Board Game Arena

Yes, we know, humans are terrible pattern spotters.

Which is why I didn't post this until I'd noticed the phenomenon, registered that I had, and then started looking for its absence as well as its presence. Now at the stage where I thought I would ask others.

What I observe is, when an opponent on the bar rolls and can ot move, I appear much more likely than normal <approx. 1 in 20 natural odds each time> to then roll exactly the same as they did.

It's just possible that a call to a slowly updating time dependent random number donor service would suffer this. Has anyone noticed the same thing?

In order to understand you correctly: If your opponent is on the bar and he rolls say 3-5 and he fans (i.e. he cannot move because your homeboard is blocked at 3 and 5) then in the next move you roll a 3-5. Is that the pattern you have noticed?

Since you realize yourself that we humans are bad pattern spotters (i.e. we are too good at seeing patterns everywhere): What are the numbers? Have you for instance counted all your opponents fanning-situations in your, say, last 20 games and counted how many times you repeat his roll?

(edit, i missread the claim, skip this post, ty )

This claim has been done twice since last week, so I would put way more value to it.

But if it's true, and since that claim is really specific, can someone do any proper maths/statistics on that? Thanks.


Note it's still pretty likely it's just because of luck.

MisterTom wrote: ↑14 October 2022, 22:00I appear much more likely than normal <approx. 1 in 20 natural odds each time> to then roll exactly the same as they did.

What is "normal"? I probably made some mistake, but...

There are 21 possible results:

1, 1
1, 2 or 2, 1
1, 3 or 3, 1
1, 4 or 4, 1
1, 5 or 5, 1
1, 6 or 6, 1
2, 2
2, 3 or 3, 2
2, 4 or 4, 2
2, 5 or 5, 2
2, 6 or 6, 2
3, 3
3, 4 or 4, 3
3, 5 or 5, 3
3, 6 or 6, 3
4, 4
4, 5 or 5, 4
4, 6 or 6, 4
5, 5
5, 6 or 6, 5
6, 6

The chance of any double is 1/36, the chance of any other specific result is 2/36. The chance of any specific result twice in a row is the same number squared, so 1/36^2 for a double and 4/36^2 for a non-double. Now, summing them all gives (6*1 + 15*4)/36^2, or 1/19.6363... One in 20 seems like a pretty normal chance.