Good idea, but it should give some context and/or a link to the blog post. Your different set of players in the future are going to be mystified if they read this cold.

If as Ais says dice with powers-of-2 sides are unaffected, we could still roll those without worrying.

There’s no issue with referencing blog posts in the ruleset, is there?

Never mind, I figured out a way to make sure that the blog post in question isn’t going to be interpreted as being part of the ruleset or gamestate.

We’re talking about differences of 0.000005% here

Seems a really silly thing to worry about

I’m not quite up to date on this matter and am willing to defer this one while I’m still catching up

as it would prevent even just making test rolls, and the benefit seems incredibly marginal

  It’s something that should be fixed, but not so urgently that we have to put the whole concept of random chance on hold.

The basic problem at present is that you can’t actually make a dice roll legally (unless the number of sides is a power of 2), as the rule requires both exactly correct probabilities and the use of the dice roller, and it’s impossible to comply with both requirements at the same time.

That said, I don’t think this proposal actually fixes the problem – I’d be quite happy to play with the dice roller’s actual probabilities as they are so close to correct (out by less than 1 in a billion per side on the dice), so I think the correct fix is to just treat the dice roller results as accurate even though they technically aren’t quite.

if it’s just 1-in-a-billion per side. I’d misread Ais’s large-number posts as this being a bigger deal, with the history of players occasionally noticing that the Dice Roller had repeatedly picked the first item from a list several times in a row.

We should just note that the Roller is imperfect but close enough.

I would prefer a rule that states that the outcomes of the dice roll are fair enough for an otherwise legally executed roll, or, more likely, that we all just accept it and get on with our lives. I’ll be honest, my prior when considering any random selection is that it isn’t going to be 100% ‘fair’ as true fair random selection has historically been extremely hard to secure - but the differences are so slight as to hardly matter. That is what I believe to be the case here.

I do appreciate that this is probably frustrating in the light of ais’ DoV in the last dynasty; when I made the comment in that thread I was responding to my own perception that there was a casually observable bias towards low numbers. It looks like that might not be the case.

per others

I had not realized some of the points brought up. CoV.

I appreciate the calls for revision, but if most everyone is actually ok with the Dice Roller and it’s observed flaws, I don’t see any urgent need to call it into question.

Withdrawn