dice analyzer addition

[friendly1]

Dice analyzer:

Hmmm,

Not sure but does anyone running dice analyzer have above average percentage on categories? Especially when playing against lower ranks?

I really dont care how much CC insists the dice are random, i know every time i run a series of games with lower ranks my dice averages drop, but when i play against the same or better ranks than my score my dice averages improve (actually i roll like a lotto 649 winner from time to time).

And I apologize but IN NO WAY do i believe this to be coincidence. And again I apologize, but if CC is truly concerned about dice randomness then damn well start tracking and posting results that are site specific. Its very simply done and an obvious concern why this function is not performed.

friendly1

[insomniacdude]

CC isn't concerned about dice randomness, because it knows that the dice are random. You are the one concerned about dice randomness. Shouldn't it logically follow that you should be the one to track data. Why should CC legitimize your argument when you have the power to do it yourself?

Every few weeks a new dice complaint thread pops up and it is always filled with people who insist that some random dice tilt, but they have no evidence to back it up. We need numbers if you want to be taken seriously!

[cicero]

I really dont care how much CC insists the dice are random, i know [they are not]"

The dice ARE random, lots and lots of analysis has been done on them both internally and by community members.

The statement in the sticky isn't there because CC don't understand that many people feel as you do friendly1, it's because this subject has been brought up repeatedly, researched (to varying levels of thoroughness) repeatedly and argued about repeatedly ... and to date the most widely held conclusion is that the dice are ... random.

start tracking and posting results that are site specific. Its very simply done and an obvious concern why this function is not performed.

CC isn't concerned about dice randomness, because it knows that the dice are random. You are the one concerned about dice randomness. Shouldn't it logically follow that you should be the one to track data.

friendly1, if it's easy to do as you say then insomniacdude's suggestion is your only option ... I appreciate that CC themselves have better resources, but they don't have a problem believing in the randomness of the dice.

Every few weeks a new dice complaint thread pops up and it is always filled with people who insist that some random dice tilt, but they have no evidence to back it up. We need numbers if you want to be taken seriously!

Actually I think that recently the number of dice rants has gone down. If so then perhaps the new boilerplate text in the B&S form should take the credit in that it makes people think twice.

Even if some then think a third time and post anyway

[insomniacdude]

Every few weeks a new dice complaint thread pops up and it is always filled with people who insist that some random dice tilt, but they have no evidence to back it up. We need numbers if you want to be taken seriously!

Actually I think that recently the number of dice rants has gone down. If so then perhaps the new boilerplate text in the B&S form should take the credit in that it makes people think twice.

Has it? That's a relief. I haven't bene in the forum much since around the php update (been busy) so I just made an assumption.

Friendly1, if it's really a big concern for you, I urge and plead for you to do some more research. I would love to be proven wrong on this. There is a Firefox add-on that tracks some dice stats, if that might help. But literally, there would be no way to prove randomness without first factoring in EVERYTHING that could affect the dice, including player rank, seniority, color, premium/freemium, order in the play sequence, etc. That would have to be weighed against the various game types and techniques, like the map, player amount, bonus cards, fog, fortifications, team/singles, private/public, etc. There could also be factors at work in the game itself, like continents owned, number of armies, etc.

If the dice really aren't random, it would have to be shown WHY they aren't random. And nobody has EVER been able to show that.

[Thezzaruz]

If the dice really aren't random, it would have to be shown WHY they aren't random. And nobody has EVER been able to show that.

Yes they have. By definition you can't produce randomness through an algorithm. You can however produce something that is close enough to random for our purposes. The biggest issue on here is that the dice throws aren't generated when they happens but instead taken from a large file of pre-generated numbers.

But literally, there would be no way to prove randomness without first factoring in EVERYTHING that could affect the dice, including player rank, seniority, color, premium/freemium, order in the play sequence, etc

None of those things should affect the dice throws though (and from what I've seen they don't).

[Vozzo13]

I don't know what else to say because it is so obvious that there are inaccuracies with the 'randomness' of the dice, and I for one cannot understand how or why CC will not admit it or correct the problem (why not try a different program?).

FYI: I just lost 8 to 2 rolling 3 dice against 1. In addition, last night I l ost 10 armies in a row TWICE rolling 3 against 2 and sometime 3 against 1.

My tenure, and many others I assure you, is coming to an abrupt end on what is otherwise a GREAT website.

CC, please admit and correct such an EASY and OBVIOUS problem.

[Arbotross]

If you look at the source of there "RNG" you'll notice that it isn't truely random (not that anything really can be) but Impossable to predict and more or less accurate. I have dice analyzer running and I'll admit my dice tend to favor the defender(not me) but very slightly in the end it wouldn't make or brake a round by its self.

