Board Game Arena

Gulchen wrote: ↑09 November 2023, 05:55 on copy-pastes of ​ the analogue of https://boardgamearena.com/gamereview?table=437304303
from your 11 most-recent CATAN games gives p-values

0.853095 ​ , ​ 0.491452 ​ , ​ 0.248882 ​ , ​ 0.134306 or 0.141482 ​ , ​ 0.375248 ​ , ​ 0.446907
, ​ 0.0496748 ​ , ​ 0.451987 ​ , ​ 0.388373 or 0.414333 ​ , ​ 0.254257 ​ , ​ 0.0922030 ​ ​ ​ .


I did 11 rather than 10 because 0.491452 was from a game where ​ ​ ​ Someone left it before the end ​ .

The audience you are trying to convince likely does not include people who have even heard of p-values. By that, I mean you're just preaching to the choir without an explanation of why it's relevant.

Catan is a pretty random game. People who do not like randomness, should not play it. Especially those who do not understand how randomness works.

Many people just don't want random results. They want results, that are meddled with in order to have more "smooth" or "fair" results. They think randomness would give fair results, which it doesn't. So what they actually really want is the site to meddle /manipulate results and then lie to them and say those (more) fair results are random.

Randomness isn't fair or evenly distributed on a single-game-basis (or on a few-game-basis).

Meeplelowda wrote: ↑09 November 2023, 01:46 I appreciate you trying to fight the good fight, but I've yet to see someone respond by saying, "oh, I understanding now and withdraw my conspiracy theory about how the site is rigged." They are only ever looking for someone to agree with them. I suppose these responses still help guide people who are willing to learn.

Just because they aren't willing to post a retraction doesn't mean they don't hear the correction. It's entirely possible they sheepishly accept it and say nothing out of embarrassment. It's also entirely possible they get frustrated that nobody believes their conspiracy theory and give up. Both are positive outcomes.

Saying nothing just encourages them that they are right, along with other people who may see the post and not understand statistics.

Meeplelowda wrote: ↑09 November 2023, 06:38

Gulchen wrote: ↑09 November 2023, 05:55 on copy-pastes of ​ the analogue of https://boardgamearena.com/gamereview?table=437304303
from your 11 most-recent CATAN games gives p-values

0.853095 ​ , ​ 0.491452 ​ , ​ 0.248882 ​ , ​ 0.134306 or 0.141482 ​ , ​ 0.375248 ​ , ​ 0.446907
, ​ 0.0496748 ​ , ​ 0.451987 ​ , ​ 0.388373 or 0.414333 ​ , ​ 0.254257 ​ , ​ 0.0922030 ​ ​ ​ .


I did 11 rather than 10 because 0.491452 was from a game where ​ ​ ​ Someone left it before the end ​ .

The audience you are trying to convince likely does not include people who have even heard of p-values. By that, I mean you're just preaching to the choir without an explanation of why it's relevant.

It would help somewhat if Gulchen also explained what these p-values means.

ok here's a dice distribution from my recent game

2 0
3 15
4 4
5 20
6 16
7 5
8 7
9 11
10 5
11 15
12 4

seems very strange to me and rather unexpected, anyone wanna run an analysis on them?

table id 445859988

yfimc wrote: ↑03 December 2023, 00:51 ok here's a dice distribution from my recent game

2 0
3 15
4 4
5 20
6 16
7 5
8 7
9 11
10 5
11 15
12 4

seems very strange to me and rather unexpected, anyone wanna run an analysis on them?

table id 445859988

You are quoting percent values of game 445859988 without telling the number of moves in this game (there were 55 moves). Unfortunately, there is not much value of analysing a series of only 55 moves because the number of rolls is very low.

There are 36 "different" outcomes when rolling 2 dice. Randomness does NOT work in a way that you get each of the 36 outcomes in 36 rolls - that has a chance of perhaps 1 in a billion (just a crude guess). It is to be expected that the results of 36 rolls will cluster heavily in most cases. 55 moves is not much more...

Your dice numbers were (pls apologize rounding errors):
2 0
3 8
4 2
5 11
6 9
7 3
8 4
9 6
10 3
11 8
12 2
Nothing special to say about such a sequence.

In my last tournament game I had the following series:
2 1
3 3
4 3
5 5
6 17
7 15
8 10
9 4
10 3
11 0
12 0
Well, a large amount of 6s! 6s and 7s together are more than 50% of all rolls, far more than expected. If I remember correctly, the first 8 rolls (it was a 4player game) were only 6s and 7s. Nothing more to say about it...

It was a real life tournament, played with ordinary dice. Not played on BGA.

Silene wrote: ↑09 November 2023, 11:50 Many people just don't want random results. They want results, that are meddled with in order to have more "smooth" or "fair" results. They think randomness would give fair results, which it doesn't. So what they actually really want is the site to meddle /manipulate results and then lie to them and say those (more) fair results are random.

Randomness isn't fair or evenly distributed on a single-game-basis (or on a few-game-basis).

that's the option of playing catan with cards instead of dice. because the cards have a fixed set of numbers so you're guaranteed a smooth distribution if you go through the whole deck.

euklid314 wrote: ↑03 December 2023, 01:46 Nothing special to say about such a sequence.

huh i thought it was very strange! so many 3, 5, 11 and very few 7 and 8. actually there were as many 3s as 6s! there were as many 11s as 6s! there were more 3s than 7s and 8s combined! that's wild.

euklid314 wrote: ↑03 December 2023, 01:46
In my last tournament game I had the following series:
2 1
3 3
4 3
5 5
6 17
7 15
8 10
9 4
10 3
11 0
12 0
Well, a large amount of 6s! 6s and 7s together are more than 50% of all rolls, far more than expected. If I remember correctly, the first 8 rolls (it was a 4player game) were only 6s and 7s. Nothing more to say about it...

It was a real life tournament, played with ordinary dice. Not played on BGA.

your distribution is totally normal. 6, 7, 8 are the most common numbers which is what you would expect. 5 and 9 are next most common, followed by 10, 4, and 3. that's basically what i see in most offline games. that's the opposite of what i showed in the example that i shared. (admittedly it's a more egregious example but still indicative of the type of pattern i see online.) rather than 6, 7, 8 the most common numbers it was 3, 5, 6, 11.

euklid314 wrote: ↑03 December 2023, 01:46
You are quoting percent values of game 445859988 without telling the number of moves in this game

huh, i just shared the numbers from the end of game statistics. i thought those were the count of rolls, not percentages??