Just because you're paranoid...

[dantonsdeath]

Okay, so I'm managing Derby County and in 2018 we're currently (somehow) top of the Premiership with a few games to go. Hit a slightly dodgy patch of form, unlikely to win it, but have got a lot of decent young players and it's not exactly a fluke.

So we're playing Chelsea in the League Cup final and we suffer one of those 'odd' losses that seem weirdly predestined. Score three times and each time they equalise within a couple of minutes. Then in the second half my defence starts behaving very oddly and they score three more.

I'm sure we've all had moments where we've assumed the AI is conspiring against us, and just moved on.

Normally I'd do that, too. But this time I thought I'd do an experiment. So I went back to my previous save just before the game. Played again, lost. Played again, lost. And so on for half a dozen or so times. Often taking a lead which proved impossible to defend for more than a few minutes.

So then I tried another experiment. Going on holiday for a day before the game and seeing who won, and repeating the process if I lost.

And here's the strange thing: so far I've done that 42 times and lost the game on every occasion.

Bear in mind I'm top of the league and they are in seventh. And not top of the league by cheating - in the actual game I always accept whatever result I get. In the pre-match 'betting' for the game, my team are (slight) favourites.

That's very odd, right?

Enjoy reading the forum, and all the best. Just needed to rant.

 

[WJ]

It's odd, but not impossible.

 

[Lippo255]

Two things spring to mind.

1. You said you have a lot of good youngsters so perhaps they are not able to cope with the pressure of a cup final just yet.

2. You also said that you have hit some dodgy form, so no doubt that will play its part as well.

When you combine the two its really not that surprising, there is always next year.

 

[MANUMAD]

The definition of stupidity somebody said is to keep doing the same thing but to expect different results.

Did you change anything those 42 times?

 

[Sherkey]

The definition of stupidity somebody said is to keep doing the same thing but to expect different results.

Did you change anything those 42 times?

Albert Einstein says hello =P

 

[MulleyMFC]

I think it was insanity, not stupidity?

If you're changing tactics/players, you'd probably expect to win once out of 42 times surely even if by fluke

 

[MANUMAD]

I think it was insanity, not stupidity?

It was insanity. It just fitted the thread ...

 

[cheeks]

i think i know what you are getting at here. With the number of times people mention about luck, players not being motivated, players not turning up and someone just having an amazing game you would expect to win at least 1 just out of an anomaly scenario. As everyone keeps saying on here tactics are not the only thing which determine how the match goes.

I actually did my own little experiment because of your thread and i have found the same result as you, i have drawn the same game every time, either 2-2 or 4-4 and all of them have involved a late goal. however i changed my tactics for the matches to see if tactics would impact, i can see why you created this post

 

[Dynamo_ZahnZee]

I can explain your entire experiment with two words: Confirmation Bias.

 

[WJ]

It's frightening how entire matches are talked about as if it is one singular thing. Matches are dynamic. If you take the lead, the AI will try and come back. If your tactics can't cope, they will get back into the game.

 

[dantonsdeath]

Yeah but here's the thing. 42 times is an awful lot of times for the same outcome every single time.

As an example, the odds of tossing a coin and getting heads 20 times in a row is one million to one.

The odds of tossing a coin and getting heads 30 times in a row is a billion to one.

So you'd think the odds of 42 games being won by the same team would be pretty remote, whatever the role of tactics, team selection etc etc.

It's really, really extraordinarily unlikely. Highly suspiciously so, in mathematical terms.

 

[GodCubed]

So you reloaded a match that you lost, changed nothing and then expected to win?

Definition of insanity indeed.

Also I swear to God if anyone mentions scripting they're getting infracted.

 

[dantonsdeath]

To answer the question of whether I did the same thing each time:

Most times just left it entirely to the assistant by going on holiday and not setting any preferences. The results were different each time (ie Chelsea won by different scores each time).

A few times near the end I set different formations, team selections etc and ordered the assistant to use these before going on holiday. Again, different scores each time (but always Chelsea won).

I understand entirely the point about the ME being reactive, tactics, form, etc etc etc.

All I'm saying is that the odds of losing the same match 42 times in a row would be incredibly small even if there were a huge mismatch between the teams. When the two teams are even roughly evenly-matched, we're talking ridiculously small probabilities. I'd love to think it was nothing more than confirmation bias, but there's not really a good way of explaining the same result in a two-horse race 42 times times in succession.

