Which number do you roll the most?

[loutil]

We were told that the list is replaced every hour. So one possible failure mode is if, for some reason, the list is no longer updating and got stuck on a list that was particularly non-uniform. But if the lists are still updating once per hour, then your explanation would require the sum total of those lists to be non-uniform.

The results I've seen are exactly what I would expect if the list was not being replaced. A small bias in a list 50,000 numbers would be reinforced every time the list was re-used. It would take just 139 fewer 1's, 66 extra 2's, etc. to achieve this result. Well within what you could expect from a random list of 50,000.

As you posted in the dice thread back in 2010, "we'll believe that there's a problem if you show us the evidence. ...demonstrate that there's some sort of bias in the dice. If you do that, we're not going to ignore the evidence". (http://www.conquerclub.com/forum/viewto ... 0#p2748566) I realize that you were talking to people who felt that the dice were biased against them personally, but I think I've provided plenty of evidence that the dice have a small, but consistent bias against rolling 1's.

You can trust that I have taken notice, and I hope it's evident that I'm not ignoring your results. But this result is still not enough to prove that the dice are non-uniform when considered sitewide, since there could be some hidden selection bias in your sample (which is small compared to the site as a whole). More importantly, since this effect is relatively small, if you want this to be a site priority then you need to collect the data suggesting that the dice are not fair.

Did you read my post above? I pulled 50 random players across many ranks and found clear evidence of dice bias. 48 out of 50 players had more 6's than 1's. Only 1 player had a dice average below 3.5 but many had an average above. How many players would I need to check to provide you evidence? My evidence is on top of Degaston's.

[Agent 86]

We pull a list of 50,000 random numbers from random.org. We then read the list out in sequence every time a roll is performed, and let the list wrap around continuously.

If this is true, then that's what the problem is. Using the list of 50,000 random numbers over and over again gives you biased results. This seems to be what is happening instead of using a new list of 50,0000 random numbers when the first list is finished.

Seems this could be the only logical solution to uneven numbers showing over time.

[Metsfanmax]

We were told that the list is replaced every hour. So one possible failure mode is if, for some reason, the list is no longer updating and got stuck on a list that was particularly non-uniform. But if the lists are still updating once per hour, then your explanation would require the sum total of those lists to be non-uniform.

The results I've seen are exactly what I would expect if the list was not being replaced. A small bias in a list 50,000 numbers would be reinforced every time the list was re-used. It would take just 139 fewer 1's, 66 extra 2's, etc. to achieve this result. Well within what you could expect from a random list of 50,000.

As you posted in the dice thread back in 2010, "we'll believe that there's a problem if you show us the evidence. ...demonstrate that there's some sort of bias in the dice. If you do that, we're not going to ignore the evidence". (http://www.conquerclub.com/forum/viewto ... 0#p2748566) I realize that you were talking to people who felt that the dice were biased against them personally, but I think I've provided plenty of evidence that the dice have a small, but consistent bias against rolling 1's.

You can trust that I have taken notice, and I hope it's evident that I'm not ignoring your results. But this result is still not enough to prove that the dice are non-uniform when considered sitewide, since there could be some hidden selection bias in your sample (which is small compared to the site as a whole). More importantly, since this effect is relatively small, if you want this to be a site priority then you need to collect the data suggesting that the dice are not fair.

Did you read my post above? I pulled 50 random players across many ranks and found clear evidence of dice bias. 48 out of 50 players had more 6's than 1's. Only 1 player had a dice average below 3.5 but many had an average above. How many players would I need to check to provide you evidence? My evidence is on top of Degaston's.

Your post is non-responsive to what I said. A small amount of dice non-uniformity is not a major issue. The only way that this could be an immediate priority is if you can provide clear evidence that the dice are not fair.

[Agent 86]

I would and most would agree that we would like to play full random, not partially random

[degaston]

You can trust that I have taken notice, and I hope it's evident that I'm not ignoring your results. But this result is still not enough to prove that the dice are non-uniform when considered sitewide, since there could be some hidden selection bias in your sample (which is small compared to the site as a whole). More importantly, since this effect is relatively small, if you want this to be a site priority then you need to collect the data suggesting that the dice are not fair.

I appreciate your attention to this issue, but I don't see how you can deny that the dice are non-uniform, given all the data that has been collected. You don't need a very large sample size when ~98% of the samples show 1's being rolled less than the expected rate. (probably close to 100% for players with 50k+ rolls) I've never said that the dice are not "fair" (in the sense that no one is getting an advantage over anyone else), just that they are not random (by definition, since the numbers are not being chosen with equal probability).

As for whether having random vs decidedly non-random dice is a priority for the web site, I guess that's up to the admins. I would think that if it was important enough to create a page for people to see their dice stats, then it would be important to make sure that the dice aren't biased. If it were my site, just basic curiosity would make me want to track down the cause of this anomaly. And considering how many forum threads people start to complain about the dice, I would want my dice randomness to be unimpeachable. But that's just me.

