Random generator is broke

[Sky Jetpack]

This game's dice random generator is broken. The percentage of doubles rolled is not statistically accurate. We see 20% to 30% regularly. Also, I rolled 3&5 three times in a row. The odds of rolling a 3 and a 5 with two dice three times in a row are approximately 0.00214% or 1 in 46,656. Not only did the game provide this wonderful outcome but did it at a time there were no tiles to take from 3 and no 3 spaces in my duchy. This is just one example of the broken programming of this game. It is a great game except for this problem, imho.

[iBlameLuck]

A quick glance at your CoB stats shows you at 16.29% average doubles rolled, which is almost exactly the average. Also your calculation of rolling the same numbers is off. You're calculating 1/(36*36*36) but you're failing to account for the fact that the numbers are interchangeable (3-5 AND 5-3), making the odds 1/(18*18*18) which is roughly ~0.02%, meaning every game you play you have about a 1% chance of the same roll happening 3 times. Entirely plausible. Sounds like this is just a baseless rant.

[RicardoRix]

The players statistics speak for themselves.

That's the odds of rolling any combination 4 times in a row. There is no need to particularly look at just 3-5.
So looking at any combination 3 times in a row is just 1 in 324 (18*18).

You're going to have to justify that the RNG somehow knows about which numbers need not to be rolled, and taking into account that this bias doesn't show in the player statistics.

That's some high accusation thrown at the developer and their code. Their code is also freely available to check.

[Jellby]

Even if it was 1 in 46,656, so what? You witnessed a rare event, that proves nothing, except that it is possible.

But as noted, the odds were much higher. (Although it's not exactly 1 in 18*18, you need to take into account that doubles can only occur in one way, while non-doubles can occur in two ways.)

[Pistol Star]

53 (35) example => For my understanding it doesn't matter what the first of the 3 consecutive identical rolls was, except for the 1/6 chance of rolling doubles, because doubles are much less likely to hit 2 more consecutive times. These non double combinations will be rolled 5/6 times per roll. Then you only need 2 more matching consecutive rolls.

e.g. you roll 42 or 24, then you need two more 42's/24's to reach 3 consecutive identical rolls. 1/18*1/18=0.30864197530864197530864197530864%
100/0.30864197530864197530864197530864=324

So after an initial non double roll (83.33% probability) a player is expected to hit 3 consecutive rolls once every 324 times.
As a little reminder:
2 player game => 50 dice rolls per game
3 player game => 75 dice rolls per game
4 player game => 100 dice rolls per game

So we can easily see that we expect to see this event every couple games and twice as quickly in a 4 player game compared to a 2 player game.
Of couse sometimes we will see it multiple times per game and other times we won't see it for 20 consecutive games.
That's normal and will even out over a large sample size.

Let's also not forget, that it's easy to play this game multiple times per day online, as it is quite quick to play for most players.
This leads players to experience these kind of events tremendously more frequently compared to offline play and then it's easy to think the RNG is malfunctioning.

Maybe my logic is flawed though.

[Sky Jetpack]

So... I appreciate all of the input and I do have to admit, I posted after a very difficult game so, yes, my comments were out of frustration. Evidently, I am lacking in understanding of statistical calculations. There are however, what seems like some very unlikely incidents while playing this game that, to my observation, do no happen as frequently in real life. Like rolling double 6's twice in a row at the moment neither the main board or my duchy provide any opportunities for a 6, despite the fact that I have a 9 tiles stringing across my duchy for adjacency. I will now resign my grievance and try look at any future anomalies as a just "challenge enhancement" to overcome while playing the game. All in all, I am grateful for such a great gaming service provided by BGA and their contributors. Game on!

[Pistol Star]

So... I appreciate all of the input and I do have to admit, I posted after a very difficult game so, yes, my comments were out of frustration. Evidently, I am lacking in understanding of statistical calculations. There are however, what seems like some very unlikely incidents while playing this game that, to my observation, do no happen as frequently in real life. Like rolling double 6's twice in a row at the moment neither the main board or my duchy provide any opportunities for a 6, despite the fact that I have a 9 tiles stringing across my duchy for adjacency. I will now resign my grievance and try look at any future anomalies as a just "challenge enhancement" to overcome while playing the game. All in all, I am grateful for such a great gaming service provided by BGA and their contributors. Game on!

I've played the game 875 times on BGA and I would bet a lot of money that the RNG is working correctly. I am sure that the vast majority of players, especially the ones who understand statistics quite well, share my observations.