thing that changed was that teams started playing more defensive against us.

[knap]

The same result will happen with a 4231 wide tactic.

It would be a bit odd if the AI just changed to the formation you are using, and the so called exploit tactics are similar to Chelsea tactics this year, so not a surprise popular.

Knock out competitions always have an element of luck, and the frustrations are just part and parcel of supporting a side in real life, although the AI tactics probably are one cause of the frustration.

From the German league I do not see the Looping example of winning then a steak of losing then winning again

 

[Svenc]

1 hour ago, knap said:

The same result will happen with a 4231 wide tactic.

It would be a bit odd if the AI just changed to the formation you are using, and the so called exploit tactics are similar to Chelsea tactics this year, so not a surprise popular.

Knock out competitions always have an element of luck, and the frustrations are just part and parcel of supporting a side in real life, although the AI tactics probably are one cause of the frustration.

From the German league I do not see the Looping example of winning then a steak of losing then winning again

 

You can narrow the pitch and overload centrally with 4-2-3-1 wide too. I personally barely know a single TFF type tactic that is built on any real world football tactic (in fact, TFF makes the point applying football logics was a waste of time), and I don't know a single side in football that positions like on the screenshot from a Mr L TTF tactic linked to on the last page [which spawned my comment about "going back five a couple dms" to draw that some "bust" just by  a simple switch in formation.]  As for the Bundesliga, Frankfurt barely losing in the first half of the season to barely win in the second, Bremen threatening to be relegated barely winning a thing, then winning 9 out 11 in spring. That was mainly posted as a point of reference to go by, as outside of FM, you are barely greeted with streaks being made this obvious (by having clearly color schemes attached to it all). I ideally want this to be as closely to football myself (which includes the engine and AI reactions, and agreed, AI mans just copying your tactics would be weird... unless you were the more successfully side perhaps), but you've got to look somewhere, and of this is all just going on gut feeling. Doubly interested as that will be tied to AI management some which could be improved (for some better competition too). If there's something hugely off, that were a decent place to start.

[knap]

Just to clarify the tests were not done with just TFF tactics. 

The test just shows how defensive AI tactics give them a better chance of picking up points, as reputation then decreases they revert to less effective tactics, creating a yo yo effect.

It raises the question that rather than copy user tactics, the AI should just use defensive tactics all the time.

 

[looping]

51 minutes ago, knap said:

Just to clarify the tests were not done with just TFF tactics. 

The test just shows how defensive AI tactics give them a better chance of picking up points, as reputation then decreases they revert to less effective tactics, creating a yo yo effect.

It raises the question that rather than copy user tactics, the AI should just use defensive tactics all the time.

 

Do you mean this looks like an unintentional rubberband caused by the ME playing defensive and narrow?

For picky users: other people don't experience that because they play better the game than me. I'm not finding an explanation for my bad overall results (which is only me making mistakes) but for my extreme winning/losing streaks.

[Ackter]

Can't believe people still think this is a real thing.

[sporadicsmiles]

53 minutes ago, looping said:

Do you mean this looks like an unintentional rubberband caused by the ME playing defensive and narrow?

For picky users: other people don't experience that because they play better the game than me. I'm not finding an explanation for my bad overall results (which is only me making mistakes) but for my extreme winning/losing streaks.

Other people do not experience this because they know how to snap losing streaks, which again is not anything to do with rubber banding. . I usually go WDWDWD when I am in decent form, since I often struggle to get results away from home. That is not because teams are defensive, but aggressive, and my defence in my current save is not very good when it is severely stressed. At home, when teams are more conservative (and my team is more confident), I usually can win. Not always, but often. When we are in great form, that is when we win comfortably away. When I am in poor form, I lose away, but damn I very very rarely lose at home to all teams of my level or lower. 

I can do this not because I am a tactical genius (I am not, I often have no idea what to do to change a match), but because I know how to manage my squad. I know what signs I am looking for with complacency, nerves, etc. I know how to use the media to try to mitigate that; take the pressure off on big occasions (I used to be bad at that too), light a fire under players when required. I can also guess roughly how a match will go based on these factors and how well we have been playing previously. I am always, always on the look out for complacency when I am winning. I am always carefully watching things when we are playing badly, because it is then that my tactical input can actually make a difference between losing and winning/drawing. 

