Conquer Club

have they changed the attackers dice/odds somehow? Myself, and everybody Im playing seems to be having terrible attackers dice today since the update.......whats up? Is this some horrible cosmic co-inkydink?

glide wrote:have they changed the attackers dice/odds somehow? Myself, and everybody Im playing seems to be having terrible attackers dice today since the update.......whats up? Is this some horrible cosmic co-inkydink?

its because they gave me alllllll the good dice yesterday.... hahahah !

My odds are fine still - nothing to get too raging about

I agree....the dice seem to go on horrible tangents against the attacker and the they go back to normal (or-in rare cases-strongly for the attacker) either way...its frustrating to go 12 vs. 3 and auto attack and have 3 or 4 vs. 3 and wonder what the hell happened!!


I second the revision of the dice program

Nope, the dice still come from random.org. If you don't think they're random, go complain to them.

ok.....thanx iceman.........just curious......no complaints.....except for my luck......

this may be a little old but i fidn the dice are really bad for me. i deployed 9 armies onto a 2 army territory and attacked a 3 army fortified territoy, in the end i was down to 3 and it was down to 2. I MEAN OMFG!!!!, is that the worst luck ever or what?!

Everyone complains about the dice. Sometimes you just have one of those days where you have really bad luck. I think people like to complain about having bad luck because its something that they can't control.

I dont think that the dice are from Random.org. i just sat there for 5 minutes pressing roll on a 5 dice generator. And not once did the last 2 dice roll double 6's more than once in a row.

JMody wrote:I dont think that the dice are from Random.org. i just sat there for 5 minutes pressing roll on a 5 dice generator. And not once did the last 2 dice roll double 6's more than once in a row.

i just had a 10 vs 3 where twice in a row all 5 dice were ones...........that means i lost 4 but still its odd to get 10 1's on the trot.

I agree. It don't make sense to get such odd dice behavior if the dice really are from random.org

For what it's worth, according to the Dice Analyzer, my attacking luck has improved ever so slightly of late.

I still have barely worse than average luck attacking 3 v 2 and barely better than average luck attacking 3 v 1. As for the rest, I don't make those attacks often enough for the stats to mean anything. Well, except the rather strange stat that I finally won my first 1 v 1 in the 10 tries I've had since I downloaded the tracker.

Hahahaha, Congrats.

I would like to see a dice recorder. I am sure it could be done by stocksr with a nice little script.

all of my stats are worse than they should be, by about 1% for most, considering thats with nearly 30,000 dice its kinda depressing, apart from my 1 vs 1 rolls, i have won 114 times and lost 191 times (i know! wat the hell am i doing attacking 1 v 1 so often?!?!?!?) which puts me about 4% better than i should be

alex_white101 wrote:

JMody wrote:I dont think that the dice are from Random.org. i just sat there for 5 minutes pressing roll on a 5 dice generator. And not once did the last 2 dice roll double 6's more than once in a row.

i just had a 10 vs 3 where twice in a row all 5 dice were ones...........that means i lost 4 but still its odd to get 10 1's on the trot.

The probability of getting 10 1's in a row is:

(1/6)^10 = 0.00000001653817168792

which you could expect to happen randomly on average once every

1/0.00000001653817168792 = 60,466,176 rolls of the dice.

By the way the probability of getting any 10 particular numbers in a row is also:

(1/6)^10 = 0.00000001653817168792

In this case, they all just happened to be the same number, namely 1.

Given the number of dice thrown at CC daily:

146 pages of active games times
100 games per page times
4 turns per game per day times
10 dice thrown per turn =
584000 rolls per day

what happened to you would be expected to happen to someone at CC on average every

60,466,176 / 584,000 = 104 days.

So it is indeed a rare occurrence, but not unheard of.

JMody wrote:Hahahaha, Congrats.

I would like to see a dice recorder. I am sure it could be done by stocksr with a nice little script.

A dice recorder would be neat. Unfortunately the dice results are not currently recorded in the game log which is where the greasemonkey scripts get almost all of their data from. Also, I believe it's correct to say that the greasemonkey scripts don't currently remember anything (other than configuration data), that is, they don't store any data, it's all calculated anew with every refresh event. Finally, the biggest obstacle is that the game only shows the results of the dice for the attacker's attacks. The defender's dice are shown during the attacker's turn, which makes keeping track of the dice more complicated. It could be done, but the greasemonkey script would become much more complicated than they currently are.

JMody wrote:Hahahaha, Congrats.

I would like to see a dice recorder. I am sure it could be done by stocksr with a nice little script.

It's been done all ready (and not by me)

http://www.conquerclub.com/forum/viewtopic.php?t=5655

Stocksr, I mean a script that actually remembers every roll. I.E.

Roll 1: 2 - 2 - 5 vs 6 - 6
Roll 2: 5 - 3 - 2 vs 5 - 5

Don't really need it to say where was being attacked just the roll per roll break down.

I apologise in advance for long post, but bear with me...

I imagine this had been asked several times, but I had a search and didn't turn up anything - could lack or anyone comment on how the calls to random.org are done? Are the values for the 5 dice of a roll completed with a single call, or 5 seperate calls?

Reason I ask is I've noticed a rough (and only partially founded) skew in my dice, where I get all 5 dice in a 3 vs 2 attack as the same. Ie, all sixes, all ones, etc.

