Is there really too much randomness in the game, or is it all a myth?

[PaulC]

I'd love to see a pkm of the 2-2-2-2-2 tactic in action btw.......

[CPS]

Setting up a 2-2-2-2-2 formation in the first place is farcical and I sincerely hope the players assigned to this formation had the nous and requisite mental attributes to disregard it (to an extent) and take up reasonably sensible defensive positions.

[glamdring]

I dont understand what you mean by "pre-coded influence". The match engine plays out second by second to create a result, of which it has no knowledge ( or interest in ) at the start. My conclusion is that regardless of the inevitable existence of match engine bugs, the end result is still influenced by the data sent into the match - which has to be the desired outcome surely?

The big issue though seems to me to be how much of that data sent into the match is defined by the user (in terms of player selections, tactics) and how much comes from hidden and possibly random data or from things like current form? e.g. If a team has been performing consistently well over a few games is there some unknown input factor that increases the chances of that run coming to an end the longer it lasts, via something hidden like players all becoming more complacent or reacting less to a team talk or that type of thing?

Obviously I don't want a 100% predictable game in which everything that happens you can pin down to something very specific which you can then fix and win every game, but there are times when I feel totally clueless as to what to change that might have an effect on my performances.

I know I'm appalling bad at getting my defence set up correctly (we scored > 100 goals and conceded ~90 in our last Championship season, incidentally no fewer than 4 teams scored over 100 goals, two more in the 90s and Derby finished with 97 points, but had to go through the playoffs - that kind of final table is fairly anomalous though from my experiences of other seasons!), but my tactics clearly work quite a lot of the time so they can't be just universally bad!

[Rupal]

Setting up a 2-2-2-2-2 formation in the first place is farcical and I sincerely hope the players assigned to this formation had the nous and requisite mental attributes to disregard it (to an extent) and take up reasonably sensible defensive positions.

But if it's just as effective as any other tactic in practice in FM, why is it 'farcical'?

It OUGHT to be farcical but we have been told that the results make it appear that this isn't the case.

If that's so, surely it shows that there is something wrong somewhere.

[makelele696]

i had an odd experience yesterday - i was 2-0 up away to QPR (im 6th they are 7th)when my computer crashed, i then played the game again went 2-0 after 10minutes so i quit because i want to win the game like i did earlier, i then send go on holiday and the assistan manager loses 4-3, so i load again and he loses 4-3 again with the goalscorers all in the same time, i then do the same and get thrashed 3-0, so i play the game myself and draw 0-0, and i just want to move on. BUT those 4-3 matches were well odd and it reminded me of this post

[andysafc]

Paul C, you take an overall view of the league tables at the end of a season and if they seem reasonable then, from your point of view, your simulation is a success. I completely understand why you take that view of things.

But I'm looking for a game Paul, not an attempt to replicate 'reality'. If I feel that my input is overwhelmed by factors over which I have no control, then, however accurately FM 'simulates a football world and puts you in that world' then, as a game it isn't as good for me as it should be.

I suppose it depends on what your customer base is looking for. Many people are clearly satisfied with your approach - but some others aren't. Currently, FM09 is the only management simulation available (I use that word advisedly). Those of us who would prefer a game will await the new CM and hope (rather against hope) that it will provide more fun even at the cost of less 'realism'. Or maybe some new competitor will emerge with a product which more closely matches what we are looking for.

Horses for courses.

This is where, IMO, FM has gone wrong. Trying to stay with realism, with realism Sunderland will not win the EPL, European Cup etc etc. But it was fantastic to do that in FM. Ok so it's not realism but it is what i wanted from the game.

Kick out the realism bring back the fun!

[CPS]

This is where, IMO, FM has gone wrong. Trying to stay with realism, with realism Sunderland will not win the EPL, European Cup etc etc. But it was fantastic to do that in FM. Ok so it's not realism but it is what i wanted from the game.

Kick out the realism bring back the fun!

Its perfectly feasible to win the Premier League on FM09 with Sunderland given a little bit of perserverance. Read some of the threads on the Challenges forum if you have time and you''ll find users who have won the Premier with the likes of FC United.

I remember the first Championship Manager game I played back on the Amiga and won the treble in the first season at Southampton - Colin Cramb was the best player in the world that season. I think I went back to Tracksuit Manager after that for a long time.

[Rupal]

I remember the first Championship Manager game I played back on the Amiga and won the treble in the first season at Southampton - Colin Cramb was the best player in the world that season. I think I went back to Tracksuit Manager after that for a long time.

I wish he was now - he doesn't even make the first team at East Stirlingshire at the moment!

[CPS]

I wish he was now - he doesn't even make the first team at East Stirlingshire at the moment!

Ha, I know - how the 'mighty' have fallen! I'm managing in the Scottish league and was tempted to sign him just for old times sake.

[Rupal]

Ha, I know - how the 'mighty' have fallen! I'm managing in the Scottish league and was tempted to sign him just for old times sake.

Noooooo! Don't do it, don't do it...........

[Hammer1000]

I dont understand what you mean by "pre-coded influence". The match engine plays out second by second to create a result, of which it has no knowledge ( or interest in ) at the start. My conclusion is that regardless of the inevitable existence of match engine bugs, the end result is still influenced by the data sent into the match - which has to be the desired outcome surely?

Obviously, as bugs get fixed I would expect similar results but with more realistic looking football along the way - which is always our prime objective with the match engine.

I fear this is where we lose each other, Rich(wwfan) understood where i was coming from regarding this issue, i wonder if it would be best to ask him to explain it?

I'll try a mock scenario in the meantime

Lets say for instance that despite leading the table, my team have lost their last 3 games away and in each match we had managed 20 shots at goal with 15 on target and created 10 CCC's, whilst the opposition had not had a single shot, but won 1-0 thanks to an O.G. in each game by one of my defenders.

I'm sure(well i hope) that you would agree that this would be unacceptable? if you agree, then we have made a start.

On the other hand, if you disagree and find the scenario perfectly plausible and acceptable, then i will most likely never get to enjoy this game ever again.

Of course the scenario i described is way over the top, but from what i am seeing on the pitch and the match stats that are produced to go with it, not a million miles away from the actual ME in FM09.

If you were to ask me where the problem lies? i would have to answer that i thought the game was still too easy and the best way to back up that theory is to admit that i am absolutely clueless when it comes to tactics and understanding the sliders, yet i go into most games knowing that i will most likely outplay the opposition both home and away, creating more efforts at goal as well as CCC's along the way.

You could of course counter argue that my lack of tactical nouse could be the reason i am not winning such games, but that just backs up my argument that the ME is not producing believable matches or stats and in the meantime i am still pretty much over performing with any side i choose to play as.

[Proteus]

Chopper is doing excellent work by taking the time to run these tests and present them to the FM community. Great stuff, chopper.

I retract what I said about the ME in various other posts on these forums. I was disparaging about it, citing its complexity and apparent randomness for my frustrations. I was wrong, and my inability to play the game properly, or run an experiment competently, caused my frustration with FM09.

In the other thread I mentioned an experiment I'd undertaken with Aston Villa where I assigned 20 to a number of attributes, such as consistency, big matches, determination, making a player's CA match their PA, etc, in order to ascertain how consistent they'd be over the course of an entire season. I wanted to ensure that the consistency of my squad was largely contingent on the consistency attribute of the players who made up that squad. I felt that if I remained inconsistent despite making every player in my squad possess 20 for consistency, then there was something seriously wrong with the ME.

And the results weren't good. My team were far from consistent despite using Matt's v4 tactic and only using players who had 20 for consistency, big matches, determination, maxed out CAs, and being healed before every game.

The reason the experiment generated erroneous results is because I'd used a heavily edited db to run the experiment.

Having read Chopper's responses to my posts in the other thread I decided to try the experiment again, but with the vanilla 9.2.0. database.

