Jets de dés équiprobables ? / Equiprobable dice rolls?

Bonjour tout le monde, Je poste en français, j'espère que quelqu'un pourra me comprendre... Je me posais une question quant aux jets de dés : est-ce qu'ils sont bien équiprobables, équitables ou est-il possible qu'il y ai des bugs ? J'ai une joueuse sur ma partie de Warhammer, qui enchaîne les échecs critiques, de manière complètement impossible. Imaginons 10 jets à 50%, son ratio est peut être de 5 échecs critiques, 3 échecs et 2 réussites... C'est tellement étonnant que ça ne peut pas être juste de la malchance... Peut être y a t-il une fonction "dés pipés" ou quelque chose comme ça ? Je trouve ça vraiment bizarre... (ça n'arrive qu'à elle, mes autres joueurs on des jets plus "normaux"...) Hello everyone, I am posting in French, I hope someone can understand me ... I was wondering about the dice rolls: are they really equiprobable, fair or is it possible that there is bugs? I have a player on my game of Warhammer, which has critical failures in a completely impossible way. Imagine 10 rolls at 50%, its ratio is maybe 5 critical failures, 3 failures and 2 successes ... It's so amazing that it can't be just bad luck ... Maybe there is a "loaded dice" function or something like that? I find that really weird ... (it only happens to her, my other players have more "normal" rolls ...)

Salut, Je n'ai pas un reponse pour votre question.  Ma francais n'est pas bon et si tout est bon, j'ai plusiers choses a discouter en Anglais.  Je regrette. I have found the same thing as you in our D&D games.  Sometimes it's ridiculous.  :)  When we first started using roll20 last December, I know I found I preferred to just roll as did many of our players.  We've since gone back to using roll20 dice because we started using a character sheet.  But the issue continues. Computer generated random numbers require a "seeding method" that sets up a series of  numbers.  I know in my own case, in the past, when I've done it with MS Access (don't laugh), each time a random number was called, I reseeded the sequence based on the current time and it worked well enough.  I don't know if that's a common method or not. Hopefully someone can comment on how roll20 generates random numbers.  It certainly seems to be an area of potential improvement! All  the best. -- Tim

Hi Both, Roll20 uses Quantum Roll , which is about as random as you can get with computer generated random numbers. There's an interesting stats page that shows exactly how the rolls balance out across a d20. I personally have a player that never seems to roll higher than a 10, and I often have days where none of my monsters seem to hit anything, but I've had those moments too with physical dice. Sometimes the universe seems to conspire c:

In the famous words of Ziechael, "/r 1000d20/1000". 10.43 seems fair to me, working as intended for now. edit - Le_Djoff said: Imagine 10 rolls at 50%, 10.5 is the average on a d20, not 10.

Thanks for your answers ! :-) So I'm not the only one to have this impression sometimes ... I use d100, not d20... So, I will see next games...

The small number of rolls made in a session, combined with human bias, is not a good way to gauge probability. Just throw this in chat if you think the Quantum roller is cheating, if you get an average of 10 over 1000d100, something fishy is definitely going on, possibly a passing spaceship activating their Probability Drive.