Board Game Arena

I have had games where the face 7 of the dice appears up to 3 times, one after the other. I've had times where the 7 doesn't come up until I collect a lot of resources (up tu 7, losing the resources); this has happened in multiple games. There are users who draw the number 7 up to 5-8 times per game.

This can't be completely random. The random dice system needs to be fixed. It is broken, coming to favor a particular player.

More than random or not, it seems to be broken, because there is no 7 face!

All of those events are not only possible but inevitable with completely random dice.

This is what happens with randomness. It's why some people don't like to play game like Catan.

If you have played Catan a lot, you've probably had the luck go in your favor in some games. But people rarely take note of that.

In your most recent 8 games of Catan as of when I started on this post, 2 were cancelled, and the 6 completed games had a total of 425 rolls.
That gives a sample average of ​ 70 + 5/6 ​ rolls per game. ​ ​ ​ ​ ​ ​ ​ The games can have 3 or 4 players, and ​ ​ ​ (70+(5/6)) / (3+(1/2)) ​ = ​ 20 + (5/21) ​ .
Accordingly,

in the three-7s-in-a-row paragraph, I pretend each game has 71 rolls
and
in the many-7s-by-users paragraph, I pretend each player gets 20 rolls

.


For any particular sequence of 3 rolls in a row, the probability of that sequence being all 7s is 1/216. ​ ​ ​ A 71-roll game has 69 sequences of 3 rolls
in a row, and treating those as independent would give a probability of ​ 215/216 ^ 69 ​ that there is _no_ sequence of at least three 7s in a row.
That would give a probability of approximately 0.27 that there _is_ a sequence of at least three 7s in a row. ​ ​ ​ Nearby events from those 69
are positively correlated (given that A,B,C is three 7s in a row, the conditional probability of B,C,D being three 7s in a row is 1/6), so 0.27 is an
an overestimate, but even restricting to the 23 non-overlapping sequences of three rolls in a row gives a probability of ​ 0.1 . ​ ​ ​ I ran Python code
to compute the exact probability for 71-roll games, and it's between ​ 0.2370 ​ and ​ 0.2371 . ​ ​ ​ https://paste.ee/p/CVWVl has a paste of that run.

Sevens out of 20 rolls, on the other hand, is a simple binomial. ​ https://en.wikipedia.org/wiki/Binomial_distribution
I wrote Python code to compute the extremeness of the times you gave, and it reports that numbers at least as extreme as 5,6,7,8 have
total probability of approximately ​ 0.3617 , 0.1279 , 0.03714 , 0.01125 ​ respectively. ​ ​ ​ https://paste.ee/p/vIXc5 has a paste of that run.
Bear in mind that ​ ​ ​ you have played over 10 games of Catan ​ and ​ in each of them, you had 2 or 3 opponents, rather than just 1 opponent ​ .


Give us links to some of the games referred to in your resources sentence. ​ Additionally, bear in mind
that the number of rolls per player varying from 20 will make it easier for players to get seven or eight 7s.
If you still think those have happened too much, then also give us links to games where those happen.

If you can't understand that RNG works like this (and I'm sure you've been on the lucky side of this before), Catan is not going to be the game you want to main on here. What I find that helps is my secret shrine to RNGesus that my cat and I pray to each night.