I used Wigan this time, using FMRTE to assign the same attributes to the players in my squad as I had with the Aston Villa players in my other experiment. I also used the same tactics: Matt's v4.

And this time Wigan were almost unstoppable. They dominated the EPL with their default squad, and they were easily the most consistent team in the EPL.

Sorry for wasting anyone's time in the other thread. I apologise for posting results from a flawed experiment.

Thanks

[Rupal]

Just as a matter of curiosity, Proteus, why does the edited or non edited status of the start database make a difference to the consistency ratings of a team where all the players are subsequently edited to 20 consistency?

[matt_132]

how can you say the match engine is too random when you cannot save and reload in real life? You may think that Man U will beat West Brom at home every time- maybe they will, maybe they could lose 10-0. You can never know

[Proteus]

Just as a matter of curiosity, Proteus, why does the edited or non edited status of the start database make a difference to the consistency ratings of a team where all the players are subsequently edited to 20 consistency?

I'm not sure, Rupal. The db I was using was heavily modified. It was the 6th iteration of the 9.2.0. db and I'd edited a massive amount of players, staff, and clubs to match my own perception of football. Some of the edited players also had twenty for consistency. The EPL contained the greatest number of edited players in my db.

It's a good question, Rupal. The difference in consistency of Wigan on the vanilla 9.2.0. db was staggering. With the modified attributes they were scarily effective in the EPL. The Villa team, on the other hand, were really relatively inconsistent.

I'd noticed that I could win and remain consistent in Serie A or the CC Championship, for example, but trying to achieve the same level of consistency in the EPL in my edited db just wasn't possible. Compared to the EPL both of the other leagues I mentioned above had a much smaller quantity of edited players. Hence the conclusion I came to re. the influence heavy editing had on my wee experiment with Villa.

Will continue to test things out to see what happens.

Cheers

[ashp]

I can sympathise with all the comments in this thread. I don't feel, despite knowing differently, that my tactics have any effect. Once I hit the second half of the season slump I just ineffectually slide the sliders around almost at random in the hope of magically effecting a change, because I can find no way to make anything actually change.

Sometimes I set myself real defensive and sometimes ultra attacking, watch the match and feel like I'm seeing the exact same play from my team. It's really a discouraging thing and regardless of the ME and how it creates the results, I feel an absolute disconnect from anything I do having any effect, outside of swapping the players.

[macca72]

Evening all.

If we forget about the words "luck" and "random" for a while and consider the word "chance" instead.

Let's take the last high profile game to be played, man u and Chelsea at Old Trafford. Let's assume that the first goal that Chelsea conceeded was indeed a tactical mishap, one that in a round about way Scolari admitted with his Man versus Zonal marking comments this week.

The numbers in brackets denote a potentially massive change in the match...

Following the first goal that was disallowed, Rooney ran fully 20 yards to scream at the assistant that he took the corner. From this position there was a chance that the referee could have taken Rooney to one side and booked him(1). From this point on there is a chance that the game could have been completely different, man u maybe don't score(2), it's 0-0 at half time(3) and Scolari's team talk is very different(4). Chelsea start the second half with a very different tactical approach, they're 0-0 away at man u and have recently been struggling, they chase the game less(5), there is a chance the result would have been very different.

One scenario in about 20 minutes of real time that throws up 5 potentially match changes situation. If you asked alex ferguson if he felt he was "lucky" that Rooney wasn't booked, he'd laugh. You ask him if there was a "chance" Rooney might have been booked, he might just entertain that.

FM's only real problem is that you lose this perspective. You play a ninety minute match in 5-10 mins, and because this is a simulation it has none of the opinion, conjecture, discussion, controversy and press coverage that you get in real life before and after a game.

Now, that's not to say the game can't be improved (tactics can be improved, but I won't go into that here), I personally think we need more teamtalk options, particularly at half time. In addition the text commentary is something that seems to have been missed recently with the dawn of 2d & 3d.

Don't get me wrong it's good, and you can see a lot ("x missed his interception and they made him pay!", "what a goal.....pure inspiration" etc) but I think we could see more ("x slipped", "x left y with far too much space", "the fans are furious that the ref didn't book x").

The player / manager attributes in the game (and this is a guess and my opinion only) seem to be a way of increasing / decreasing the probability (or chance) of any scenario playing out. It increases / decreases the possibility, but will, and never should, eradicate it. This is why a lot of people who play the game can still produce "unrealistic" scenarios (teams winning the premiership after 1-2 seasons, thus forcing out one of the top 4, which creates another scenario where Liverpool / man u / Arsenal go bankrupt).

We are all in control of the game, because every decision that is made has a knock on effect somewhere else, maybe not straight away, but most certainley over time. To suggest that "luck" plays a bigger part than a series of well thought out decisions, over a reasonable period of time, is simply wrong.

[Smac]

Some people have claimed that they can replay the same match and get wildly varying results, from losing 4-0 in one game to winning 4-0 in the next one.

I would take it as a starting hypothesis that some tactics and match-ups will produce more variation than others. If this isn't true, ever, then something is likely wrong. One axis of the tactical game is risk-taking, and a manager should be able to put forward a tactic that limits his risks as much as possible.

:

There is the nested question of "Does this risk-limiting usually require lots of tactical responses during a match, or can an overall tactic be played throughout a match without responding to the opposition tactics and match events?".

Often when people put forward their experience of wildly varying results it is countered with 'Its your tactics'. The question is though, can you actually demonstrate tactics that produce less variation, or not? I think it would go a long way towards giving a sense of control to FM players if they felt they could at least start a match with little risk taking and then 'up the ante' from there, depending on how things develop. We often see this approach taken in the real world when closely matched teams meet: The first XX minutes produce almost no chances as teams size each other up.

On the other hand, overly risk-averse tactics can sometimes be exploited and produce the very thing they were attempting to avoid. This can be luck, or it can be a sort of tactical counter-measure to prise open the faults in the risk-averse tactic (usually at some risk of course). It is often difficult to judge which is the case (luck or counter-tactics). However, with quality teams (high consistency, good leadership, etc..) it probably should be possible to develop consistently tight tactics or tactical responses that, if replayed, would have a high likelihood of producing similar results.[/tangent]

So from this first test it seems to me that there is no problem with the games random element being too prominent.

What is your null hypothesis, and what are it's criteria? To me the test doesn't show a case either way, for several reasons as mentioned above. What would 'too random' look like? What would 'too consistent' look like?

I often check out some of the match predictions in the EPL for the week ahead, and to look back at last week. Surprisingly, many people's predictions (on podcasts, etc..) are fairly accurate.

Lawro's predictions Actual results

Aston Villa 2-1 West Brom Aston Villa 2-1 West Brom

Arsenal 2-0 Bolton Arsenal 1-0 Bolton

Everton 2-0 Hull City Everton 2-0 Hull City

Fulham 2-1 Blackburn Fulham P-P Blackburn

Middlesbrough 2-2 Sunderland Middlesbrough 1-1 Sunderland

Newcastle 2-1 West Ham Newcastle 2-2 West Ham

Portsmouth 1-1 Man City Portsmouth P-P Man City

Stoke 0-2 Liverpool Stoke 0-0 Liverpool

Wigan 1-1 Tottenham Wigan 1-0 Tottenham

Man Utd 1-1 Chelsea Man Utd 3-0 Chelsea

Lawro got FOUR results right, with TWO perfect scores.

Perhaps this gives us something to aim for when looking critically at the match engine? You'll notice that he's off by more than one goal only twice (Man U v Chelsea, Liverpool v Stoke). I think this was a good week for Lawro, but not totally unusual. Therefore I'd start out with the idea that at least in the premiership, games are fairly predictable, probably more than your tests (non-prem, I know) would indicate.

.................................pt2..............................

