So...If the dice are Truly Random

[Walthobum]

Since the introduction of the dice stats the dice in the games I've played have been crazy. Really, really crazy. I've lost count of the times I've seen armies with massive numerical advantages losing in the last week or 2 weeks. When I say massive, I mean it's commonplace for me to see battles of 20 v 4 going in favour of the 4. Not all against me or anything. Not claiming it's a conspiracy. It just makes me mistrustful of the dice.

Basically if someone has been playing with the dice algorithm I can report that I don't think it's been a success. Change it back.

[White Moose]

Nothing with the dice has changed.

That CC now keep statistics over the dice results have nothing to do with the accual dice.

[Walthobum]

something has gone badly wrong recently. or i'm witnessing some statistical anomaly.

[rdsrds2120]

The way the dice are retrieved has not been changed. The statistics are created after the dice are retrieved in the same ole' fasion they used to be. They're fine, you may have just had some streaks and are looking for something to blame it on.

-rd

[shieldgenerator7]

Like I said, neutral player controls the dice, and it seems he still does. He likes long streaks, for odd strange reason.

[Walthobum]

no. like i said i have seen them go for me and against me in equal measure. not claiming a conspiracy at all. i also know a bit about statistics and chance and the way there can be 'runs' of unexpected results which is why i waited a week or so to say anything. it must be that i'm in one of those runs. it should be pointed out that several of the players i have been playing with also appeared to be in the middle of similar runs.

[Walthobum]

and it just keeps happening. 16 v 6? 6 wins. 22 v 8? 8 wins. 13 v 4? 4 wins. all in recent games. just crazy, crazy dice. every time a big army attacks a much smaller army in my recent games, they lose. every time.

[Leehar]

auto-assaults?

[Army of GOD]

Since the introduction of the dice stats the dice in the games I've played have been crazy. Really, really crazy. I've lost count of the times I've seen armies with massive numerical advantages losing in the last week or 2 weeks. When I say massive, I mean it's commonplace for me to see battles of 20 v 4 going in favour of the 4. Not all against me or anything. Not claiming it's a conspiracy. It just makes me mistrustful of the dice.

Basically if someone has been playing with the dice algorithm I can report that I don't think it's been a success. Change it back.

Anecdotes are my preferred form of false argument.

[Walthobum]

army of god? massive

[GeneralAnestetic]

conquer club are so purposely ignorant to their dice failings.
the original poster has it right.


just now, in a great lakes game, a stack of 41 vs 27 spread out like so:
4,8,1,1,1,1,1,1,7,1,1


i rolled this same set of combat TEN times with physical dice, and this how many armies the attacker had standing after winning each time:

27,26,20,14,31,21,19,26,20,23


what happened in CC? the attacker didn't even kill the defender. the defender ended with 3 armies spread 1,1,1.

not only is CC inaccurate, it's off by -17 from the worst outcome, and by -25.7 from the average outcome, and -34 from the best outcome.

i invite anyone to take their physical dice at home and roll these attacks and see how many they have left over each time after ALWAYS WINNING.

not on CC's messed up system though.

people are quitting because it's so unrealistic & well they should.

[maasman]

conquer club are so purposely ignorant to their dice failings.
the original poster has it right.


just now, in a great lakes game, a stack of 41 vs 27 spread out like so:
4,8,1,1,1,1,1,1,7,1,1


i rolled this same set of combat TEN times with physical dice, and this how many armies the attacker had standing after winning each time:

27,26,20,14,31,21,19,26,20,23


what happened in CC? the attacker didn't even kill the defender. the defender ended with 3 armies spread 1,1,1.

not only is CC inaccurate, it's off by -17 from the worst outcome, and by -25.7 from the average outcome, and -34 from the best outcome.

i invite anyone to take their physical dice at home and roll these attacks and see how many they have left over each time after ALWAYS WINNING.

not on CC's messed up system though.

people are quitting because it's so unrealistic & well they should.

10 times? Come back when you have done it 10,000, then we'll talk.

[GeneralAnestetic]

conquer club are so purposely ignorant to their dice failings.
the original poster has it right.


just now, in a great lakes game, a stack of 41 vs 27 spread out like so:
4,8,1,1,1,1,1,1,7,1,1


i rolled this same set of combat TEN times with physical dice, and this how many armies the attacker had standing after winning each time:

27,26,20,14,31,21,19,26,20,23


what happened in CC? the attacker didn't even kill the defender. the defender ended with 3 armies spread 1,1,1.

not only is CC inaccurate, it's off by -17 from the worst outcome, and by -25.7 from the average outcome, and -34 from the best outcome.

i invite anyone to take their physical dice at home and roll these attacks and see how many they have left over each time after ALWAYS WINNING.

not on CC's messed up system though.

people are quitting because it's so unrealistic & well they should.

10 times? Come back when you have done it 10,000, then we'll talk.

i don't have to, fool. it's called extrapolation.

CC provides results that only happen once out of 10,000 times, half the time.
they aren't even REMOTELY close to typical.
not REMOTELY.
but have fun playing with a shrinking userbase, lame-o apologists.

[maasman]

Extrapolation does not work very well in statistics or probabilities. You can't say "I have rolled this die 10 times and have not gotten a 6, therefore I will never get a 6/this die is broken, etc." It just doesn't make any sense to do that.

Looking at your dice stats, everything checks out as far as luck is concerned, so you really have no argument besides a select few instances amongst thousands of others that your dice are somehow worse than anyone elses.

