Dices probability

Forum rules
Please DO NOT POST BUGS on this forum. Please report (and vote) bugs on : https://boardgamearena.com/bugs
Posts: 1
Joined: 29 December 2022, 05:15

Re: Dices probability

Post by WuhanLabTech »

I agree with OP. The die rolls on this site are not random. Too many low probability events happen (at least in my games).
User avatar
Posts: 773
Joined: 28 August 2020, 12:27

Re: Dices probability

Post by voriki »

WuhanLabTech wrote: 29 December 2022, 23:09 I agree with OP. The die rolls on this site are not random. Too many low probability events happen (at least in my games).
You have played a total of 127 games on this site. I think you should expand your pool of examples, or at elast give some examples in your games of very rare rolls(such as someone rolling 6 same faces multiple times in a so called random manner).

What people are not grasping is that dice rolls are not related to the game itself. Dice probabilities are entirely random as the only thing the game is is to ask the BGA-server to give a random number between 1 and 6. And that's the same probability server every other game uses on this site. Then the game translates those numbers to die faces and values equal to what was returned.

Claiming the dice are not random would mean that it is a problem for all games, and clearly it is not.
Posts: 3
Joined: 31 October 2022, 03:46

Re: Dices probability

Post by TeamBlueHeart »

Ez0ah wrote: 26 December 2022, 17:52
adrianbouvier wrote: 26 December 2022, 15:00 I am not speaking about feelings here, i am doing maths.
Yeah, you are speaking about feelings here. You are biaised to see the unlikely successes and forget about the majority of average results.

Also, 20 games is a statistically small sample compared to the thousands of games every day. So you could have had very lucky events in your games and it still wouldn’t prove anything. Accusing players of cheating based on a few random games is just pretty funny.
Thanks, I found it funny, too.💙
Posts: 5
Joined: 21 January 2023, 04:46

Re: Dices probability


Well, uou just need to stop thinking about probabilities. And see what are you tended to roll. Sometimes in one game you are tended to reroll many claws, sometimes different dices that helps you with panda Express or clown,and sometimes your dices have tendency to get points.
This is a pro tip that hopes help you.
User avatar
Posts: 2
Joined: 14 July 2022, 18:08

Re: Dices probability

Post by ahrnge »

Romain672 wrote: 29 December 2022, 13:56 Probability to get exactly 2 hearts in first roll: 1/6*1/6=1/36
Probability to get exactly 1 heart in first roll: 1/6*5/6+5/6*1/6=10/36
Probability to get exactly 0 heart in first roll: 5/6*5/6=25/36
You can see: 1/36+10/36+25/36 add up to 36/36, so 1.

Then you roll a second time:
If you rolled 2 hearts in first roll, the probability to get 0 more heart in the second roll is 1/1, 100%. So we keep that 1/36
If you rolled 1 heart in first roll, the probability to get 1 more heart in the second roll is 1/6, so 10/36*1/6=10/216
If you rolled 0 heart in first roll, the probability to get 2 more hearts in the second roll is 1/6*1/6, so 1/36. 1/36*25/36=25/1296
Then we add up all three, and get: 36/1296+60/1296+25/1296=121/1296.

Which give 9.34% in your simple situation.

I gave you why you are not the first one to think that probabilities is wrong in my first post, and the probability to get 6 hearts in 3 rolls (I will not detail though, since it's pretty complicated, but you can just apply the same thing that what I did here).

Perhaps an easier way to approach it is to think of each die individually:

So for your example of "What is the probability of rolling 2 hearts on 2 dice given up to 1 reroll for each die?"

We can just say that the probability for a single die with up to one reroll is 1/6 + (5/6 * 1/6) = 11/36, so the probability for both dice is (11/36)^2 = 121/1296 like you said.

I think this way of thinking about it scales up more naturally to more complicated questions like "What is the probability of rolling 6 hearts on 6 dice given up to 2 rerolls for each die?"

By considering each die alone then multiplying them together as before we get the single expression:
(1/6 + (5/6 * 1/6) + ((5/6)^2 * 1/6))^6 ~= .56%, like you mentioned above.

Your way of reasoning is also absolutely correct, but I thought maybe thinking about the dice individually was a thought worth raising.

Also the dice aren't rigged, but I get why it feels that way sometimes haha.
Post Reply

Return to “King of Tokyo”