For me this element of randomness is a non-issue. My irritations come from what appears to be too much randomness from game to game - e.g. when I play a similar strength of opponent with a similar tactic and against one we create loads, against the other we create next to nothing. Then also a recent example in which my Scunthorpe team usually win at home and lose away (fair enough, no problem with that), then suddenly we start losing home games and playing like Brazil away from home with seemingly nothing having changed and leaving me feeling like I'm just a passenger, left at the whim of hidden attributes or random elements that are influencing results far more than any changes I make in tactics.

That is probably just frustration on my part, but it does highlight, for me, the difference between a simulation and a game. Inexplicable things happen all the time in real life and thus in a simulation, but for a game I prefer a situation in which I feel it is down to me to work out what is wrong[/b], try to fix it and observe the results of that, etc. Maybe I just have a load of players with high hidden inconsistency attribute or too many of whom don't like big matches or some other strange goings on that I can't see. Too big a random element is very frustrating though (my bold)

I've felt similarly since FM07, from time to time. A series of unfortunate events should leave a manager feeling like he has something to blame! That would be realistic, no? And it would be more fun. If, after losing or drawing mysteriously for a spell the manager could come away 'knowing' something, it would be easier to take. "I blame my inconsistent captain". "There was poor refereeing again at Old Trafford". "I'm going to need new fullbacks next year". "My players are unmotivated. Time to threaten their places in the team". Something.

I do think it says something negative about the simulation and the game that even very good players often "feel like passengers". It isn't just the newbs. I've tried to change my expectations to distance myself from this bored-and-confused ugly experience, but it comes back even while I'm over-performing. Does this have anything to do with randomness in the game? Only on a macro level. On a real-time level it just feels like "being a passenger" or something quite out of control.

[DamianJMcGrath]

The only problem I'm reading in this thread is people are not unhappy about the perceived randomness in the game, they are unhappy about the lack of control they feel they have over their own team and results, and obviously feel that if they're not in control, the AI must be, and their control is random.

We don't feel like we're in control because our input into the game (through tactics and player instructions) don't result in what we expect. Maybe what we expect is wrong or our inputs are wrong, but none of us are actual football managers, and none of us have had professional training so all we can go on is our own instincts and our own thoughts. For example, if we watch the match, see that the opposition strikers are having a lot of the ball, an obvious reply would be to change some "closing down" settings. However, what we don't realise is that this has a knock on effect on another part of the game that causes us to lose anyway.

What we truly need is player interaction. For example, if I tell my central defender to close down always, but play with a deep defensive line, why cant he raise an objection, or say "Boss, I can't really do both. These two commands conflict with each other. Which one would you prefer?"

That way, we don't get caught out by subtle knock-on effects and we aren't left feeling our defeat was down to random reasons.

[Smac]

Well said DamianJMcGrath

Also, perhaps getting to know the limits of individuals would be great. Its one thing to note that your DC has a 7.10 rating and a Marking ability of 16, but perhaps he's particularly good at marking targetmen out of a game, but woeful with pacey wide players due to his personality (not pace). With such individuality in teams we could build to strengths and mitigate weaknesses. At least 'on paper'.

[macca72]

For example, if we watch the match, see that the opposition strikers are having a lot of the ball, an obvious reply would be to change some "closing down" settings. However, what we don't realise is that this has a knock on effect on another part of the game that causes us to lose anyway.

This is true, however this is the essence of the game. An obvious response would be to increase closing down, however a considered response might be....

If I increase closing down there's a good chance my DC's will step out of line and get punished, so maybe i'll...

1. Drop a MC into the DM position and man mark

2. Increase my defensive line to reduce the space

3. Show him onto his weaker foot

There's a chance that none of this will work, but the essence is don't go with your gut instincts, have a short think about any downsides to the decision.

I'm not saying your wrong, just offering a different perspective

[DamianJMcGrath]

I've no problem with it not working, but why not have in-game player interaction (or even better assistant interaction) to say "Boss, you told me to close him down but he's getting behind me too much - any instructions?"

That way, we have a chance to get feedback from the players or assistant on how the game is going and we can try something to stop it.

If it doesnt work, we have a reason for the defeat so it's not random.

[DamianJMcGrath]

This is true, however this is the essence of the game. An obvious response would be to increase closing down, however a considered response might be....

If I increase closing down there's a good chance my DC's will step out of line and get punished, so maybe i'll...

1. Drop a MC into the DM position and man mark

2. Increase my defensive line to reduce the space

3. Show him onto his weaker foot

There's a chance that none of this will work, but the essence is don't go with your gut instincts, have a short think about any downsides to the decision.

I'm not saying your wrong, just offering a different perspective

Your 3 points - how do you know how to do that? Maybe you watch a lot of football? Maybe you're fairly experienced at FM. Imagine if you weren't. You probably wouldnt even think of those 3 things would work together. You might just think of 1 of them, and then wonder why you're still losing the game, for example. You might not realise you need a combination of all 3 to have any meaningful effect.

If you got better assistant feedback or player feedback on how your changes are actually affecting the game, you could use their feedback and their professional opinions about what you should do, together with your own thoughts to put in place a decent response to the opposition and be fairly confident that the response is now taking into effect more things than your thoughts only may have conjured up.

At the only thing people want is a reason for their win, draw or loss. They don't want "I tried this, it didn't work because the AI is pre-programmed to win this game", or "nothing I did worked, so it's random if I won that or not".

[Wayne\'o]

im not sure if this aids to the topic etc, when testing out tactics that worked for my team. i reloaded the same match many times to see the outcome of using different tactics, i remember it beeing a game against leeds, this was a hard game to win, i think i settled for a 1-0 win in the end, but my best win was 4-0..

i got promoted via the play offs to the prem, where i was instantly relegated, becuase i held on to my star players, i made an instant return to the prem, breaking goals scored and points gained etc.. thing is i had been using the same tactic that i used for 5 season's with the odd change here and there...

it was clear as day that having a hand full of better quality players in my team gave me better results..

in all 5 season's i have played. the top 4 IRL have finished top 4 on my game every season bar 1, and they are always over 10 points clear of 5th place, 90% of the time the teams promoted get relegated..

I'm not sure how si have done it but they have certainly created the player CA to out play the lessor teams with lower CA players..

there are random eliments to any games, if two team IRl played 10 times in a row, im sure results would be the same based on team quality etc...

i have found tactics dont work if you dont have players good enough for that league no matter what you do.... people who say they have a great tactic have great players, i don't see any one winning the prem with Championshp players.

[macca72]

Your 3 points - how do you know how to do that? Maybe you watch a lot of football? Maybe you're fairly experienced at FM. Imagine if you weren't. You probably wouldnt even think of those 3 things would work together. You might just think of 1 of them, and then wonder why you're still losing the game, for example. You might not realise you need a combination of all 3 to have any meaningful effect.

If you got better assistant feedback or player feedback on how your changes are actually affecting the game, you could use their feedback and their professional opinions about what you should do, together with your own thoughts to put in place a decent response to the opposition and be fairly confident that the response is now taking into effect more things than your thoughts only may have conjured up.

At the only thing people want is a reason for their win, draw or loss. They don't want "I tried this, it didn't work because the AI is pre-programmed to win this game", or "nothing I did worked, so it's random if I won that or not".

Apologies, I wasn't inferring that a combination of all three was the answer, i was saying there are more things to try than the obvious initial reaction. Even if the player in question is new / inexperienced, then people will make the wrong decisions, and hopefully remember and learn, so that next time something similar happens, you apply your recently gained knowledge and perhaps trial something else.

The assistant feedback is ok at the moment, it could (and probably will) be much better in the future, and i'm sure SI welcome suggestions on how to develop.