 

[dantonsdeath]

So you reloaded a match that you lost, changed nothing and then expected to win?

Definition of insanity indeed.

Also I swear to God if anyone mentions scripting they're getting infracted.

Well, I was on holiday. It was in the hands of my assistant. But my point really is that the game WAS different every time. Sometimes 4-0. Sometimes 2-1. Sometimes 3-2. Sometimes 2-0.

Once it was 7-0. Once - only once - there was a penalty shootout after a 5-5 draw.

The only thing that never happened is that my team - despite leading the Premier League - managed to win the tie. And that is statistically very unlikely.

 

[cheeks]

on a rather amusing note i just calculated the odds of it happening without even seeing you mention about the coin toss and odds business. lets just say its longer odds than the lottery

 

[dantonsdeath]

on a rather amusing note i just calculated the odds of it happening without even seeing you mention about the coin toss and odds business. lets just say its longer odds than the lottery

Ha.

Full disclosure - tried it again a while ago; won (on penalties) on the sixth attempt this time. Didn't do anything except go on holiday.

So it seems it was possible to win the game, just incredibly unlikely.

And safe in that knowledge, I'm going back to the original, morale-crushing defeat and trying to pick things up again.

Might need _just a bit_ of luck...

 

[jameshlfc]

on a rather amusing note i just calculated the odds of it happening without even seeing you mention about the coin toss and odds business. lets just say its longer odds than the lottery

How did you manage that when you don't know the odds of Chelsea winning the game?

 

[mersk]

How did you manage that when you don't know the odds of Chelsea winning the game?

They obviously regard the outcome of a football match the same as tossing a coin. Maybe they'd prefer it if, instead of the referee tossing a coin to determine who kicks off, he determines the outcome of the match? And all the players just waltz back to the wardrobe, hit the showers and listen to the managers post-match team talk, while the supporters get in each others faces yelling stuff like "it was all down to the refereeing!"

 

[Dynamo_ZahnZee]

You've drawn a false correlation here with your coin tossing metaphor.

A simplistic way of observing this would be:
A coin will be n(a)^(b); where 'n' = number of variables and 'a' = the factors affecting that variable. and 'b' will be the number of times you'll be repeating the event.

If flipping a coin 20 times and getting heads 20 times is 1,000,000:1

1(a)^20=1,000,000
a=1.00000299573

A football match will be 25(a)^1= (I've no idea what this number could possibly be since chaos theory is applicable as a football match is many dynamical systems that are highly sensitive to initial conditions: Ie, the butterfly effect)

My point is that the sheer number of variables exponentially distorts your metaphor.

---

The point is that you've not lost the same match 42 times in a row. You've lost 42 individual games of football played by the same players 42 time in a row. That is an incredibly important distinction!

To be fair, you're just comparing holidayed results here. To tell me you compared 42 sets of results like this and you've "spotted a pattern" is not enough. You have absolutely no idea how these matches played out or who took the upper hand, when or how. Viewing full time statistics means nothing when to comes to getting the feel of a match as that's where games are won and lost. It's possible that in one of these games, you would have read the game better than your assistant at a critical moment, changed something subtle and then destroyed them.

---

P.S. This is why we get annoyed when people tell us the game is fixed or scripted. If you could develop a mathematical system to prove that FM (a giant probability engine / random number generator) has a internal logistical system to it, you should be in a university pioneering the study of quantum physics or high-end mathematical equations.

 

[Death Ball]

As an example, the odds of tossing a coin and getting heads 20 times in a row is one million to one.

The chances a tactic and team combo meets a tactic and team combo that's superior or its counter will lose thirty times in a row is nine hundred ninety nine thousand nine hundred ninety nine in one million.

Tossing a coin is a random act.

Winnning a match is not a random act, but the meeting of two squads and two tactics. In FM, each tactic is working over a layer of random rolls, but they do push those chances to one side of the other. If in the meeting of team A versus team B, the meeting of tactics A vs tactics B work so the chances are the kind that favour team B, team B will win most of times.

To finish with a joke, using the phrase: «I fought Mike Tyson thirty times and lost every match. It can't be possible. If you toss a coin, that it falls tails thirty times in a row is one in a million» - Woody Allen