Well not since you edited no .
Must of misread it then , I didnt see the word exactly for “some” reason
People who talk D1 are actually targeted less often. This is because of the Hunter rule I mentioned. Even if it’s false, people think it’s true so they don’t target the D1 speakers.
You may be able to predict “people” as a whole, but you cannot predict (with at least a 50% similarity) a single person in particular.
For instance, I might target people who talk day 1 regardless of the hunter thing
I’m not saying your wrong. I’m saying that I can get an idea on how different classes will usually act. There are exceptions to every rule and my predictions are probably useless if FoL for example. That doesn’t however mean the the predictions are useless as they mean I can influence the game from D1.
I find it interesting , people stop doing Action A because everyone expects you to do Action A which makes you predictable . Then it becomes nobody does Action A because its too predictable . Then because everyone has stopped doing Action A its ok to do Action A , because nobody does action A because its too predictable , so they wont see it coming . That’s why I always do Action A , except when I don’t do Action A .
https://en.wikipedia.org/wiki/Statistical_significance
Random chance actually does even out for everyone over a large sample size because every individual is subject to the same variables. You sound like you live in a world of “only BAD things keep happening to ME!”
1 Like
BBC’s Sherlock has a fantastic episode illustrating the infamous WIFOM problems.
In the end, it’s all chance. But if somebody guesses correctly, they can claim it was because they were clever enough to “mindread” the other person. If they fail, they just say they were unlucky. It’s always chance.
Is popularity randomly reading a thread I have nothing to do with and find a comment about me?
3 Likes
This is probably incorrect, simply because:
- few support classes help them, because they don’t know if they are on their side
- investigators cannot visit them
- killers probably won’t kill them, because they might be on their side
do not NECESSARILY even out.
Most people will fall in the middle; 50% bad luck, 50% good luck. However, some people will be on the other sides, with better or worse luck
Rope I have neither the time nor the crayons to explain to you why you are so unequivocally wrong. You are literally arguing against basic statistical math just shush.
LOOK AT YOUR OWN GRAPH.
See the part off the hump? Someone is going to be at that area.
each point on the “hump” as you call it rope, does not represent a person, it represents an event.
say all events have a value, positive negative or zero, where the greater something is, the better it is for you (positive values are considered good) and the lower, the worse (negative values are bad) and if its zero, the event is neutral against you.
if ToL events follow a normal curve (i believe that is the correct term), that means there is and exactly equal amount of events that are good and bad, every bad event has exactly 1 good counterpart of equal significance, because the curve (or hump) is symmetrical.
so, as you play more and more games, more and more events effectively counteract each other, and more are insignificant enough that you can basically (but not fully) ignore them.
so if you play an arbitrarily large, but finite, number of games, the probability of you having significantly more bad events becomes less than .001%, enough someone could practically guarantee it didnt happen, and reasonably bet $100 on that
its not an actual proof, but i think it’ll work here
(for those wondering, the standard deviation would probably be something like 5)
If you consider it, however, even if it is a minority, some people will experience greater over all luck and others negative luck.
When determining the future, you apply the average, but if you take a month of data and look at it, you would see some people having greater luck.
if someone, somehow, plays an infinite number of games, and added up all the values of all the events in every game you play, the total will, 100%, be exactly 0
if you play 100000 games somehow, i can tell you with 99.9999% certainty that the total of all the events that happened to you is 0, and with even more certainty less than 1 away from 0 (or rather, that the absolute value of that values is less than 1)
1 Like
That’s absolutely correct, Rope.
If you had read the article instead of just looking at the picture, you would realize that the significance level of that particular two-tailed test is 0.05. The typical sample size for a 0.05 α is about 30, iirc. It’s been a while since I’ve taken stats, so somebody is free to correct me.
I’m assuming that you have played more than 30 games of ToL and that most players have played more than 30 games. They’ve been subjected to a large sample size and therefore have the allowance to reduce those levels of significance much lower than 0.05, and eventually you can reach levels of 0.00001. Once you do, that number becomes zero.