The AI isn't pre-programmed to win the game, you can change the course of every match by making a decision / change. That's not to say you will make the right decision and won't make things worse!!! But again that's all about trial and error and being willing to learn how to deal with certain situations. People can now go back and watch a full game, if someone wants a reason why they lost a game, remember what decisions you made at what point in the game, then watch it back and see if you would have done anything different OR there was a piece of controversy, a missed interception, anything that is "common by chance" in football and maybe a less sinister reason than "i was beaten by random events" might present itsself.

[DamianJMcGrath]

The AI isn't pre-programmed to win the game, you can change the course of every match by making a decision / change. That's not to say you will make the right decision and won't make things worse!!! But again that's all about trial and error and being willing to learn how to deal with certain situations. People can now go back and watch a full game, if someone wants a reason why they lost a game, remember what decisions you made at what point in the game, then watch it back and see if you would have done anything different OR there was a piece of controversy, a missed interception, anything that is "common by chance" in football and maybe a less sinister reason than "i was beaten by random events" might present itsself.

But wouldn't it be good if you didn't have to do that, but instead had your players, who experienced it first hand, to come back into the dressing room and tell you, and for some of the more senior players to suggest ways we could avoid such an event happening in future games?

[macca72]

But wouldn't it be good if you didn't have to do that, but instead had your players, who experienced it first hand, to come back into the dressing room and tell you, and for some of the more senior players to suggest ways we could avoid such an event happening in future games?

I like the idea of senior player feedback...Would be a good addition to perhaps have some captains feedback as well as assistants. Not sure what type of role the Captain has in FM...

[Stephanie McMahon's Secret Lover]

I'd love to see a pkm of the 2-2-2-2-2 tactic in action btw.......

I'll try and post it over the weekend. Hopefully after my real-life footy management on Saturday.

I guess my only complaint or perhaps "query" would be a better word is how come the AI can play just as well even playing a ridiculous formation like 2-2-2-2-2 with all players wide.

[Squirmy Rooter 2.0]

But wouldn't it be good if you didn't have to do that, but instead had your players, who experienced it first hand, to come back into the dressing room and tell you, and for some of the more senior players to suggest ways we could avoid such an event happening in future games?

You sir, get the award for best post I have read so far today. (Although the day is currently only 24 minutes old...)

Throw in some post-match comments from pundits (or "expert analysers") for good measure as well.

As for this thread in general...

There will always be randomness and chaos in the game, and it's not something that SI can "tone down" or be controlled/reduced for us. Even with no random numbers at all, the fact that the human player can make changes to their side during a match, and that the AI is programmed to react appropriately means chaos theory comes into play regardless, thanks to our actions.

The reason that AI vs AI results/scorelines are generally realistic, as are the league tables at the end of the season, is that the AI on both sides are constantly fighting to restore "order" from "chaos". For example, if one team falls behind early on, they may try to push forward more in search of goals, and the opposition will try and protect that advantage. And more often than not, the better team will win.

We humans also need to react to what is happening in the game, whether it being freak "random" incidents, or simply being out-played etc., and make the necessary changes to keep the chaotic behaviour reigned-in and maintain order. In this respect, the game is no different that real-life.

What it boils down to I guess, is that it isn't that the random element is too strong, or chaos theory is to blame and should be "removed", it's that we need to be better at steering the play back on course when things do take an unexpected twist. And I do agree with the people that have posted in this thread and others, that we are arguably lacking the tools by which to do this - in terms of both analysing the match, and the means by which we communicate our tactics/instructions.

[wwfan]

Let's assume, for sake of argument, that the human user has an excellent starting tactic, has given a good team talk, is playing at home and favourite to win. The chance element might well be as follows:

Human team to score first 66%

AI team to score first 19%

No one to score 15%

In 7 out of 10 occasions, the human team will grab the opening goal. From that point on, the rest of the match will be related to the decisions of the human and AI manager. Perhaps, with no change in decisions, the human manager will have a 50% chance to score another during the first half, and the AI remains at 15%. We'll assume that no major decisions are made, so the human user is generally going to be leading 1-0 or 2-0 at half-time. During the second half, the AI will try to get back into the game. If the human user fails to make any changes, perhaps the AI's chance of scoring raises to 40%, as the human tactic is vulnerable to the counter. If the AI scores, this might unsettle the human team and increase the likelihood, with no change made on the human's side, of the AI grabbing an equaliser.

However, if the human team scores a third, perhaps that kills off the AI team's ambitions, leading to an uneventful remainder of the match, or perhaps a late 4th when the AI team is exhausted and demoralised. We have now generated realistic possibilities of the same match, with the same opening score, ending 1-0, 2-0, 3-0, 4-0, 2-1, 2-2 as long as the AI is making changes and the human isn't. However, we haven't taken into account the possibility of the AI scoring first. If they do, then it gives them an increased chance of holding on to a lead and perhaps increasing it, partly down the the inhernet advantage of already having a goal and partly downthe the user not changing things around. We can add 0-1 and 0-2 to the scenario. However, holding the lead isn't guaranteed, so perhaps we can add 1-1. We also have a 15% chance of the match ending 0-0.

Thus, the same match, depending on the opening goal, can realistically finish 1-0, 2-0, 3-0, 4-0, 2-1, 0-0, 1-1, 2-2, 0-1, 0-2 as long as the human user isn't making any changes. That gives us a rough 5-3-2 WDL record over 10 matches at the human users advantage, suggesting that the AI's changes have improved its chances of a result by about 15%. This spread means that the human manager isn't doing his job. With the better side, at home, he should reactively and proactively change things to accentuate his advantage, in an attempt to get the WDL sequence to 8-1-1, 8-2-0, or, if he is very good, 9-1-0 or 10-0-0.

I know my maths isn't exact, (and my assumptions are slightly iffy as well) and I'm not going to go into the calculations required to make it so, but the above neatly fits into Chopper's 10 match sequence. The issue we need to be focusing on is the management side of things, not the randomness of the opening goal. By doing nothing, the user is likley to give the AI an advantage of 15% over the assigned randomness/luck elements. By acting, or managing, the user should be able to increase his own chances of a good result by 20-30%.

[Rupal]

That's interesting, wwfan but there, is, of course, the other side of the coin, that the human manager's changes may actually make things worse not better. And as Macca72's and others' contributions suggest, you may need to make a very sophisticated decision indeed and when you combine this with the feeling that the sliders are unintuitive, a feeling of 'lack of control' can easily result. The AI manager has (maybe!) changed his/her tactics at half time, but we don't know how and it can be extraordinarily difficult to pick up what that change is simply by observing the behaviour of the players. In that respect, it's an oddity that the text will indicate changes in style/formation adopted by the AI team during a half but doesn't give the same information immediately after half time.

Ashp makes the point that he can set the sliders at any point between ultra defensive and very attacking and that the team appears to play in exactly the same way. Certainly that can be true - how is it logical for me to set my team to play counter-attacking football and find the players swarming around the opposition penalty area for a great deal of the time? Yet this can and does happen.

People's feelings that they are 'detached' or 'not in control' are a genuine issue. In this regard, perception is all. I believe that we feel this way because of real lack of clarity about what effect tactical changes actually have. What many of us seem to need is a clearer visual relationship between what we do (changing settings) as managers and the way in which our players act on the pitch. If this came about, the nagging feeling (misguided though it may be) that the position of the sliders really doesn't matter and that everything is predetermined would be far less likely to arise. For better or worse, we would have decided to do something, our decision would have observably changed the way in which our players behaved and we would feel that, even if we messed it up, we weren't at the mercy of the fates.

[wwfan]

Aye, but it is a different argument, relating to intuitiveness (or lack of) of the slider system and the ability for the manager to make correct decisions. I've lost count of the number of times a forum user has stated 'Ferguson never changes his tactics', suggesting their inability to read a professional football match extends far beyond FM's 2/3d representation. That immediately removes a significant chunk of the user base from our calculations, as their unwillingness to change approach in any circumstance will never lift them beyond the 5-3-2 split.