The game is not really as simple as "pick a team, stick with it, pick a tactic, stick with it, and hit continue". I mean, it can be if that is how you want to play. You do not need to spend hours analysing every single part of your tactic, or micromanaging each individual training regime. You do have to pay attention to form (both yours and your opposition), moral (not just the "superb" or whatever, but how you expect your players to react in a game. Nervous, complacent, fired up, etc). I mean take your example. If I know a team is going to play defensive and narrow against me, do I just change nothing and shrug and go "meh, rubber banding here, I will lose". No. I will try to find a place where I can stress their defence. I will target a weak player. I will overload certain areas of the pitch (I can change this during the match as well), I will try to draw them out, I will try to be direct. I will select the players I think can do the job best (super slow defence, pick my pace 19 finishing 10 striker instead of pace 10 finishing 19 for example). There is no one thing, and maybe not all these things will naturally come to you (it took me maybe 4 years to get to the point I am now), but doing nothing is setting yourself up for exactly what you see.

I will give you one last example from a save of my own, which is contrary to rubber banding, and pro manager influence. I recently took over LA Galaxy on a save. They were languishing 20th in the MLS, they had just been dumped out of the cup by a non-MLS side, and had lost perhaps 6 or 7 in a row at this point. They were in awful form and had a super unbalanced squad. I came in, changed what we were doing tactically (which is normal), initially set up to be defensively sound so we could hopefully get a moral boosting clean sheet. The first game we ended up winning against a team in form just as bad as ours (a lucky first match). I was really supportive to the team. Congratulation and praise all round, both in team talks, press and individual conversation. Let's get some players happy. From that point, we lost only 3 matches in the league (out of 14, if my memory serves) and made it into the MLS Final, in the end (we lost that game, on penalties). The point of this example? If there was nothing you can do to stop an extreme losing streak, then my taking over as manager should have had little impact. It clearly did not, I clearly helped transform the season of this team, by carefully thinking about tactics, by managing the happiness of my players, and by luck (which you make, in the end). The point is not to sing my praises, but to point out that although not every loss can be chalked down to the manager, long term trends in performance almost certainly can (within reason, obviously if your team is really bad, you will struggle whatever you do).

[Svenc]

2 hours ago, knap said:

Just to clarify the tests were not done with just TFF tactics. 

The test just shows how defensive AI tactics give them a better chance of picking up points, as reputation then decreases they revert to less effective tactics, creating a yo yo effect.

It raises the question that rather than copy user tactics, the AI should just use defensive tactics all the time.

 

You should really forward this to SI for them to have a check, might be interesting, raised some concerns as far as AI going defensive as well, which is new in FM 2017. I don't have an overview but the ones I have seen and tried should never work that efficiently, in particular when teams would just drop deep and pack it (which defensive AI, typically, does). They compress the space incredibly, in both dimensions (going extremely narrow for zero width, plus encouraging every player forward for no depth), without the defending team even needing to contribute to it (making every player then sit on each others toe occupying the same space, which would make it really easy tto get a foot in anytime in real football). And as defensive teams quickly drop deep, they also can't get quickly behind their lines, which means against defensive teams every attack comes to this. Players pushing forward in masses also raises concerns about the defensive integrity of those. I don't have a complete overview, mainly prompted to them through others and some testing of my own. For instance, more recently in the German comm somebody argued why he should bother if you could simply "break the ME" with a TFF tactics (champions with Nürnberg, 3rd season). Of course you then go looking.

However, as it's a flawed ME rather than football, this would be worth investigating! I am personally of the opinion tactic testers / exerimental players should get involved in testing, as I think the main testers on this as well as the AI is playing this a bit differently! Seems it's still too easy to get into territory where neither the ME nor AI responds that well. =)

Re: Streaks, and that RL Bremen one, one part that can influence this is actually put into FM, it's edited to be. The team is without both it's by far best forwards, which also happen to be two of the three best players in the squad, until months into the season, as they all suffer long-term injuries. Which is also something watching out for. Whilst on FM injuries outside of isolated crysis overall are fairly toned down and have been for quite a while, how does the AI deal with preventing injuries from key players, keeping them fresh and fit for the crucial matches? How good is their possibly rotation, how well can they think in "low" vs "high" priority games etc. or is that an inherent edge to any human player who puts some more thought into this (myself I've never done overly much).

