euklid314 wrote: ↑24 March 2026, 20:03
Deckeon wrote: ↑24 March 2026, 18:31
Doubt anyone really cares about this reply but I had to cut my game play short ( some health issues)
Basically i can conclude I am wrong about doubles/triples. So much so I might consider the reverse true...
Of course I care because you initially called us forum users to simply trust your feeling because it was so obvious for you. I wanted to reply that you should simply trust me (as a mathematician) - but it is not a matter of trust. You can simply make tests and look at the numbers.
If BGA does use the cryptographically secure PHP function random_int() - see FAQ - and all their tests and those of game developers, made on millions of random calls show the function does what it is supposed to do - one can simply trust BGA.
Whenever a random number is called by some BGA game the function random_int() does not know what the result is used for. When the next number is called in the same game, hundreds of other random numbers will have been generated by the same function and was used in different games. What you see as a double in your game is not a double for the random function at all. There will be hundreds and thousands of random numbers produced inbetween. The function does not care of your ELO, if you attack a space ship or roll a Catan dice. It does not know that the sum 7 has a special effect in a game of Catan or a 1 has a special meaning for your game.
Have fun with further analyzing your dice rolls. You will find lots of interesting patterns to focus on!
These threads, which keep popping up over and over and over again, are a great illustration of the problem raised in
The Death of Expertise. They fall into two broad category: 1) the ones based on the "evidence" that something "feels off"; and 2) the ones where the person at least attempts some kind of analysis.
You can't really do anything to persuade someone who's just going on vibes. But you'd think the second type of person, who is at least attempting to be analytical, could be persuaded.
But nope. They are shown that their sample size is too small, or they used the wrong statistical test, or their probability calculations are just plain wrong, sometimes by bona fide experts in this type of analysis. Yet it falls on deaf ears. For some reason it doesn't sit well with some people that, however smart they think they are, there are people who literally do this for a living and know what they are talking about. It doesn't mean you aren't smart just because you don't understand it, but people take it personally.
You don't necessarily need a PhD in probability theory. But you need some kind of study where your understanding is
assessed to make sure you get it. It doesn't have to be a university course, but you need to crack open a physical or digital book and actually work problems.