Arguably, FM is a learning simulation in which, over time, the human user will learn what changes to make during certain match scenarios and begin to implement them more and more successfully. Continuously making bad or no decisions rightly gets punished. However, the question has to be 'is it too hard to learn what these decisions should or shouldn't be?'. I'd argue that, in general, it isn't and, bar how mentality works, pretty much every other slider has a logical use based on how real life professional football is played and applying changes correctly will improve managerial and team performance. Some users will never learn, simply because they don't want to. However, most will and, in doing so, significantly reduce the perceived randomness of the game.

[Rupal]

I believe that your point about the human user learning to make the correct changes would be more in line with what, in fact, occurs if it was clearer from what one saw happening in the match what effect one's change actually had.

Let's take one of the examples which people find annoying.

One is involved with one of those matches where one's players are hitting the post, having wonder saves made, etc, etc and are unable to score. You end up with 68% possession and 10 shots to their every one and lose 1-0.

Now I have found that, in practice, it sometimes seems to help to nick the attacking slider back one or two notches. However

1) When that works, I really have no idea why.

2) There appears to be no obvious difference in the way in which my players or the opposition players behave on the pitch. The only thing which is different is that the ball goes in rather than hitting the bar.

3) It doesn't work all the time.

4) Sometimes, but I think more rarely, in fact, it can appear to pay to move the slider up instead.

So I am left wondering exactly what is going on here. Has, in fact, my changing of the slider had an effect on the result or not? If I had left things alone, would I have scored anyway? Has my gut feeling that moving it back helps any basis or is it purely coincidence? Obviously, I have no means of knowing this. There is no clear cause and effect here.

A very large number of variables are outside the player's control (this is where we come back to the question of the relative importance or otherwise of the human manager's input). This, combined with the aforementioned lack of intuitiveness, is what leads to the feeling that one's input does less than it ought to.

[Smac]

Arguably, FM is a learning simulation in which, over time, the human user will learn what changes to make during certain match scenarios and begin to implement them more and more successfully. Continuously making bad or no decisions rightly gets punished. However, the question has to be 'is it too hard to learn what these decisions should or shouldn't be?'. I'd argue that, in general, it isn't

Amount of FM players who 'tweak' during a game: probably 1/30.

Amount who push beyond the 5-3-2 to the 8-1-1?: probably 1/1000

Amount who 'learn what changes to make during certain match scenarios: 1/50

etc..

It doesn't matter whether or not it is too hard objectively. If its not actually done by better margins, its not right. I'd love it if FM was "a learning simulation in which, over time....." for many users. The fact (stretching the term) is, it is only thus for a tiny handful of users. The potential reward of the game isn't realized by most, and I'll continue to rant against this state of affairs until it is corrected. Because you are right: Such a game would be brilliant. One in which Joe User gradually gained insight and wisdom due to his experience, and as such gradually got better at the game.

Joe User: Downloads tactics, if hooked up to the net at all.

Joe User: Doesn't tweak.

Joe User: Has no idea why he wins or loses.

Joe User: Never really improves, except in non-tactical areas such as squad management.

Am I wrong?

[RT--]

Amount of FM players who 'tweak' during a game: probably 1/30.

Amount who push beyond the 5-3-2 to the 8-1-1?: probably 1/1000

Amount who 'learn what changes to make during certain match scenarios: 1/50

etc..

It doesn't matter whether or not it is too hard objectively. If its not actually done by better margins, its not right. I'd love it if FM was "a learning simulation in which, over time....." for many users. The fact (stretching the term) is, it is only thus for a tiny handful of users. The potential reward of the game isn't realized by most, and I'll continue to rant against this state of affairs until it is corrected. Because you are right: Such a game would be brilliant. One in which Joe User gradually gained insight and wisdom due to his experience, and as such gradually got better at the game.

Joe User: Downloads tactics, if hooked up to the net at all.

Joe User: Doesn't tweak.

Joe User: Has no idea why he wins or loses.

Joe User: Never really improves, except in non-tactical areas such as squad management.

Am I wrong?

Smac, I agree with some of what you're saying (that it's likely that most users don't play around with tactics too much) but again, this comes back to the tactics interface perhaps not being as intuitive and transparent as it could be. It needs to be a lot easier and less painful to use if most people are going to engage with it, for sure.

The argument that some people are putting forward is that they have no real control over the outcome of the match, which wwfan is doing a good job of arguing against, for my money.

Others are saying that they don't feel like they can see what effect certain changes are having -- again, I think this could be partly countered by a tactics interface which is much clearer about what instructions are intended to do, with some sort of visual element to avoid confusion.

I suppose the assistant feedback was meant to provide some sort of help with where tactics were going wrong/right, but I'm not convinced by it's utility myself - I still find watching full games far more helpful, in that regard.

[Smac]

Smac, I agree with some of what you're saying (that it's likely that most users don't play around with tactics too much) but again, this comes back to the tactics interface perhaps not being as intuitive and transparent as it could be. It needs to be a lot easier and less painful to use if most people are going to engage with it, for sure.

The argument that some people are putting forward is that they have no real control over the outcome of the match, which wwfan is doing a good job of arguing against, for my money.

Others are saying that they don't feel like they can see what effect certain changes are having -- again, I think this could be partly countered by a tactics interface which is much clearer about what instructions are intended to do, with some sort of visual element to avoid confusion.

I suppose the assistant feedback was meant to provide some sort of help with where tactics were going wrong/right, but I'm not convinced by it's utility myself - I still find watching full games far more helpful, in that regard.

OK, argument 1: "I can't see what effect certain changes are having".

Counter: "Your changes have effect, you just need to learn what effect" (its your tactics, lol).

Argument 2: The majority of people don't get any better at all at FM.

Counter: None.

So yes, its apples and oranges (or lemons if in the UK). But they are related. Chess isn't for everyone, for instance, but if you suggest that you need to be good at chess to be able to get better at enjoying Lego Building Blocks, we've raised the bar too high. Even Joe User should (if I were designing a game) be able to gradually increase his sense of control and success, without consulting a forum, without anything but his natural interest in the game. If he shows us he's not willing or able to do the work to get better, is it his problem or the game's?

Ultimately, it is the game's problem. Joe can play any game he wants. He'll eventually find a game that challenges him but rewards him, even if that means making some sacrifices (like playing Fallout instead of Footie). Wwfan and anyone else can blame Joe all they want for his failure. We can either change Joe, or the game. Which is easier to program?

But more to the point, if FM is a game, then isn't part of it's fundamental aim to be a good game? And isn't being a 'good game' shorthand for 'being fun and accessible to many people'? We can all go play 4 dimensional Go and enjoy our genius but I'm not going to tell anyone that 4-dimensional Go is a great game in general. Its a niche game for those nerds with the time and patience to learn and master it (under 4 dimensional tutelage of course).

Again, it doesn't matter if Joe can learn the game. Its Does he? No. Then the game isn't good enough. The 'randomness' argument touches upon some of the reason for this, as does it's counter-argument. FM, IMO, is just barely satisfying enough to keep the diehards going. It is slowly losing ground year after year by cluttering up with irrelevant features (harsh, but less words is more sometimes). When subjects like 'randomness of results' are brought up we have a chance to argue again the basic difficulties of the game being subjectively 'a good game'.

To continue your aside: I like the AssMan feedback actually. Probably my favorite feature this year along with 3-D.

[Rupal]

Honestly, I find the assistant's feedback of little help.

It's of little use him saying 'So and so is an accomplished crosser of the ball, he must be closed down or one of those crosses will connect' or what have you when you have already got so and so on tight marking, hard tackling, closing down and showing onto his weaker foot.....

[Daley]

So from this first test it seems to me that there is no problem with the games random element being too prominent. If anyone else has done similar tests, or would like to, then please post more results here so we can build up a bigger picture. And all opinions of course are welcome.