[HUNT3R]

13 hours ago, knap said:

Just to clarify the tests were not done with just TFF tactics. 

The test just shows how defensive AI tactics give them a better chance of picking up points, as reputation then decreases they revert to less effective tactics, creating a yo yo effect.

It raises the question that rather than copy user tactics, the AI should just use defensive tactics all the time.

 

In general these tactics are ridiculously attacking/aggressive, overwhelming the AI (hell, it'll probably overwhelm most users too) and of course the AI starts doing better when going more defensive because these tactics then over-commit.

A case could be made for the AI to realise it sooner in a match and also sooner into the season and it is always improving, so in time this will be a lot better.

[Svenc]

This isn't only about AI, @HUNT3R, it's also about how central spaces are defended as of FM 2017 in general. The acual "overcommitment" on occasion can also still have a flipside naturally. With this one, imagine the opposition playing aggressive wingers of some calibre and pace in tendency always sitting behind the wbs, waiting for a clearance/opportunity to counter [e.g. why did I suddenly lose to struggling Swansea 1-3 after having just thrashed United, Barcelona and Chelski, bad match engine, not recommended]. With some overhauls as to defending, also in terms of modelling the more physical side (truly collision model including truly blocked runs etc.), SI might experience another "FM 2013" again though, which was user after user logging on how the game were broken as nothing of prior editions would work that well anymore, forcing one of the tactical mods to write another "how to play FM" guides as a quick workaround.

However: It'd still be interesting how this is overall balanced longer term. Personally I've been advocating for better and more robust defending for years, likely a few others too. There were also improvements made as to how sides can break after an interception, as well as to punishing aggressive d-lines. In spite of all, it can mean that at a point, the balance may have tipped. Who considers for instance whether sides sitting deep would be too hard to break down? Or is it as "simple" a case of gauging player feedback, filtering it, and also looking at the results / tables the game would produce after 1000s of the famous soak tests involving AI managed teams [afaik, many ME builds never made public oft produce pretty interesting results, also in terms of goals scored overall]. Crucially in TFF's testing league, atop of any tactic put through it, it is always done with the worst team in the league. Which means it is, bluntly put, crap players possibly overcommitting, plus even without it, crap players coming against class defenders not gifting easy space, but denying it. The idea to somehow test specifically how "defensive AI" would perform over "attacking AI" is still pretty interesting, but that naturally is a pretty vital piece of context, regardless of the tactics filtered through it.

 

edit: Mid-season collapser taking over Feyenoord on FM 2015. :-P (Can Si somehow measure the average streaks of sides going without a win?)

Novara doing a "reverse looping™" when reaching the play-offs 2013.

[Mitja]

I wouldn't care about TTF tactics if I was at SI. if people want to play that way it's their decision. there were absoulutly no big issues with full backs in FM16 as long as you used FB A+S combination. and didn't employ ultry narrow tactics with wing backs (who stay wide) instead of full backs. people forget that AI starts adopting to your tactics and if you use wing backs whole league will use it also, especially in games against you. on my 16 save, all my fullbacks don't have more than 10 assists per season and I exploit the flanks, nor is any AI fullback leading the assist charts, also no team in the league is using wing backs. the only issue was that it was too easy to switch the ball to the flanks without sufficiant player quality or enough width. the only thing needed to fix this issue was to improve defensive positioning  and anticipation. but defending cries for improvements for years now anyway. from what I have gathered defending looks horrible on 17. we need to be carefull what we wish  

[knap]

Changing manager can completely change how a tem plays whilst keeping everything else the same.

edit: Mid-season collapser taking over Feyenoord on FM 2015. :-P (Can Si somehow measure the average streaks of sides going without a win?)

I am now wondering how much Reputation increase accounts for

2nd season poor form after improving team

Losing against relegation teams

Losing in Cup games against lower reputation teams.