[Metsfanmax]

You can trust that I have taken notice, and I hope it's evident that I'm not ignoring your results. But this result is still not enough to prove that the dice are non-uniform when considered sitewide, since there could be some hidden selection bias in your sample (which is small compared to the site as a whole). More importantly, since this effect is relatively small, if you want this to be a site priority then you need to collect the data suggesting that the dice are not fair.

I appreciate your attention to this issue, but I don't see how you can deny that the dice are non-uniform, given all the data that has been collected. You don't need a very large sample size when ~98% of the samples show 1's being rolled less than the expected rate. (probably close to 100% for players with 50k+ rolls) I've never said that the dice are not "fair" (in the sense that no one is getting an advantage over anyone else), just that they are not random (by definition, since the numbers are not being chosen with equal probability).

As for whether having random vs decidedly non-random dice is a priority for the web site, I guess that's up to the admins. I would think that if it was important enough to create a page for people to see their dice stats, then it would be important to make sure that the dice aren't biased. If it were my site, just basic curiosity would make me want to track down the cause of this anomaly. And considering how many forum threads people start to complain about the dice, I would want my dice randomness to be unimpeachable. But that's just me.

Recall that the dice stats information was set up by a previous administration. And a new administration will have different priorities. All I can say is that, while you've provided enough evidence to warrant the code base being eventually looked at to ensure nothing is wrong, I can't promise it will be an immediate priority with this administration unlesss there's evidence of unfairness.

[Agent 86]

That's cool, the current admin are doing amazing stuff. When they get the time maybe they could have a look at the current random generator.

[loutil]

We were told that the list is replaced every hour. So one possible failure mode is if, for some reason, the list is no longer updating and got stuck on a list that was particularly non-uniform. But if the lists are still updating once per hour, then your explanation would require the sum total of those lists to be non-uniform.

The results I've seen are exactly what I would expect if the list was not being replaced. A small bias in a list 50,000 numbers would be reinforced every time the list was re-used. It would take just 139 fewer 1's, 66 extra 2's, etc. to achieve this result. Well within what you could expect from a random list of 50,000.

As you posted in the dice thread back in 2010, "we'll believe that there's a problem if you show us the evidence. ...demonstrate that there's some sort of bias in the dice. If you do that, we're not going to ignore the evidence". (http://www.conquerclub.com/forum/viewto ... 0#p2748566) I realize that you were talking to people who felt that the dice were biased against them personally, but I think I've provided plenty of evidence that the dice have a small, but consistent bias against rolling 1's.

You can trust that I have taken notice, and I hope it's evident that I'm not ignoring your results. But this result is still not enough to prove that the dice are non-uniform when considered sitewide, since there could be some hidden selection bias in your sample (which is small compared to the site as a whole). More importantly, since this effect is relatively small, if you want this to be a site priority then you need to collect the data suggesting that the dice are not fair.

Did you read my post above? I pulled 50 random players across many ranks and found clear evidence of dice bias. 48 out of 50 players had more 6's than 1's. Only 1 player had a dice average below 3.5 but many had an average above. How many players would I need to check to provide you evidence? My evidence is on top of Degaston's.

Your post is non-responsive to what I said. A small amount of dice non-uniformity is not a major issue. The only way that this could be an immediate priority is if you can provide clear evidence that the dice are not fair.

You would seem to be correct in your assumption that the dice are fair as there does not appear to be any discernible advantage to the obvious bias of the dice. However, having said that, I would think that proof showing dice bias and a clear lack of "randomness" would be a high priority. You poison one piece of fruit on a tree you tend to give the perception that the whole tree is poisoned.

[degaston]

Recall that the dice stats information was set up by a previous administration. And a new administration will have different priorities. All I can say is that, while you've provided enough evidence to warrant the code base being eventually looked at to ensure nothing is wrong, I can't promise it will be an immediate priority with this administration unlesss there's evidence of unfairness.

I like the new features that have been implemented lately, but what could be more important to the game than having good quality random dice? It's the one feature that players use on almost every turn, and it shouldn't be that hard to fix. Even a local pseudo-random number generator would produce better results than what we're seeing. Or mod two sources together - random.org data with a PRNG so that if one fails in some way, the other will compensate for it.

[AslanTheKing]

the only solution on internet dicerelated games would be to change the numbers on the dice,
no more maximum of 6, there should be 10 as highest number, this is technically possible
and would make people stop comparing it to a normal dice used on real life board

u might be laughing at this comment, but this is the future of internet dice gaming
it will be more even

[BigBallinStalin]

this is the future of internet dice gaming

it will be more even

[AslanTheKing]

can u see the naked woman in my avatar?
its like a hidden mona lisa.

[Vid_FISO]

Another that gets 2s with 1s lagging well behind

me - 2,4,5,3,6,1
opponents - 2,6,4,5,3,1

might explain why a third of my ranking points have disappeared over the past couple of months (other than those lost by crap play of course)

Surely a straightforward task to check whether a) the same numbers list is being used constantly/ frequently and b) whether the list(s) contain a bias?

