After 20 times playing, I am done with the BGA version

[Meeplelowda]

Btw, here is what I did yesterday with silentProtest: https://docs.google.com/spreadsheets/d/ ... edit#gid=0

I took all the games present on the bug report and reported by silentProtest. Then I took all the logs, separated the dice rolls, and regrouped them in the main page you see.
Then for each turn, I keeped the last three unique numbers rolled, and if the next roll is one of them, i display it in white.
And you can see the results, and see by yourself.

There is one blank line between each game to add comments on it if you want.
Feel free to pm me if you want I explain deeper.

(and my doc can be used to extract dices more easily, like to count the double in backgammon games, just ask me)

You should drop this link in the bug report.

[Romain672]

I would be very suspicious if I saw a bell curve across multiple games. On average it should be a triangle. The number of combinations for rolls 2..12 is 1/2/3/4/5/6/5/4/3/2/1 which most certainly isn't any sort of curve.

On average, it should look like that: https://www.zupimages.net/up/22/29/xu8k.jpg

[steve5332]

I would be very suspicious if I saw a bell curve across multiple games. On average it should be a triangle. The number of combinations for rolls 2..12 is 1/2/3/4/5/6/5/4/3/2/1 which most certainly isn't any sort of curve.

On average, it should look like that: https://www.zupimages.net/up/22/29/xu8k.jpg

Precisely. Excellent chart! The orange background nicely shows the "average" triangle, and the blue sets of data very nicely show how wild some random sets of results can be. If the rolls from the 275k+ games of Catan on BGA were put together it would tend towards that orange background.

[SluggerBaloney]

Deleted

[steve5332]

Does that look like a triangle to you?

The orange expected outcomes chart is a perfect triangle, yes, because the expected outcome is a triangle (not a bell curve, not random, not anything else). Each individual chart of blue bars isn't a triangle, no, but on such a small sample size that is to be expected.

[GeneralIgnorance]

https://boardgamearena.com/table?table=299049866

57 rolls. Two sixes.

Everything's normal.

[Romain672]

https://boardgamearena.com/table?table=299049866

57 rolls. Two sixes.

Everything's normal.

That claim will be easy.
Here is my bernouilli table: https://docs.google.com/spreadsheets/d/ ... 1846324187
31/36 in C3 (probability to not roll a 6)
line 62 (57 rolls)
And we add the colmns C (roll no 6), column D (roll one 6), column E (roll two 6).

I find 1.027% . So this happen more than one every 100 games.

But we can do the same claim for every number from 2 to 12. So I can multiply that probability per 11 ( 1-((1-0.01027)^11) to do it proprely).
Which give 10.7% of occuring.
So that happen more than once every 10 games
Everything is normal

Please give me probability under 1/10k and I will start putting value on your weird games

[muntzer]

Please refer to my new thread: Board Game Arena, Catan, virtual dice and randomness - a short trial

viewtopic.php?t=27679

[bagofnails]

I have read this thread, and went on BGA to play a quick game. It is not BGA's version that I am done with, it is Catan in General. Three player game, the first twenty or so moves we are all fairly equal, on 3,4and 5 points. Then I have an extended string of bad luck on the dice. I stand still on five points and both players overtake me, to the point where the person who was on three points is now on nine points and I am fully blocked in, unable to expand in any direction. I did nothing wrong, and my game is effectively over. I had a sort of similar instance of this in real life a few weeks ago, but with an opposite result for myself. I think my position was ok, but the die were just terrible for me early in the game. The dice even out a bit after a time and I claw my way back to aybe being on five points versus seven and nine for mny opponents. Then the dice just fall for me pretty much every turn, and I win the game. The person who was on nine points just stood still for twenty minutes, not being able to do anything at all.

May as well play snakes and ladders.

[stellalevigata]

I completely agree with you.

I've played several games in different boardagame on BGA and after a consistent number of games I see a pattern of the number or a pattern of card randomic generation.

For example this is one of the laste game I played for Ticket to ride.
https://boardgamearena.com/gamereview?table=320789345

It is really rare to have all of that objectives all together in one game. The same occurs sometimes to me, this is how I sometimes win.

I could have more example on Cubirds, Railroad, Azul, etc

It is really broken and it is really disappointing for a platform like this.

I'm sure there is someone willing to help them in order to avoid this kind of event.