Mid season dips in form (re-ranking)

Losing to a team of greyed out players

All these things should be reflected in game but is it a reflection of how reputation (+other factors) works rather than tactical

[Svenc]

17 hours ago, knap said:

All these things should be reflected in game but is it a reflection of how reputation (+other factors) works rather than tactical

Reputation and tactics (at least on an AI level) are linked. [I could be mistaken, but this works via the pre-match odds arriving in anybody's inbox prior to kick off, in which reputation is a factor, and by which AI can "see" whether it is approaching a match where it's thus favorite / underdog / level, to generalize]. Reputation I think also influences the player's perception of an opponent that ways, e.g. they may not be as pumped up against that small side they expect to beat than in that home match against the title holders, which can nbe shifted  a bit by man management.  So rep is a driving factor in all the dynamicism in the game world (naturally also transfers, etc.), and thus can be a contributing factor, you don't need to google long for "fm and randomness" to find something old but worth reading. (It was written by one of SI's former most valued testers gained some insights too. Bugger the game doesn't deal in displaying such percentages. 
 

Quote

Perhaps the shift in personnel dropped that 60% chance of scoring the first goal to 53% as the team were less gelled. Perhaps other teams started more defensively as your reputation had improved, dropping it to 45%. Add this to related form and morale boosts/drops, and you will have a poor season.

The general ranking in terms of importance of input from my stance (assuming the AI manager decisions were roughly level / you wouldn't override the ME/AI to the point that it hasn't much of fighting chance -- as that can go several ways, you could argue that many in-depth stuff in the tactics forum when understood lets you "exploit the AI" in a sense. It may be using the same tools, but such a level may be beyond it).

1: Player quality; typically, the better side wins more points [obviously, on more level competitions, more level results would be expected]
2: Tactics; a fitting approach may eek out a draw out of a loss and a win out of a draw and perhaps contribute to a giant killing
3: General man management; if a side is badly managed, it may lead to inconsistency all by itself [as could bad tactics / panic decisions] How SI all balance that is their secret, however to me every time I see a big side in-game underperforming with some consistency, my first instinct is, it's got to be how the AI manager lets that side play, and imo you oft go no hugely wrong on that even like me without being a tactics guru  -- same as any of the odd sides on FM 2015 initially that had hockey scorelines every week was connected to their tactics which I reported back then)

Part of the problem to some is conceptualizing it all and making a logical decision, but also identifying what is a tactical issue and what isn't and the game has been fairly ambiguous on that [but then so can be football]. E.g. was that pass misplaced/error committed because of the player getting isolated, because of his crapness, because he hasn't adapted/gelled,  because, duh some passes/tackles/headers go wrong, or simply BUG? The problem can already begin at when you start out, you may not know what can be much influenced, and what is more down to general decisions or flat out hard-coding. E.g. to use a fairly obvious example currently, wide-midfielder positioning thread, why can I push any combination of buttons I like and still can't get them to stay much narrower in defending?

[Sunstrikuuu]

30 minutes ago, Svenc said:

Part of the problem to some is conceptualizing it all and making a logical decision, but also identifying what is a tactical issue and what isn't and the game has been fairly ambiguous on that [but then so can be football]. E.g. was that pass misplaced/error committed because of the player getting isolated, because of his crapness, because he hasn't adapted/gelled,  because, duh some passes/tackles/headers go wrong, or simply BUG? The problem can already begin at when you start out, you may not know what can be much influenced, and what is more down to general decisions or flat out hard-coding. E.g. to use a fairly obvious example currently, wide-midfielder positioning thread, why can I push any combination of buttons I like and still can't get them to stay much narrower in defending?

Also, difficulty in translating the vision of football you have as a player into Football-Manager-ese.  Players often see what's happening, recognize a problem and want to change it to look a certain way but can't navigate the steps of converting an principle that's easy to articulate into the necessary instructions for the match engine.

[Svenc]

1 hour ago, Sunstrikuuu said:

Also, difficulty in translating the vision of football you have as a player into Football-Manager-ese.  Players often see what's happening, recognize a problem and want to change it to look a certain way but can't navigate the steps of converting an principle that's easy to articulate into the necessary instructions for the match engine.

