What if it's not the dice?

wcaclimbing · Post by **wcaclimbing** » Sun Jun 17, 2007 1:52 pm

maniacmath17 wrote:It also explains the "streakyness" of the dice. If someone out there has a run of great rolls, then whoever rolls after that person is rolling from a page with less good rolls left and is more likely to hit a streak of bad rolls.

There's got to be a better way.

but you have the same probability of being the guy with good dice as the guy with bad dice, so its still random.

Sgt. Drake · Post by **Sgt. Drake** » Sun Jun 17, 2007 3:02 pm

maniacmath17 wrote:
papadopo wrote:I think its the TIME...

sometimes i win every battle and sometimes i loose them all and i think it makes sense too if the following is true:

The way CC gets the random numbers is that they get a 'page' with all possible dice rolls in 'lines'. Each possible dice roll in each line'. The lines are placed on the page in a random order.
Then each dice roll request by CC uses the next available 'line'. When the 'page' is used up, another page is generated by Random.org and used by CC.
The thing is that when a specific dice roll has been used up from the page (say, three sixes) it will not come up again until the page is changed. This means that when the request is made by a game for a dice roll, you don't really have equal % for ALL possible dice rolls... you will never get the used up ones

Is the above true?

Wow, thanks for the info. Now I've actually got a strategy heading into attacks. If I lose 2 then I'll quickly roll again because there is 1 less bad outcome on the page. If I win 2 then that means there is 1 less good out come, so I'll wait for a while for a new page to generate in order to increase my chances of a good roll.

Thats actually what I try to do. If I get favorable dice, I pound them until the go against me, and I'll often find times where I'll take a bunch in a row. If they are unfavorable, I'll take my time with them, maybe go back to my games page or log out and back in. In a game I just took a turn in, I did the very same thing, was losing even with him for a little bit, then hit a streak and took 9 in a row by attacking as fast as I could over and over. Now then, there will be sometimes where no matter how much I wait, I will only get bad dice.

I wonder if that info is true, and if so how big are the pages and with what frequency are they refreshed.

nietzsche · Post by **nietzsche** » Sun Jun 17, 2007 10:41 pm

wow, be careful, wicked will ban you only for talking about the dice

maniacmath17 · Post by **maniacmath17** » Mon Jun 18, 2007 1:28 am

wcaclimbing wrote:
maniacmath17 wrote:It also explains the "streakyness" of the dice. If someone out there has a run of great rolls, then whoever rolls after that person is rolling from a page with less good rolls left and is more likely to hit a streak of bad rolls.

There's got to be a better way.

but you have the same probability of being the guy with good dice as the guy with bad dice, so its still random.

True... but think of it this way: Before each turn you are told ahead of time that you will either lose 100 straight troops or kill 100 straight troops. Now the chances of you getting the good streak is the same as the chances of the bad streak. Is this really what we would call random?

Obviously this is a bit more extreme than what is actually going on but the point is that while the numbers will even out eventually, it's important to not have streaky dice.

decoulombe · Post by **decoulombe** » Mon Jun 18, 2007 8:08 am

-=LRK=- wrote:No, it's not fair. Fair, as we expect to play, means win some/loose some. I just hit auto-attack and the single defender, against my 13+ army, rolled 6,6,6,6,5 before I finally got him. I was JUST complaining about those odds yesterday. If you go to Random.org directly, you will NOT get dice rolls like that. BTW- it was also a neutral owner, which I also mentioned is harder to kill.

I just went again, another map, and he rolls 5,5,6,6,6,6,4,6,6 all beating me, before I got him with, another 6. That was how many guys to take out a little blip of land? Once is pretty neat. Regularly and there are issues. Maybe it's not the issue that it randomly picked 6, 6 times out of 9 but that 8 were high numbers. As a defender, ties go in his favor.

I just went to random.org and picked various numbers of dice. Those numbers went up and down, some duplicates, but mostly changed. It doesn't seem that way here, even when it isn't going against me.

Are you saying another source is rolling the die?