[GeneralAnestetic]

Extrapolation does not work very well in statistics or probabilities. You can't say "I have rolled this die 10 times and have not gotten a 6, therefore I will never get a 6/this die is broken, etc." It just doesn't make any sense to do that.

Looking at your dice stats, everything checks out as far as luck is concerned, so you really have no argument besides a select few instances amongst thousands of others that your dice are somehow worse than anyone elses.

you're completely ignorant and not worth replying to except to clarify: you can't even comprehend what other people write. i never said anything about 'luck' or 'my dice'. they weren't even my rolls in the great lakes game. the dice suck site-wide b/c the algorithm used by CC blows.

[shieldgenerator7]

dice algorithms are a weird thing. They actually use mathematical reasoning and stuff to come up with a "random" answer, which isn't completely random, but to us it is. I remember I was getting a list of random numbers using my TI-84 calculator, and a lot of numbers appeared more than once, and some numbers never appeared at all. Same with Windows Media Player. I used to have it set to "Play Random Song" but got frustrated because it would play the same songs over and over again while ignoring others. So I now shuffle the list each time and play it chronological, at least this way its "random" and I get through all the songs (300+! yay!). So yeah computer algorithms are wacky in this sense when creating random numbers, but I think random.org uses something else or other (?).

[natty dread]

There is no algorithm.

[Walthobum]

auto-assaults?

does it matter? if it does matter, then that surely proves the dice are not random. i have one game where i am considering attacking 126 v 18 to prove a point. though that isn't really fair to the other people playing on that game.

[jammyjames]

The thing with the dice stats though is that they only show the average roll you have achieved. What is there to say that i roll a 5 but my opponent rolls a 6..

My dice stats show that i am actually lucky and i'm getting really high rolls, whereas in reality i'm still losing every battle?

Obviously this doesn't affect me i was just thinking about it the other day

[natty dread]

The thing with the dice stats though is that they only show the average roll you have achieved.

Click image to enlarge.

[SirSebstar]

About the dice

19. How does the dice work?

We have a series of 50,000 true random numbers from Random.org. Each time the game engine processes an assault or auto-assault, it select a random spot in the series to read from using a pseudo-random computer function. Each time the game engine rolls the dice, the next number is read in sequence from the series (e.g. in a 3v1 attack 4 numbers are read sequentially). The series of 50,000 true random numbers from Random.org is replaced every hour. As of June 2010 we process 1,000,000 assaults per day.
20. Seriously, how could I possibly lose X troops while assaulting a region that only had Y? Are the dice broken?

The dice on this site are randomized by Random.org and by being an active member of Conquer Club you experience a very large number of dice rolls so it is natural to occasionally get your ass kicked. It is human nature to remember the losses more than the wins.
21. Why is the amount of data on my dice stats more/less than I expect?

First of all, we only started recording dice stats as of May 2011 so battles before that date are not included in the data. We only store the last 100 battles per player over all opponents as "recent" stats, so if you filter recent stats by opponent over 25 battles there there might not be 25 "recent" battles with that specific opponent and less data will appear than expected. Also, auto-assaults are considered a single battle so recent stats may contain more data than expected if there were auto-assaults.
22. How is luck calculated on my dice stats?

For dice distribution, luck is the actual average roll over the expected average roll (i.e. 3.5). Luck can range from -100% for an average roll of 1.00 to +100% for an average roll of 6.00.

For battle outcomes, luck is based on the ratio of kills over losses. For each type of battle (e.g. 3 vs 2, 1 vs 1, etc.) we know the expected ratio of kills over losses for both assault and defend based on the table below.
3 vs 2 3 vs 1 2 vs 2 2 vs 1 1 vs 2 1 vs 1
Assault 54.0% 66.0% 39.0% 57.9% 25.5% 41.7%
Defend 46.0% 34.0% 61.0% 42.1% 74.5% 58.3%

Luck is how many percentage points the actual ratio is above or below the expected ratio. To calculate total luck (i.e. assault + defend) we take a weighted average of ratios based on the amounts of assult battles and defend battles.

Red and green are used to emphasize negative and positive luck, respectively.

[Dako]

When you attack 40 vs 10 it is not 40 vs 10 in dice terms. It is 3v2, 3v2, 3v2, 3v2 and so on again and again. Each time you have like 45% to loose both. Here is your streakyness. Loosing in 45% of the time 10 times in a row is an easy thing (and it will result in -20 armies, yikes).

[AndyDufresne]

When you attack 40 vs 10 it is not 40 vs 10 in dice terms. It is 3v2, 3v2, 3v2, 3v2 and so on again and again. Each time you have like 45% to loose both. Here is your streakyness. Loosing in 45% of the time 10 times in a row is an easy thing (and it will result in -20 armies, yikes).

Right, I think this sort of breakdown is sometimes forgotten.


--Andy

[Fruitcake]

When you attack 40 vs 10 it is not 40 vs 10 in dice terms. It is 3v2, 3v2, 3v2, 3v2 and so on again and again. Each time you have like 45% to loose both. Here is your streakyness. Loosing in 45% of the time 10 times in a row is an easy thing (and it will result in -20 armies, yikes).

Right, I think this sort of breakdown is sometimes forgotten.


--Andy

Is losing in 45% of the time 10 times in a row an easy thing? The argument has always been that this happens with a regularity one would not usually expect.

[osujacket]

On that example, statistically speaking from the gambit odds calculator the defender is not defeated 9% of the time so it is within reason.

It can happen and it will happen 9 out of 100 times....

http://gamesbyemail.com/Games/Gambit/BattleOdds?