Agreed! Found a fairly interesting one more recently, basically a demand for a "defend the flanks / avoid opposition crosses as best as possible" instruction that would do it all in a swoosh on a higher level, without the need to "tweak" the defensive shape and players on an individual level [which may be another argument for a fully-fledged out tactical assisant]. If you remember the sliders, they made that process even more painful (I had a beef with a few of those myself). To me the simplest way assistant and several more feedback can improve here is better feedback on the nature of shots, not the volumes, which is pointless. Because, eventually, those are the end product of everything. 75% of shots being long shots? Typically tactical, may not be a big issue when the opposition has them (depends on their nature and who has them). Trying to push for a goal rather than keeping things tight on your end but outside of working a few set pieces nothing really happens? Typically tactical [not only because set pieces have their own instructions anyway].

Every single opposition shot from within range, from open play and in space? Outside of bugs triggered in sequence, typically totally tactical issue, a clear sign to tighten up or concede soon (if it hasn't already happened). Speaking about shots in "space", I have a beef here that crosses connecting are if at all rarely considered as "good chances", as in-game too if they connect and the shot is on target, that's a far harder challenge for the keeper than that dreaded narrowed 1 on 1. That's assuming that the sides were roughly level, as outside of keeping the ball from him you can't stop Messi who according to the league stats averages 5, 6, 7 dribbles / runs alone per match cutting through your defense all the time, can you [with Messi being the substitute to any average super dribbler of your league level].

From there on you could decide whether the assistant would be able to spoon feed causes, which also implies AI in general were at a level where it would equally be able to spot causes and context better, improving the experience for all (e.g. too many long shots typically being the result of play being too compressed and/or player running out of options, e.g. boss, stop cramming every single player inside the box against this opponent packing it and/or open some additional passing options by making this wide player push some up).

[Sunstrikuuu]

1 hour ago, Svenc said:

Agreed! Found a fairly interesting one more recently, basically a demand for a "defend the flanks / avoid opposition crosses as best as possible" instruction that would do it all in a swoosh on a higher level, without the need to "tweak" the defensive shape and players on an individual level [which may be another argument for a fully-fledged out tactical assisant]. If you remember the sliders, they made that process even more painful (I had a beef with a few of those myself). To me the simplest way assistant and several more feedback can improve here is better feedback on the nature of shots, not the volumes, which is pointless. Because, eventually, those are the end product of everything. 75% of shots being long shots? Typically tactical, may not be a big issue when the opposition has them (depends on their nature and who has them). Trying to push for a goal rather than keeping things tight on your end but outside of working a few set pieces nothing really happens? Typically tactical [not only because set pieces have their own instructions anyway].

Every single opposition shot from within range, from open play and in space? Outside of bugs triggered in sequence, typically totally tactical issue, a clear sign to tighten up or concede soon (if it hasn't already happened). Speaking about shots in "space", I have a beef here that crosses connecting are if at all rarely considered as "good chances", as in-game too if they connect and the shot is on target, that's a far harder challenge for the keeper than that dreaded narrowed 1 on 1. That's assuming that the sides were roughly level, as outside of keeping the ball from him you can't stop Messi who according to the league stats averages 5, 6, 7 dribbles / runs alone per match cutting through your defense all the time, can you [with Messi being the substitute to any average super dribbler of your league level].

From there on you could decide whether the assistant would be able to spoon feed causes, which also implies AI in general were at a level where it would equally be able to spot causes and context better, improving the experience for all (e.g. too many long shots typically being the result of play being too compressed and/or player running out of options, e.g. boss, stop cramming every single player inside the box against this opponent packing it and/or open some additional passing options by making this wide player push some up).

I'd like to see a feature - I assume it'd have to be a full feature - where you could ask your assistant for advice in implementing a philosophy.  "I want us to cross for runners attacking from deep."  "I want us to maintain a tight defensive shape and go for the 0-0."  "I want us to maintain possession until we can create a channel to attack."  And your assistant would give you a starter pack and, crucially, explain why the shape, mentality, instructions and soforth were chosen.  Something like that would explain in-universe that sometimes Defensive is the right mentality for controlling possession, even though it doesn't suggest that anywhere.  The onus would still be on the user to identify the correct style of play for their team, or to alter their team to fit the style, and to tweak the system for in-match situations, but it would make it less likely that players would end up with systems that don't reflect the way they want to play the game.