Firstly can I say that I think this is a really interesting experiment Chopper, I apologise if someone else has already pointed this out but I started reading this thread yesterday but haven't gone through what I missed but I have had a thought which I wanted to share:

Surely the fact that you are not changing tactics or anything else means that it is the AI who is dictating the result of the game? now I know that you are looking for randomness but as the AI is reacting and you are not then for example when you get 'lucky' and score early then the AI (if it was good enough) should change his tactics so that he then scores as you will not have reacted to his change in tactics?

I spose what I am saying is that because you are not changing your tactics then I don't think it's a random result as it is the AI's competance that is really dictating the result. If you played the game 20 times and reacted in exactly the same way to any occurrence that happened in the game then you could tell how random it was.

I am not sure that I have really explained my thoughts very well there but I hope that some of you get my meaning.

[bieritarier]

To add my two cents, I still do think that certain areas, especially the "mental" lapses are over emphasized in this game.

In addition to that, a user will easily get frustrated when his team constantly does stupid things that you wouldn't consider as being a tactical problem when watching the match.

Some examples:

- one-on-ones: Only using common sense, not experience with the game, I bet there aren't many people that link success of one-on-ones with tempo.

- Poor running of defenders: A user just expects a defender to judge the direction of the ball correctly _most_ of the time, especially at professional level. Yet, even world class defenders constantly run sideways or stop dead I understand that this is a problem with the ME representation, but tbh, if players just crawl/run sideways/stop dead many users will _feel_ that the AI just pushed the "I win" button, especially when suddenly their defenders seem to become superhuman, which they weren't until half time.

- I say it again, team talk effect seems over the top for me. It ma just be another representation issue, but as said a little further up, when the AI (or the user likewise) manages to fire his team up on HT, that doesn't just mean that the affected team plays better and more aggressive, it lso means that the other team becomes a bunch of headless chicken, booting the bal against each other, missing easy tackles or hold up the ball until they are closed down and tackled despite being on quick, direct passing and rarely run with ball.

- tactics aren't intuitive. An example (following on from the above): My players get tackled too often. I do not care if they give the ball away (well I do, but first of all I want to avoid that situation, no matter if it causes other problems or not). As I said, I put them on higher tempo, direct passing and remove run with ball. I also untick tight marking (just to make sure they aren't too close to their opponents when we win the ball). Yet, they don't even try to move the ball around quickly. Players still wait until they are tackled.

You now could say I didn't find a way to break the AIs tactic. Fair enough and I can accept that I'm not a tactical guru. However, I do watch a lot of football and I'm not only interested in superb flowing moves, I can have fun with some result-oriented football. I would consider myself as a person that is able to spot strenghts and weaknesses of players and teams. Yet, I rarely see a game where all of a sudden _a whole team_ forgets how to play football. I can accept that one, two or three players don't have their best day or that the other team punishes their weaknesses, but the game currently makes you whole team behave like idiots when it calculates that the AIs tactics are now better than yours.

That is exactly what makes me feel like a passenger. I do believe that tactics make a difference. I am however not able to spot what I need to improve. I (and I've read other threads that tell me I'm not alone) feel that a change that does not seem related to the problems you notice on the pitch can make the difference.

One thing that mad me angry for example was that my players neve closed down as much as the AI did. I started to randomly move one slider after the other around, and finally a combination of high closing down and high _creative freedom_ did the trick. Now common sense would tell me that creative freedom should mainly affect offensive moves. What I also don't understand is that you can read everywhere that low creative freedom means that your players obey to your tactical settings more, not less. So, if anything I would expect to see them close down even less (bearing in mind I had my fullbacks on very high closing down and relatively little cf before).

Of course there is more that makes the experience "random", and most of it has been mentioned before. One thing that I want to add is that realism is good, and many people (uncluding myself) want it. But you can't have pure realism on one aspect of the game where another, related aspect is clearly underdeveloped. That means, if you take the match engine and feed it with all effects that are there IRL, then you need to be able to influence the same things to the same extent as IRL as well. That in turn would lead into micro-management. I don't know if FM09 is a micro-management game. I usually keep it fairly simple, mainly because I am yet to see the difference between a team tactic and a tactic where I carefully setup individual player instructions or differences in player development when setting up my own training schedules (a completely different story..).

I compare it to a racing sim.. It'S great to setup a car with all the settings you'd go in a garage. What's not good is to find out that reducing the spoiler angle only increases your top speed if you reduce tire pressure of the rear left tire as well....

[nymanr]

Great post by bieritarier who sums up my feelings about the game better then i could have expressed it my self.

Just to add to that.

It's very hard to really pinpoint the issues with the game experience that make it feel so random. Like i said earlier i feel it's random and that makes me feel out of control. If that is because i don't understand the game engine then fair enough, but as i am willing to go the extra mile and learn the new engine through tests but get mixed results more often then not it leaves me frustrated. I'm no novice and i understand you have to react in game and make changes(have played over 500 season with the previous versions), but in this version i just can't see any "pattern" (what my tactical change actually has done).

Another big issue is the way my input on "moral" issues can have such huge effects on the outcome of games.

This is a huge "random" event that can turn any game/player/team compleatly uppside down.

Manager interaction, Press conferance and team talks are a part of real life but their effect on outcome in games is highly debatable, and i say that with over 20 years of experience as a athlete in team sports.

In reality the effect is extreamly small and usually effects mostly young/inexperienced players who might be playing their first games of their career at a senior level.

I understand that for them to have a "point" in the game they have to have some sort of effect. But as it is now it has to much effect.

If it was up to me i would rip it all out and leave the features away as i find them boring and tedious and as this is a game i want it to be fun.

This is just one aspect of the mantra "make the game as realistic as possible" that i feel has just made the game worse and less fun to paly.

I might sound overly critical but i find the game in it's current state okay/good, but it just isn't inspiring.

[macca72]

One is involved with one of those matches where one's players are hitting the post, having wonder saves made, etc, etc and are unable to score. You end up with 68% possession and 10 shots to their every one and lose 1-0.

The only comment I would add to this particular statement is linked to what wwfan stated above. Take the example where the user has a perfect tactical set-up, it's not uncommon (or unlikely) that you may dominate the game in the first 15-30 minutes, have 6-7 shots on target and create 2-3 CCC's, and yet not score, that's not unreasonable.

At this point you (or the AI) make make a decision to change things, which from that point on will fundementally change the way the game is played. The AI may switch to a more defensive game, which in turn could lead to you having another 5-6 shots, maybe 1-2 are on target and 1 CCC. Now at half time the stats would be massively weighted in your favour, quite rightly to show your dominance.

You make a half time team talk of whatever. At 0-0 at half time, it's effectively the start of the game again, despite the fact that the stats show your total domination.

From this point on the AI may change again, or be inspired by a teamtalk, and catch you on a counter attack, or begin to dominate you. These scenarios are more than reasonable and lead to the point i've borrowed !

The biggest thing i've learnt from playing this game is that a match needs to be broken down into smaller parts, say 15 minute "chunks" of play. The people who are the most sucessful are the people who can consistantly create chances, control a game and understand that changes and decisions need to be made consistantly.

Just looking at the end stats and saying..."I lost because of randomness or pre-programmed realism" isn't the way to view the game

BTW Rupal - I'm not commenting on the validity of your post, just borrowing a comment of yours to state a point.

[Rupal]

I agree that the game OUGHT to work so that you look at the match, find that your tactics need adjustment, take the appropriate action, rinse and repeat..... If it DID work that way I'd be happy enough. That's what I want to see in the game. The trouble is that it doesn't pan out that way in practice.

Sometimes what you do seems to have no effect whatsoever.