By partially founded, I mean in my 130-something games, I know for fact I have rolled 5 sixes at *least* 10 times - by at *least* I mean 5 sixes is the one that sticks out in my mind the most, as it invariably results in a small tantrum to roll 3 sixes as an attacker, and still lose 2 armies. Rolling 5 ones also sticks out as unusual, and I would estimate at least 6 times, also being conservative.

So that is (conservatively) 16 events of a roll that has 1 in 7776 probability - not to mention rolls of 5 x 2,3,4,5 that I don't recall vividly. I certainly have not rolled 124,416 times in 130-something games.

SO THE REASON I ASK, is understanding the method used by random.org (atmospheric noise), I can also see how potentially it could generate similar results given 'approaching zero' interval between samples. While atmospheric noise seems random, it inherently becomes less random as you reduce sample interval towards zero. Scale this reasoning towards taking 2 samples at the same time, and you have the same input and hence same output. So I can imagine that several calls at small (near-zero) intervals would generate similar numbers.

HOW DOES THIS AFFECT CC? Due to the rules, and equal defenders dice wins, and perhaps a reason why we are getting a skew towards "bad dice" whingers over "good dice" braggers...?

Sorry for long post - I am generally pro the "user error, shut up and go away" argument, but I have a nasty dint in my dice statistics after some 4000 recorded rolls that shows otherwise.

Could any of the CC coders comment on the original question, as to how the calls to random.org are made...?

Sorry I should also say, that if my theory does in fact hold water (unlikely, but am interested in comment) then I do not know the solution.

The sites performing randomness by entropy of radioactive decay would probably fix it, but I do not know of one with an open HTTP call...?

if tahitiwahini is right and the probability of getting 10 same dice in a row is 1 in 103 days then i guess something is really wrong with the dice cause it happened to me 3 times as far as i remember and 2 of those times were in a single turn.

i rolled 3 1's and the defender rolled 2 1s then same thing happened again.

various other dice happened then in the same turn i attacked another country and again it happened 10 1's in a row. don't call me crazy cause i have eye witnesses.

and one other time it happened to roll 10 6's in a row.


this is only from what i remember there may have been even more because i never look at auto attack rolls.

i'm not complaining here and i don't really care if the dice are changed or not but something is fishy around here.

also it happened to me when attacking single armies terits, that the defender rolls the same dice several times in a row. 3,4 and even 5 times in a row.
i don't know what's the chance f that but i bet it's slim.

It definitely does fluctuate, but I have been having the same good luck on the attack too. Yesterday I beat my friend in two 2 player games because of ridiculous attack runs and defensive wins in my favor.

You also must remember that the rules favor the defender. The rolls are random from a list, which as I understand it, is a list of all possible combinations. Since the rules favor the defender, the combination of higher defense die and even defense die makes more defender victories.

oran0007

the dice are from random.org and are supposed to be random.


while i don't question the fact that they are really random, i do have some doubts when i see strange dice. like the fact that sometimes you hit 10 same dice in a row. being a risk fan and actively playing it for almost 10 years means i have rolled thousands of dice using real dice and i'm telling you in all these years i've never seen such strange dice behaviour.

i have the dice analyzer installed. and i've seen it at others and i know it evens out at ~16.66 for each dice. at first i said it will even out at 10k throws. i'm at 15k and still i have large differences (2% is large)

currently the analyzer tells me the defender rolls 6s in 17.45% of the cases while i roll 1s in 17.2% of the cases. so i mainly throw 1s and the defender 6s. not to mention i'm under the ideal stats at all battle outcomes.
again this is not a problem as i'm used to having crappy dice. i have them in real life risk so why not on CC? i've learned to live with it. what disturbs me are the statistic anomalies. things that are supposed to happen really rarely but in fact they happen too damn often. like the defender rolling 5 straight 6s.

i'm sure that if every CC user had the dice analyzer and if we took the stats we would probably be very close to the ideal average but still the anomalies puzzle the heck out of me.

tahitiwahini wrote:

alex_white101 wrote:

JMody wrote:I dont think that the dice are from Random.org. i just sat there for 5 minutes pressing roll on a 5 dice generator. And not once did the last 2 dice roll double 6's more than once in a row.

i just had a 10 vs 3 where twice in a row all 5 dice were ones...........that means i lost 4 but still its odd to get 10 1's on the trot.

The probability of getting 10 1's in a row is:

(1/6)^10 = 0.00000001653817168792

which you could expect to happen randomly on average once every

1/0.00000001653817168792 = 60,466,176 rolls of the dice.

By the way the probability of getting any 10 particular numbers in a row is also:

(1/6)^10 = 0.00000001653817168792

In this case, they all just happened to be the same number, namely 1.

Given the number of dice thrown at CC daily:

146 pages of active games times
100 games per page times
4 turns per game per day times
10 dice thrown per turn =
584000 rolls per day

what happened to you would be expected to happen to someone at CC on average every

60,466,176 / 584,000 = 104 days.

So it is indeed a rare occurrence, but not unheard of.

^ great post
and that only takes into account people from cc calling random.org, if include all the calls made to it each day then it is possible far many things which seem skewed to occur to one person while the overall statistics are not skewed at all.