As to shots, the annoying thing is that in my experience with this ME, long shots are really quite good.  They're easy to generate in from unmarked players and often lead to corners and parries or rebounds, which can create chances themselves.  They also lead to own goals.  Watching the AI's Manchester United over the last season I played was an interesting experience.  United set up with a defensive midfielder and a defend duty in central midfield.  The other CM and the two attacking wide players thrashed long shots away with abandon, often reaching the kind of numbers you'd see only from crazy attacking user tactics; in one match, they had 24 in a half and a touch less than 40 for the match.  Those long shots frequently went in since they were being hit by Pogba and Coutinho, and the ones that didn't led to corners; United scored a sixth of their goals in 2021/22 from corners, and they had the second best attack in the league. 

I agree with your beef regarding accurate crosses counting as chances.  I have another one.  The user needs feedback for final balls.   Often, a given tactic will actually be creating reasonably good opportunities which don't lead to shots because a dangerous pass is blocked, cleared or intercepted.  They don't show up as key passes, because they aren't completed; perhaps a key passes attempted stat would be useful (likewise key tackles and headers).  Additionally, some way to track dangerous runs off the ball would be quite useful as well in determining whether someone's not scoring because of lack of supply, failure to finish chances, lack of movement or simply bad luck. 

[waldywaldy]

17 hours ago, Svenc said:

Novara doing a "reverse looping™"

Works on so many levels

 

Exquisite

[Svenc]

13 hours ago, Sunstrikuuu said:

As to shots, the annoying thing is that in my experience with this ME, long shots are really quite good.  They're easy to generate in from unmarked players and often lead to corners and parries or rebounds, which can create chances themselves.  They also lead to own goals.  Watching the AI's Manchester United over the last season I played was an interesting experience. 

Long shots with a couple of genuine specialists you blew all your budget on could actually be consistently decent before. Still, maybe a connection to this thread (haven't really taken a look at them)? Naturally, you can't code any advice/feedback to possibly "imbalances" in the ME, on FM 2010-2012 you could totally turn every of your CBs into top scorers due to buggy corners (which in turn also could settle many a close match in your favor, giving you a huge edge over any AI). But long-term that's not very good advice. I agree about your feature suggestion, suggested similar. It's something that may happen in real football too, managers leaving some more nuanced details to a tactical assistant [Klinsmann/Löw], with the broader decisions and risk/reward decisions up to the manager to set in motion. You could still completely misjudge a match (your insistence to push for a thrashing after 2-0 completely backfiring after you have missed how they brought on a few key attacking players only in the 2nd half and went more risky, seen this with AI before, likely connected to them resting those initially due to thinking the starters would do), say. The fundamentals for this are already in the game to some degrees, like every AI manager reacting to scorelines, every AI having a bias for styles of attacking defending play and roles, or going into matches targeting draws, wins, and sometimes just not getting thrashed. To a degree it's also set in motion already in FM Touch's match plans for the assistant manager, as you can give him instructions to go for the stalemate, etc. It's just not advanced to the next stage yet -- the same you could say to the AI to an extent, who likewise oft struggles to replicate a really "possession based game" much purely on the "style of football front". However, anybody having a truly superior understanding on things on a micro level would still perform better long-term, which is the trade off you would have to deal with.

[Goosewinkle]

On 7/24/2017 at 09:10, looping said:

All I said is:

1. I see WWWWWWWWWWW LLLLLLLLLLLL streaks....

If you want to understand, nice. Otherwise, feel free to carry on if it makes your day better.

I feel like you may have misunderstood me, my good man.  I was trying to defend you since someone had claimed he never sees anyone complain about the match engine producing overly good results, even though you and I both were complaining about that in this very thread.  Anyway, wanted to clear that up!

[Chinook2000]

Is this right? ''You want SI to fix your game to make it easier for you to play''?

[upthetoon]

mid season collapse basically means you need to relook into your tactics. many people have studied and understood tactics and know how to prevent this. 

I think the ones complaining about this are players who just load tactics from others and have no understanding whatsoever.