[Agent 86]

Well looks like Random.org and the magical atmospheric pressure forgot to program 1's into the equation. So has anyone rolled more 1's than any other number and if so please tell. I bet it will be silent because I've checked the dice on most of the high rollers and it's staggering how the dice are purely not random. So I think site admin should send the data to the rip-off site who claim randomness and then make the offer of a complete explanation to their claim of random dice. What a rip-off site, complete shambles.... Oh by the way I hope conquer club doesn't pay them for this???

Stupid wikipedia even promotes this?

Random.org
From Wikipedia, the free encyclopedia
Web address Random.org
Owner Mads Haahr
Alexa rank positive decrease 6,303 (November 2013)[1]
Current status online

Random.org is a website that produces "true" random numbers based on atmospheric noise.[2] In addition to generating random numbers, it has free tools to simulate events such as flipping coins, shuffling cards, and rolling dice. It also offers paid services to generate sequences of random numbers and act as a third-party arbiter for raffles, sweepstakes, and promotions.

The website was created in 1997 by Mads Haahr,[3] [4] a doctor and computer science professor at Trinity College in Dublin, Ireland. Random numbers are generated based on atmospheric noise captured by several radios tuned between stations.[5][6]

The problem is the word "true" in the statement..it's false. Conquer Club might as well use Random.Bullshit as it would be better for laughs.

[iAmCaffeine]

Ones are still the lowest for both myself and my opponents, by a few thousand out of 120k+

[degaston]

Well looks like Random.org and the magical atmospheric pressure forgot to program 1's into the equation...

I don't think random.org is the problem. You can go to the website and get batches of 10,000 numbers at a time and test them. I've done it and I see no long-term bias. What you will see is that within any set, the numbers will not all appear equally. This would normally be cancelled out by subsequent sets, but if you use the same set over and over again, whatever bias it has will be reinforced. I think this is the cause of what we are seeing, but it's not a priority for admin because it isn't giving an unfair advantage to anyone.

[Agent 86]

Well looks like Random.org and the magical atmospheric pressure forgot to program 1's into the equation...

but it's not a priority for admin because it isn't giving an unfair advantage to anyone.

This can't be true, if it isn't random it must affect many members?

[degaston]

Well looks like Random.org and the magical atmospheric pressure forgot to program 1's into the equation...

but it's not a priority for admin because it isn't giving an unfair advantage to anyone.

This can't be true, if it isn't random it must affect many members?

Did you see all the charts I made? This affects everyone, but it affects them all equally, so no one gets an advantage over anyone else.

[BigBallinStalin]

I won't be satisfied until I get to roll more sixes than everybody else.

[Agent 86]

It still continues, so random.org has a serious problem..nothing random about these dice. 2 is ultimate winner, then 4. Even for the newer members. Dice loop we are all caught in a dice loop.

So degaston decides to play ? guess that would even things up lol

[degaston]

It still continues, so random.org has a serious problem..nothing random about these dice. 2 is ultimate winner, then 4. Even for the newer members. Dice loop we are all caught in a dice loop.

So degaston decides to play ? guess that would even things up lol

Not sure what you're talking about. What did I decide to play, and how would that even things up?

[Agent 86]

It still continues, so random.org has a serious problem..nothing random about these dice. 2 is ultimate winner, then 4. Even for the newer members. Dice loop we are all caught in a dice loop.

So degaston decides to play ? guess that would even things up lol

Not sure what you're talking about. What did I decide to play, and how would that even things up?

Really, you need me to show your work..ok
Game 14455995
Game 14447584
Game 14447462
Game 14373944
Game 14356930
Game 14347707

All private games, so did you invite yourself

[degaston]

It still continues, so random.org has a serious problem..nothing random about these dice. 2 is ultimate winner, then 4. Even for the newer members. Dice loop we are all caught in a dice loop.

So degaston decides to play ? guess that would even things up lol

Not sure what you're talking about. What did I decide to play, and how would that even things up?

Really, you need me to show your work..ok
Game 14455995
Game 14447584
Game 14447462
Game 14373944
Game 14356930
Game 14347707

All private games, so did you invite yourself

Those are all from the Guide program, which has since been shelved. When someone new signed up with the site, they were offered a chance to play a game with an experienced player for no points. People who had signed up to be guides would get an email, and the first person to respond would join the game. We tried to be welcoming, but most new players never came back to make another move, so they dropped the program.

I've been freemium for a while, so I couldn't create a private game even if I wanted to, and even if I had, how would that relate to the dice stats?

[Agent 86]

So every player dead beats, no points fair enough..looks like this was a waste of time?

[Metsfanmax]

It still continues, so random.org has a serious problem..nothing random about these dice.

The dice weren't random, but as has been expressed earlier in this thread, it is not the fault of random.org.