Here is my numbers if you want them, I'll try and make them understandable

Attacker dice distribution Defender dice distribution
1s 4498 / 26990 (16.67%) 2393 / 14441 (16.57%)
2s 4644 / 26990 (17.21%) 2400 / 14441 (16.62%)
3s 4478 / 26990 (16.59%) 2481 / 14441 (17.18%)
4s 4444 / 26990 (16.47%) 2362 / 14441 (16.36%)
5s 4536 / 26990 (16.81%) 2364 / 14441 (16.37%)
6s 4390 / 26990 (16.27%) 2441 / 14441 (16.9%)

Battle Outcomes Actual Stats Ideal Stats
3v2 1901 / 1653 / 1508 (37.55% / 32.66% / 29.79%) (37.17% / 33.58% / 29.26%)
3v1 2370 / 1260 (65.29% / 34.71%) (65.97% / 34.03%)
2v2 20 / 31 / 45 (20.83% / 32.29% / 46.88%) (22.76% / 32.41% / 44.83%)
2v1 161 / 107 (60.07% / 39.93%) (57.87% / 42.13%)
1v2 5 / 36 (12.2% / 87.8%) (25.46% / 74.54%)
1v1 69 / 76 (47.59% / 52.41%) (41.67% / 58.33%)

Overall stats
Attacker threw 26990 dice.
Defender threw 14441 dice.
Wins / Ties / Loses 4526 / 1684 / 3032

[insomniacdude]

If the dice really aren't random, it would have to be shown WHY they aren't random. And nobody has EVER been able to show that.

Yes they have. By definition you can't produce randomness through an algorithm. You can however produce something that is close enough to random for our purposes. The biggest issue on here is that the dice throws aren't generated when they happens but instead taken from a large file of pre-generated numbers.

But literally, there would be no way to prove randomness without first factoring in EVERYTHING that could affect the dice, including player rank, seniority, color, premium/freemium, order in the play sequence, etc

None of those things should affect the dice throws though (and from what I've seen they don't).

I think we're approaching this from different angles. My concern isn't whether the dice are truly random or not (as you correctly noted, they are not *truly* random). However, they are random in the sense that anybody is just as likely to roll a certain number or set of numbers as any other person. In terms of this site, that's all the randomness of the dice really means. It means the even spread of circumstances across all users. In theory, I'm as likely to roll 8 trips 6s in a row as you or Lack or Twill, etc..

So when I say proof is needed that the dice aren't "random", I mean to say that there needs to be an indication on why I might be more likely than you to roll 8 trip 6s in a row. Otherwise all of the claims are simply circumstantial and subjective. There's nothing empirical or objective about them because, in theory, those same numbers are just as likely to happen to me as they are to him.

Even if there does happen to be a glitch or a bug in the distribution of the numbers, that glitch would apply to everyone equally. Everyone would be at the same advantage or disadvantage. Any collected numbers need to be weighed against other variables to determine which way, if any, the "randomness" tilts. It would have to show that it clearly does favor or doesn't favor or portion of the population.

Though like I said, I have my fair amount of disgruntled rantings at and about the dice, so I'd love to be proven wrong about the "randomness." Keep digging and doing research and observation! Don't let me stop you. I just simply think it's not a problem to worry about (mostly because, if there is a problem, it applies to everyone equally).

[cicero]

I don't know what else to say because it is so obvious that there are inaccuracies with the 'randomness' of the dice, and I for one cannot understand how or why CC will not admit it or correct the problem (why not try a different program?).

FYI: I just lost 8 to 2 rolling 3 dice against 1. In addition, last night I l ost 10 armies in a row TWICE rolling 3 against 2 and sometime 3 against 1.

My tenure, and many others I assure you, is coming to an abrupt end on what is otherwise a GREAT website.

CC, please admit and correct such an EASY and OBVIOUS problem.

this is classic "dice rant", someone lock this thread down !! ... quick before the world explodes ...
come on mods, do it quick, it's easy and ...

[friendly1]

What in the world makes you assume that there has been no tracking on my part?

The firefox addition is called dice analyzer, you know - the same as the title of this post.

I track usage of it on 2 different systems (i play from two different locations) however dice analyzer only carries two months history. (well 50 days actually) I also track usage when dice analyzer is active vs when it is dormant.

And I do not make claims the dice are unfair - I simply have noted several progressions: when playing against lower ranks over a concentrated period of time I consistantly drop in percentages on both attacking and defending on both systems. When playing against the same or higher ranks, my dice percentages increase. I can only offer 4 months of tracking, and realistically 2 months of curiosity in checking for a trend.

As per my first post, I asked simply if anyone had above average percentages on dice analyzer when playing consistantly against
players with lower ranks.