I've just played a match. Up until that match, I had not lost for 10 in a row, so my tactics clearly aren't THAT bad. Now I lost this one 1-0. I tried various changes of slider. The problem which I have with this is NOT that I lost but that the pattern of play which I saw on the pitch didn't discernably alter! I made considerable adjustments throughout that match to the attack mentality, tempo adjustment etc and none of them appeared to make any difference to what I was seeing on the pitch. The ratings of my strikers and midfielders remained stubbornly low no matter what I did.

Now I wouldn't have minded even if the changes I had made had made matters obviously worse. For one thing, that might have given me a clue to alter things the other way. But more importantly, I would have felt that my input was doing something, even if it was the wrong thing. The fact was that I couldn't see that they had any effect at all. Something should have changed for better or worse and it didn't.

That's when one feels out of control. I am assured by people on these forums that if I was better tactically I could have found something and I suppose that that must be true. But it didn't feel like that. It felt that I was going to lose that match all the time and that's when the suspicions arise that I lost it NOT because I was so tactically inept but because I was overperforming (top of the table when odds suggested that I should end up at the bottom).

Of course, my squad isn't that strong anyway (LLM, Scotland Div 3 - East Stirling, which is bad as they come) and I did have four first team regulars injured. But the players I had in weren't all that much worse (all my players are pretty useless), so I'm not sure it was that.

That's the sort of thing that makes you feel that you are powerless to change things. It's just a feeling and people keep saying it's got no basis, but it's something which needs to be looked at, I think. We need to see that what we are doing makes a difference, for good or ill and from time to time we don't.

[bieritarier]

"The only comment I would add to this particular statement is linked to what wwfan stated above. Take the example where the user has a perfect tactical set-up, it's not uncommon (or unlikely) that you may dominate the game in the first 15-30 minutes, have 6-7 shots on target and create 2-3 CCC's, and yet not score, that's not unreasonable."

Absolutely true and not unacceptable in any way, although We could argue about the frequency this occurs in-game. IRL, you rarely see teams, especially with proven stars up front, that miss easy chances match after match after match. It does occur, no doubt about that an hence has to be in the game, no matter how frustrating it may be, but it just seems over the top at times.

"At this point you (or the AI) make make a decision to change things, which from that point on will fundementally change the way the game is played. The AI may switch to a more defensive game, which in turn could lead to you having another 5-6 shots, maybe 1-2 are on target and 1 CCC. Now at half time the stats would be massively weighted in your favour, quite rightly to show your dominance."

Also accepted, although a team that hardly get forward and is bombarded with shots by its opponent IRL is already considered as sitting back and playing defensive. In an open battle (which you refer to here), they may be the weaker team, but would _try_ to get forward and create chances nonetheless, unless they lose the ball as soon as they play a pass forward. I understand your point though.

"You make a half time team talk of whatever. At 0-0 at half time, it's effectively the start of the game again, despite the fact that the stats show your total domination."

Thats where you lose me. The game does not start again, at least not all the time. If I controlled my opponent before HT and neither they nor I change anything major, I wouldn't expect to see a different game after HT, no matter what the HT team talk of any team was. My players may lose patience and shoot earlier, they may get the odd opportunity to go forward and take them, but they shouldn't start to control the game all of a sudden, or at least this should be a rare thing. Even if they do make major changes that should not mean that my team becomes a bunch of morons and that _is_ the case now.

"From this point on the AI may change again, or be inspired by a teamtalk, and catch you on a counter attack, or begin to dominate you. These scenarios are more than reasonable and lead to the point i've borrowed !"

No, unless I have the weaker team and my players profit from a non-focussed favorite I wouldn't epect them to start dominating me when my tatics still tell my players to close my opponent down, play quick and tear their defense apart. Just because they were fired up at HT should not mean that my players are not able to hit a ball straight after HT, nor should it mean that their defenders become faster or their midfielders magically string 25 passes together whilst my bunch just stands there with an open mouth watching them.

Yes, they may get instructed to be more adventurous and to take counter attack opportunities, but then they shouldn't be able to get forward once and score just because my defenders, that are perfectly in their positions because I told them to stay back (which is one thing that imo works quite well) think that a long ball goes down at the halfway line rather than the 18 yards line and move forward instead of properly intercepting the ball.

"The biggest thing i've learnt from playing this game is that a match needs to be broken down into smaller parts, say 15 minute "chunks" of play. The people who are the most sucessful are the people who can consistantly create chances, control a game and understand that changes and decisions need to be made consistantly."

Yes, changes are required, you need to react to what's going on on the pitch. The thing is, you see what's going on on the pitch so you take counter measures. So far, so good. However, more often then not it turns out that the things you tried to improve get worse or nothing changes at all. A good example is the defensive line. The opponent plays 2 through balls that could've led to a goal, so you instruct our team to defend deeper. However, they still let the striker run away after a through ball because they either push up too far, even further than they did with a high DL (???) or they just stop dead. Fact is that at the current state you either read the match engine correctly or not, but you're not watching a football game and react to what happens on the pitch but you react to how the ME displays certain shortfalls that in many cases don't seem to have a direct relation.

Conclusion: there seem to be too many knock on effects in the match engine/tactics.

"Just looking at the end stats and saying..."I lost because of randomness or pre-programmed realism" isn't the way to view the game"

No it is not, but the ME representation doesn't tell you either, or should I say, it doesn't tell you like a real football game would as the reason for a sucessful counter by your opponent with a pinpoint through pass and your defender stopping dead may be too much creative freedom for your wingers...

[ashp]

One thing I would like, and I've suggested before, is for the sliders to relate to each other a little more. If I pull back a player to be real defensive, it would be nice if the system would automatically tune other sliders for me. A better example might be an ability to set sliders per 'zone', rather than just team/player. I would like it if I pulled back my defensive to be ultra defensive, it would automatically pull my midfield back a little so the huge gap didn't exist.

Obviously you would be able to override the computer if you wanted, but a great deal of the confusion with sliders would alleviate if pulling back one could auto-adjust some of the others. I would actually support just ditching individual player sliders and instead allowing you to set tactics by zone. That may just be defense/midfield/attack, but you could layer additional areas on top such as 'flanks', 'middle'.

Also the time has come for 'meta-tactics', where we can define a few related styles and have the players auto-adjust to the situation. I'm pretty sure in real life players have an idea for what's expected when they have high possession, when they need to shut up shop, when they need to play catchup etc, and having to pause and load a new tactic constantly is silly.

Users don't tweak tactics constantly during the game, but you can beat they would make several tactics for different situations if the option was available.

[Squirmy Rooter 2.0]

Some good posts Smac. This section in particular rang true with me.

So yes, its apples and oranges (or lemons if in the UK). But they are related. Chess isn't for everyone, for instance, but if you suggest that you need to be good at chess to be able to get better at enjoying Lego Building Blocks, we've raised the bar too high. Even Joe User should (if I were designing a game) be able to gradually increase his sense of control and success, without consulting a forum, without anything but his natural interest in the game. If he shows us he's not willing or able to do the work to get better, is it his problem or the game's?

Ultimately, it is the game's problem. Joe can play any game he wants. He'll eventually find a game that challenges him but rewards him, even if that means making some sacrifices (like playing Fallout instead of Footie). Wwfan and anyone else can blame Joe all they want for his failure. We can either change Joe, or the game. Which is easier to program?

The tactical interface, the ambiguity in not knowing what it is that you are actually telling your players to do, the difficulty in assessing what differences your changes make, and the frustration at times of seeing the exact situations happening that you deliberately went to great lengths to prevent. In the end, the game just felt more like an arbitrary puzzle game than being in the dressing-room / dugout etc. interacting with real footballers. It felt too much like the slider positions I set were merely a way of seeding the random-number-generator.

It's funny. In a way, I felt like with the slider system I had far too much control over my players. But at the same time, I didn't feel like I was in control of how the match played out.

