Bad luck? Read this

[Dunno]

Thought I could make mods' work a bit easier

We, people, have a very good memory for bad things that happened to us. We easily forgot about good, lucky events though. I've never seen a thread entitled "whoa look at this, my adept survived an ulf!" and I've seen LOTS of threads about bad luck. So try doing this before posting an useless topic which would get locked after 5-6 hours: note down every unlucky AND lucky situation. Or do something simpler, remember one lucky battle and think about it every time you lose a fight.
Other thing is learning about probability. 70% chance of hitting is not equal to 100% chance of hitting. There's always this small 30% chance of missing. This means that if you load the game 100 times your unit will hit 70 times and miss 30. It's possible for a unit to miss 30 times in a row and then hit 70 times in a row. It's also possible (although [acronym=which doesn't mean impossible]unlikely[/acronym]) to miss every time. I think what I said is a shorter version of this. Read that too if you have some time.
Thanks

[Ferrous]

Sometimes you fail to note the lucky events because if you hit so many times in a row, you might kill the other unit before you can see the full extent of your "luck".

It's true though. Bad luck is louder than good luck. We take it for granted when things go our way, and raise a fuss when they don't.

[Kalajel]

I was going to suggest to sticky this, but the other thread you linked too is already stickied… and yet people still complain…

Oh well!

[Dave]

I think the problem is that as soon as you tell people about psychological artifacts around randomness and chance, they will agree that such artifacts exist but that they have a good and complete understanding of chance. Nobody is going to come here ready to launch into a tirade about how horribly broken Wesnoth's RNG is, then read about human perception of chance and say "oh, heh, I guess it's all my perception".

I think that the Dunning-Kruger effect is likely hugely relevant when it comes to recognizing random number patterns. The worse someone is at it, it's likely the better they think they are. Since they already suffer from the Dunning-Kruger effect when it comes to luck, they probably also suffer from it when it comes to Wesnoth playing ability. So we have someone who thinks they have a perfect understanding of luck and chance, and they also think they are a grandmaster-level Wesnoth player, and yet they continually lose to games.

No amount of evidence is going to convince such a person.

David

[Sangel]

I agree that the Dunning-Kruger effect is particularly obvious in many complaints about luck in Wesnoth. While I'm not aware of any peer-reviewed research to this effect, I believe that there is a link between the Dunning-Kruger effect and conspiracy theory. If you have cognitive biases that make you believe that you are better at tasks than you actually are - particularly if those biases are pervasive and encompass quite complex tasks and goals - then there's a necessity to account for the gap between your self-perceived competence and your inability to reach goals. Conspiracy theory is an ego-affirming way to explain that disparity.

You see this variant conspiracy theory quite often in video game forums - the idea that the developers have secret designs that they conceal from the public. While this might have some grain of truth in certain video games - an artifact of the fact that difficulty levels can be tweaked in a whole host of different ways, many of them not obvious to the user - it really makes no sense with open source software, as there's absolutely no way to hide such tinkering.

[Zachron]

You know they aren't going to read this... Then again, troll fighting seems to be one of our best past times as of late.

[Caphriel]

Eight experience apiece!

[Zachron]

Meh, just 4 most of the time.

[artisticdude]

I completely agree with Dave. This would be a good idea if humanity were less self-conceited, but everyone is naturally going to assume they're right and they rest of the world is wrong, and no amount of stickied threads is going to convince them otherwise (unfortunately). But hey, this may stave off just a few trolls that the mods would otherwise have to deal with.

[Hulavuta]

Every unit that advances for the player is a result of good luck right? I'd say units advance quite a lot.

[Gambit]

I think we should just take the Mom approach.
"Life has luck. Deal with it."

[Zachron]

You mean, admit that life is inherently unfair... why that would make them cry.

[oberon]

This would be a good idea if humanity were less self-conceited, but everyone is naturally going to assume they're right and they rest of the world is wrong,

Other people may make that mistake because they are conceited. In my case I AM always right!

More seriously, people who become convinced that the RNG is broken may either:

1) genuinely have had an extended run of bad luck
2) be bad at looking at the luck over the course of a game impartially.

Of course, that still does not explain why people who have had an extended run of good luck do not complain about the RNG

I find it hard to imagine how the RNG could be broken in a way that consistently favours one side.

[Gambit]

I came up with a new theory last night. It's a little crazy, but it explains everything.

I think that the developers aren't lying to us. I think the RNG is fine. What happened was that Crab made the AI so smart that, without him realizing, it figured out how to change the RNG post compiling! Think about it! It explains everything.

1. How the AI is so lucky.
2. Why it isn't mentioned in the manual.
3. Why it's so hard to prove (the AI erases all evidence!).
4. Why the developers don't seem to know about it.
5. Why you can't find it just by looking at the source code.
6. Why in all these years nobody has ever beaten HttT on hard difficulty without cheating.

It's all there man!

[Zarel]

I think that the developers aren't lying to us. I think the RNG is fine. What happened was that Crab made the AI so smart that, without him realizing, it figured out how to change the RNG post compiling! Think about it! It explains everything.

Not only that! He made the AI so smart that, it changes the RNG without affecting statistical analysis! So even if you calculate the probability of getting what you got, it seems reasonable, even though it isn't. We've uncovered a massive conspiracy!