Tracking can be done easily, but i do think it would affect server load and speed delivery; perhaps substantially depending on activity (probably be one massive log file as well). As none of the above is desirable from either the players or CC's perspective, it makes sense it is not done.

However it is also quite possible to run a skewed or targeted algorithm - part time or full time - actually its just as easy as running an algorithm designed to produce random calculations. So I hold the opinion that simply stating the dice are random is not in any way substantial.

Comments made regarding bad dice rolls are wasted in this post, as are comments made about what factors affect the numbers generated. The only factors which affect the numbers generated are an algorithm. Logging and trending the numbers produced specific to CC are the only true audit for this algorithm.

And let me say once more: this is an opinion. It is not a rant, nor do I claim to be correct or accurate. However I do believe it makes perfect business sense to encourage better dice results to newer player to the site so as to encourage membership. And in the same line of reasoning it makes sense to provide better dice results to lower ranking players in games.

friendly1

[cicero]

friendly1

Your latest post makes your intention much clearer.
If I take your post apart sentence by sentence it is not because I am trying to refute every element in turn rather that I am trying to recreate the conversation that we would have had if you had presented your points sequentially and I could have interjected at the time ...

What in the world makes you assume that there has been no tracking on my part?

Your first post, whilst it mentioned that you had been using dice analyzer also said "damn well start tracking and posting results" ... from this and the remainder of that post I understood that you were 'suggesting' that CC should do some greater analysis of it's own and should incorporate the results of this analysis into the site. (And that this was the main thrust of your post notwithstanding the tracking you had done yourself.)

And I do not make claims the dice are unfair - I simply have noted several progressions ... I asked simply if anyone had above average percentages on dice analyzer when playing consistantly against players with lower ranks.

OK. I think that if the dice are in some way flawed even if that is not in itself unfair (because everyone suffers the same flaw) it is still something that CC could and should investigate. But, given how many times this has been raised in the past, CC should only do so on the presentation of mathematically valid research and argument ...

Tracking can be done easily, but i do think it would affect server load and speed delivery; perhaps substantially depending on activity (probably be one massive log file as well). As none of the above is desirable from either the players or CC's perspective, it makes sense it is not done.

Agreed.

However it is also quite possible to run a skewed or targeted algorithm - part time or full time - actually its just as easy as running an algorithm designed to produce random calculations.

Now perhaps that is the kind of valid research and argument that could legitimately lead somewhere. I don't know what a skewed or targeted algorithm is in this context, but if you present one then let those with the mathematical knowledge engage in discourse ...

So I hold the opinion that simply stating the dice are random is not in any way substantial.

But CC doesn't "simply state" that the dice are random (see end of this post). It presents it's respected source of random numbers and allows them to make the argument, quite convincingly in my lay opinion, that the numbers are indeed random.

Comments made regarding bad dice rolls are wasted in this post, as are comments made about what factors affect the numbers generated. The only factors which affect the numbers generated are an algorithm.

Agreed on the first part. The algorithm you refer to here ... If I understand correctly you are suggesting that CC's random numbers, and hence its dice, are generated by an algorithm. (And that said algorithm is flawed.) That is not the case (again see end of post).

And let me say once more: this is an opinion. It is not a rant, nor do I claim to be correct or accurate.

Not trying to misunderstand you here, but the tone of your original post - "I really dont care how much CC insists the dice are random, i know ..." - implied that you weren't expressing an opinion but presenting incontrovertible fact.
My subsequent post about rants quoted another poster and wasn't aimed at you. It was intended to be somewhat tongue in cheek and quoted the post which provoked my (allegedly witty) response ...

However I do believe it makes perfect business sense to encourage better dice results to newer player to the site so as to encourage membership. And in the same line of reasoning it makes sense to provide better dice results to lower ranking players in games.

I'll agree that it makes perfect sense to have genuinely random dice results. However it would have to be shown that the current CC dice are not genuinely random and only then could it be agreed that "better" dice were possible.

Finally (keep reading until you get here ) Regarding my statement that Conquer Club do not "simply state" their dice are random ...
Click "Help" in the main CC menu ... to get to the Conquer Club Frequently Asked Questions.
The first question (Q17) under the heading "About Gameplay" is "How do the dice work?"
Following the link there to http://www.random.org I browsed to http://www.random.org/randomness/ which is helpful on why and how their numbers are truly random. Personally I find the evidence there (and elsewhere within the site as far as my maths allows me to understand it) enough to reassure me - beyond any frustrating experiences/strange observations of my own - that the numbers are truly random and our dice are truly random.
Their FAQ question 2.1 is kinda relevant
And 2.2 is interesting given how our own FAQ explains how CC uses the random.org numbers.

Cicero