I still think FM is a great game, and there are many things that are brilliant about it. As a programmer myself, I think the Match Engine is a terrific piece of work. But as a gamer, I am not so fond of how I provide it with inputs and the amount of information I get back from it.

FM, IMO, is just barely satisfying enough to keep the diehards going.

I used to consider myself a diehard fan. I bought every version from CM2 to FM06. But since then I haven't played it (apart from the demos) largely because of reasons already stated. I still like to look around these forums though for discussions like this, and look forward to the day when FM will make me want to play it once more.

[Hammer1000]

I quit my Everton game despite being top of the table at the time after 12 games in season 4.

Despite previously having come 3rd, 2nd and then somehow 5th? it just felt that i was been "allowed" to win it this time(although as in previous seasons completely dominating matches) add to that that other teams were not strengthening their squads properly, making it just a bit tedious and pointless.

I then spent a whole morning trying to decide who to be next, but knowing what to expect, i could not force myself to go through the rigmarole of setting up a game, buying players, sorting through pre season etc, etc, so i did'nt bother.

As usual the urge struck to start again and the only team i could think to start with(based on not caring if i won or lost?) was Spurs.

Having played 13 EPL games, it is obvious that we are performing just as well as i did in both my Man City game(1st in season 1) and my Everton game(3rd in season 1) the only obvious major difference is my strikers goalscoring averages.

At City my best two strikers were averaging 1 in 5.5 and 1 in 7

With Everton it was 1 in 5 and 1 in 8(Torres at Liverpool averaged almost 1 in 2 this season)

With Spurs i brought in Rossi and Lavezzi on the cheap as i was worried about Spurs lack of firepower.

My best two strikers so far are Rossi 1 in 9 and Pavlyuchenko 1 in 13 with the other 3 strikers in the squad as follows.

Lavezzi 0 in 10 and the only player in the squad who's stats are dropping

Bent 0 in 7 despite his pace he looks clueless and his average rating is just over 6

Campbell 1 in 16

I'm using the same tactic i was using as City and Everton, which creates a lot of easy chances, especially for the left sided striker if pacey?

I can't help but wonder if this is the games way of measuring results to produce realistic results(not games) and League Tables?

Obviously Spurs start out as ungelled so it would explain the difference in stats between them and City/Everton, but the games produced are not of an ungelled squad, but just of strikers who will not put the ball in the back of the net with any consistency.

Defensively and midfiels wise, its pretty much identical, with missed interceptions and players standing still/moving away from the ball, being the main cause of any goals conceded.

[Oy]

I've made an entire thread about this a while ago, because I really thought that match results are as good as set when you press the 'start match' button and leave everything as it is during the game.

All results I got were always very very similar to the same game I played in a previous load, so I thought there were some stats we cannot see (like motivation for example) that seriously affect the players. Because those stats are the same every load (given you saved right before the match starts), the outcome is always pretty much identical. Of course, I manage a team that is favourites against all other teams right now, so it's pretty logical the result isn't going to swing between 4-0 and 0-4.

So, if a bunch of players claim they can get a great variation of results every time, then they either do not save right before the match so that those hidden stats still change, or there really isn't any connection between the match results and what I think are hidden stats.

[chopper99]

Firstly can I say that I think this is a really interesting experiment Chopper, I apologise if someone else has already pointed this out but I started reading this thread yesterday but haven't gone through what I missed but I have had a thought which I wanted to share:

Surely the fact that you are not changing tactics or anything else means that it is the AI who is dictating the result of the game? now I know that you are looking for randomness but as the AI is reacting and you are not then for example when you get 'lucky' and score early then the AI (if it was good enough) should change his tactics so that he then scores as you will not have reacted to his change in tactics?

I spose what I am saying is that because you are not changing your tactics then I don't think it's a random result as it is the AI's competance that is really dictating the result. If you played the game 20 times and reacted in exactly the same way to any occurrence that happened in the game then you could tell how random it was.

I am not sure that I have really explained my thoughts very well there but I hope that some of you get my meaning.

This was pointed out to me by someone earlier, so when I finally get round to doing my next test (hopefully tomorrow now) I'll be taking over both teams, and using the exact same tactics and avoiding team talks. opposition instructions etc. That way neither manager will be reacting to anything.

[wwfan]

There are certain assumptions that need to be unpacked here if we are to continue in a constructive manner.

1: It is inarguable that human users have an advantage over the AI in terms of picking up quality players. The AI's ability to build a squad is fundamentally lacking. Thus, in general, human users will quickly be able to build a squad that should be able to dominate a division.

2: Given that a large percentage of users can achieve 1, it stands to reason to argue that even average to poor tactics enable such a squad to do OK. People are complaining that their teams do less well then expected, rather than come dead last. Thus, good players do OK with bad tactics.

3: Certain users, such as Hammer1000, tend to discover tactics that out perform the AI. Given that these users always do better then expected with good tactics, we can equally argue that good tactics do well with OK players.

4: In accepting 2 and 3, we can make the argument that good managers will do well via discovering good players and designing/downloading good tactics.

5: However, it seems almost impossible to significantly overachieve in this manner. In order to win a title, users must be able to either dominate the division in terms of player quality (which will take a few seasons) to the extent that the extra ability overcomes tactical weakness, or outperform the AI tactically to the extent that the difference in player quality is overcome.

6: Assuming that people who have won titles in either manner are not lying/cheating, we can assume that some users outperform the AI tactically, others outperform it in terms of squad building, others do both, and still others do neither.

7: Despite the above, there is still significant frustration in the user base about the ambiguity of tactical instruction. This is either because the users don't understand the instructions, don't understand professional football, or that the ME is horribly flawed/impossible to read. Given that some users can read the AI, and have produced reams of evidence of the above, we can conclude that a certain percentage of users don't understand the slider system, the decisions you should make in a match, or both.

8: From the earlier posts, the idea of 'tweaking' sliders has raised its head. Given the excellent quality of writing and opinion of the people talking about this, I think I can make the conclusion that somewhere, somehow, the idea of minor tweaks making a significant difference in performance has become the forum norm, whereas, in fact, the manager needs to make significant changes in most matches, especially with a weaker team than the opposition. I'd argue that this perspective is SI's fault and related to the paucity of the instructions in the manual.

9: Changing this perspective and helping users to understand the validity and significance of their decisions is the 'must fix' element of future FMS. However, this needs to be done without spoonfeeding, or the game loses much of its appeal.

As an aisde, and with the gratest respect, it is pure conjecture to post 'odds' relating to the number of FM users who learn to make the correct in match decisions. TT&F has been accessed, in its various English language versions, over 100,000 times, and has been translated into Dutch, Danish, Chinese, Italian, Spanish, French, Russian, Vietnamese, Indonesian, Greek, Serbo-Croatian, to name but a few. If you add the number of people doing well, either by following different methodologies or deliberately ignoring or remaining unaware of the existence of TT&F to the equation, I'd argue that the odds suggested above are way out of synch. It may well be unfortunate that TT&F is needed for so many, and could well be argued that it highlights a massive game flaw, but it doesn't help in terms of the validity of this discussion. Simply put, in order to improve performance, a human manager needs to make a significant number of good decisions, in terms of a: tactical design, b: in-match decisions and c: squad development. Weakness in any area will handicap the team. My argument is that a is not intuitve enough/too difficult, b is often related to users failing to comprehend the basics of professional football and c is too easy. Get a wrong and matches appear random as your opening tactics aren't good enough. Get b wrong and you suffer the same fate. However, get them both right and everything falls into place.

NB: This is not an argument defending mistakes in the ME. We know it isn't perfect, so it is pointless going there. Rather it is one arguing that significant decisions are required to overperform, which flies in the face of the general assumption is that minor tweaks make all the difference.

[Oy]

Despite the grandiloquent language, you're completely right WW. Ever thought of a career as a communication manager by the way?