Forum adverts like this one are shown to any user who is not logged in. Join us by filling out a tiny 3 field form and you will get your own, free, dakka user account which gives a good range of benefits to you: No adverts like this in the forums anymore. Times and dates in your local timezone. Full tracking of what you have read so you can skip to your first unread post, easily see what has changed since you last logged in, and easily see what is new at a glance. Email notifications for threads you want to watch closely. Being a part of the oldest wargaming community on the net. If you are already a member then feel free to login now.

 spiralingcadaver wrote:
Just putting it out there, after around 300 spins (numbers updated in my post linked above), I'm still seeing about the same trends. Things are a little more even, but there are definite outliers that have stayed outliers.

While I don't doubt this... 300 isn't really enough for a d20.
I started to get convergence on my excel model (formulas too annoying on dakka posts) and still wasn't getting a consistent converged result even at 2000 (used a proxy of 1% differential in expected rolls to actual as a sign of non convergence).

They started to get more consistent between 3000 and 4000.
That's a lot of spins.

Always is surprising when people are asked to pick random patterns, we expect them to be... well, more random than random. I am not worried about these, they look good.

On the possibility of these being biased based on the starting point (this is 100% true) unless you apply a random force to it... as your finger will be reasonably consistent this will be biased... but that's the same with a dice that you always roll the same way with the same facing pointing up in the same direction... not anything new there.

That's why I like my 3d6 ring . Because of how my thumb can reach it, I tend to spin the outer band more than the others, making it hard to "aim" for a specific result with 3 bands spinning. Getting tons of comments / questions on it, too

Just have to get the right size for the others, and I'll be set!

 Makaleth wrote:

 spiralingcadaver wrote:
Just putting it out there, after around 300 spins (numbers updated in my post linked above), I'm still seeing about the same trends. Things are a little more even, but there are definite outliers that have stayed outliers.

While I don't doubt this... 300 isn't really enough for a d20.
I started to get convergence on my excel model (formulas too annoying on dakka posts) and still wasn't getting a consistent converged result even at 2000 (used a proxy of 1% differential in expected rolls to actual as a sign of non convergence).

They started to get more consistent between 3000 and 4000.
That's a lot of spins.

Always is surprising when people are asked to pick random patterns, we expect them to be... well, more random than random. I am not worried about these, they look good.

On the possibility of these being biased based on the starting point (this is 100% true) unless you apply a random force to it... as your finger will be reasonably consistent this will be biased... but that's the same with a dice that you always roll the same way with the same facing pointing up in the same direction... not anything new there.

I have no idea... should the outliers be grouped, and should they stay proportionally outliers? Not questioning you, just checking since I don't know all that much about statistics.

Your results look quite alright and not really different to rolling regular d20 dice.
You could try and repeat your experiment using normal dice to see them producing patterns, grouped outliers, etc especially if you roll only a few times.

 spiralingcadaver wrote:

 Makaleth wrote:

 spiralingcadaver wrote:
Just putting it out there, after around 300 spins (numbers updated in my post linked above), I'm still seeing about the same trends. Things are a little more even, but there are definite outliers that have stayed outliers.

While I don't doubt this... 300 isn't really enough for a d20.
I started to get convergence on my excel model (formulas too annoying on dakka posts) and still wasn't getting a consistent converged result even at 2000 (used a proxy of 1% differential in expected rolls to actual as a sign of non convergence).

They started to get more consistent between 3000 and 4000.
That's a lot of spins.

Always is surprising when people are asked to pick random patterns, we expect them to be... well, more random than random. I am not worried about these, they look good.

On the possibility of these being biased based on the starting point (this is 100% true) unless you apply a random force to it... as your finger will be reasonably consistent this will be biased... but that's the same with a dice that you always roll the same way with the same facing pointing up in the same direction... not anything new there.

I have no idea... should the outliers be grouped, and should they stay proportionally outliers? Not questioning you, just checking since I don't know all that much about statistics.

you are doing it exactly right... just so hard to get statistical results on something like this. Someone did a chi-Suared test and it passed... so it's still ok.

If you rolled it 10 times in a row and it was all a single number first go... that would be compelling... but seeing some groups come up after a few hundred is normal. I always joke about the lottery that always pick at least one set of numbers next to each other (as that normally is the case) but it's more a joke as 1,2,3,4,5,6,7 is just as likely as any other set of numbers.... but we would cry foul if that occured.

Not grousing as such, but................

I had my e-mail 10 days ago (on the 18th) saying my order had been dispatched, but no tracking no. at all. Now I don't know if I should be worried at all, but has anyone else in the UK had theirs yet? Or should I be contacting them directly by now?

Worried
Lousy Britain.

I got the 2d6 as a novelty and surprise! That's what it is. At the price point I knew they couldn't be accurate, but they look nice and show that I'm a "classy" nerd.

Thinking these would be acceptable in anything but a casual setting is just idiotic.

 Bat Manuel wrote:
At the price point I knew they couldn't be accurate

We don't know (yet) whether they are "accurate" or not. The data suggests they are random enough, just like regular dice.

For all those discussing how accurate these are, I simply suggest that you spin the dial several times before reading a result. Three flicks means you're not starting on a set number.

I wouldn't try to get a casino to accept them. On the other hand, I got the rock-paper-scissors ring, and think it'll work just fine to break any unknown patterns I may fall into.

Should have posted earlier on this, by the looks of it. ssaid yesterday I hadn't received them yet, and asked if anyone knew if I should contact them about it ............ so of course, they turned up today. The only one I have tried so far is nicely sized, tight enough not to have the inner band move, not so tight as to press into my finger. It was a little 'gritty' the first time I span it, but went and washed my hands with the ring on, and now floats and spins very well. As it was a Y/N, and I have not spun it much yet, I'm not even going to try and work out how accurate it is and to tel the truth, it's not going to bother me that much.

I also have a %, and a 3D6 to try yet, but don't think they will have any problems. All 3 have been tried for fit, and fit well, and will need a good wash before I try them in anger, but I am well pleased with the lot of them.