[looping]

12 hours ago, Goosewinkle said:

I feel like you may have misunderstood me, my good man.  I was trying to defend you since someone had claimed he never sees anyone complain about the match engine producing overly good results, even though you and I both were complaining about that in this very thread.  Anyway, wanted to clear that up!

Completely misunderstood. I'm sorry for that.

[looping]

5 hours ago, Chinook2000 said:

Is this right? ''You want SI to fix your game to make it easier for you to play''?

No, nobody said that. Indeed, I'm more worried about my undeserved winning streaks. It's about the perception some of us have that the game is too streaky. Perhaps wrong perception but that's the discussion.

20 minutes ago, upthetoon said:

mid season collapse basically means you need to relook into your tactics. many people have studied and understood tactics and know how to prevent this. 

I think the ones complaining about this are players who just load tactics from others and have no understanding whatsoever.

False. Read before posting.

[lemeuresnew]

On 7/26/2017 at 21:38, Svenc said:

, you don't need to google long for "fm and randomness" to find something old but worth reading.

Love the second post. only improvement to his title winning side was a 23(?) year old Messi and a few regens not good enough for 1st team. we all hate bad transfer windows like that...

[Svenc]

6 hours ago, lemeuresnew said:

Love the second post. only improvement to his title winning side was a 23(?) year old Messi and a few regens not good enough for 1st team. we all hate bad transfer windows like that...

The focus of that thread is on the tactical side, but context is something that seems hard to judge in general (changes made to a squad, players available/used on both sides, able to gauge what separates sides in general, speaking about the MLS from the opening post, that is a league inhibiting a salary cap to begin with). As sometimes frustratingly flawed as FM is, it's a commercial video game, not a multi-billion dollar research project, as soon as somebody posts nothing but runs and results, you know he'll always have trouble (doubly so if he doesn't know how football seasons, real ones, ebb and flow). I bet not a big portion of the player base who has trouble here is the same kind of audience that believes the league tables at the end of a season were anywhere near to being a definite ranking of any team's true strengths. Personally? I'd put money that a few sides competing for European spots last BL season will be heavily engaged in a relegation fight the coming. That's not purely due to random chance, but common sense in a league where the difference between Europe and relegation can be as small as a couple additionally lucky wins, and where the comparably level competition produces streaks itself [real football betting odds barely fluctuate as wildly as the in-game ones, as bettors/bookies look for long-term key performance indicators, rather than random farts, or streaks of them, for that matter, all of which suggesing that outside of transfer windows the general "ability" of sides never changes hugely during a season, but the odds in the game serve a different purpose].

As a counter point to that argument thus that the game shouldn't allow anybody to go on a good streak if he weren't much cop: There's nothing posted in here suggesting anything truly horrid that should prevent those from being possible. [FM has made it harder to engage in basic mistakes as in real old iterations anyway, stuff that was guaranteed to produce underperformance,]. My personal opinion is it's this what separates the somewhat better from the average players on this, not that anybody can get into good runs (indeed, winning when all is going well is easy, it's when heads drop that you earn you worth imo), it's some added consistency and some less. Some then respond better to it, other less so, taking it on the chin, improving the squad and chances, and thus reducing randomness that-a-ways (you'd hope, or at least think that PSG currently trying to pick up Neymar is done in a bid to add consistency to their results, but then in the messed up world of real football, you'll never fully know who tries to benefit of what here).

[upthetoon]

On ‎1‎/‎8‎/‎2017 at 22:43, looping said:

No, nobody said that. Indeed, I'm more worried about my undeserved winning streaks. It's about the perception some of us have that the game is too streaky. Perhaps wrong perception but that's the discussion.

False. Read before posting.

the mid season collapse is certainly avoidable. look at my results right now.

When I feel like i'm going to go on a losing streak If I don't do anything, I change my tactics or drop players to do what's necessary. after a defeat at arsenal, I quickly knew QPR would be a difficult game and I won 1-0 with change of tactic.

if I did not do anything and just slam my spacebar which I think most do, i'd be moaning about 'mid-season collapse'.

https://ibb.co/cTER4v < sorry I do not know how to uploaded image.