Even if the dice are not in sync with actual die (IE. You roll more or less six's, playing like actual dice), the same will applies to player 2 and so forth. Supposing you said that wasn't fair (like trying to capture a neutral territory), and you proved that wasn't fair. Unless the die are purposely making you lose, and if player two tried to get the same territory, he would get the exact same chances. (not the same luck, but the same chance.) He might very well get that territory, but his chances would be the same. You can argue the die are acting like real die or even acting the die Random.org claims it would be, but the fairness would be the same in that everyone would have an equal chance to win or lose. If it's TIME, then everyone would have the same opportunity to play according to that time (I.E. someone said wait a couple seconds, etc.)

And besides, if LACK said the dice come from Random.org, thats good enough for me. Besides, I like playing the game. I don't care who wins or loses, just trying my best.

TipTop · Post by **TipTop** » Mon Jun 18, 2007 11:13 am

Guys, stop judging dice on short term results. Short term results prove nothing.

You cannot roll the dice 10 times and say this set of results proves anything.

You can only judge with long term results. The minimum long term result that is meaningful with which to judge the randomness of dice is 10,000 rolls!

I will try and explain why.

Take a two sided coin. Now we know there is a 50% chance it will land on heads and a 50% chance it will land on tails. As there can only be two possible outcomes to every flip we now this is a mathematical certainty, but only IN THE LONG RUN.

If we flip that coin once (the shortest of short term results) then one side will win 100% of the time the other will lose 100% of the time.

If we flip it 10 times it is still highly unlikely you will see a 50/50 split in results because there are more combinations of one side winning more than the other. It is still too short term.

Possible results: Heads/tails

10/0 9/1 8/2 7/3 6/4 5/5 4/6 3/7 2/8 1/9 0/10

If you look just at those possible results you will notice you are twice more likely to see 1 side winning 10 nil (10/0 0/10) as you are to see an exact 5/5 split! This means that it is a mathematical certainty (in the long run) you will see a 10 nil streak twice as often as you will see a 5/5 split!!!

If you roll the dice 100 times and think of all the possible combinations of results, the odds of hitting that magical 50% for both heads and tails is still huge. (99-1)

However the longer you roll, the closer to the 50% mark you will get as there will be far more combinations of results that bring you closer to this 50% mark than not.

Another thing to remember about random.

RANDOM HAS NO MEMORY

Random can not say to itself I have landed on tails 10 times in a row I best start landing on heads for a bit. What results came before has no bearing on what happens next. Daft as it sounds this is the main thing most people just can't get there heads around. Do you honestly think that if a coin landed on tails 10 times in a row that it would be more likely to land on heads on the next flip?

I hope this coin example helps you to understand more clearly how random works. It will be streaky in the short term, this is normal. Only in the long run will the stats you expect to see become reality. If those stats don't become reality in the long run (10,000 rolls minimum) then, and only then, can you claim with conviction that the dice aren't random.

decoulombe · Post by **decoulombe** » Mon Jun 18, 2007 1:26 pm

But even if you had played 10,000 or so times and realized the dice were faulty, the second player would have to play with the same dice. Therefore, going by the original question is the dice fair, they must be, not showing any favoritism between the first player and the rest.

maniacmath17 · Post by **maniacmath17** » Mon Jun 18, 2007 1:50 pm

TipTop wrote:Possible results: Heads/tails

10/0 9/1 8/2 7/3 6/4 5/5 4/6 3/7 2/8 1/9 0/10

If you look just at those possible results you will notice you are twice more likely to see 1 side winning 10 nil (10/0 0/10) as you are to see an exact 5/5 split! This means that it is a mathematical certainty (in the long run) you will see a 10 nil streak twice as often as you will see a 5/5 split!!!

If you roll the dice 100 times and think of all the possible combinations of results, the odds of hitting that magical 50% for both heads and tails is still huge. (99-1)

You are very wrong my friend. Each outcome has a different probability. Lets take a look at flipping 2 coins. With your reasoning the outcomes are 2/0 1/1 and 0/2 which would mean there is a 33% chance of having 1 heads and 1 tails.

The actual probability is 50% of having 1 heads and 1 tails because there's 4 different outcomes. HH, HT, TH, TT. That makes the chance of getting 1 of each 2/4 or 50%. With your reasoning, if we flipped a million coins, you'd be just as likely to roll all heads as you were to roll half heads, lol.

RobinJ · Post by **RobinJ** » Mon Jun 18, 2007 5:11 pm

TipTop wrote:RANDOM HAS NO MEMORY

I wish I would remember this more often when I gamble. I know only too well that random has no memory but when black spins up 10 times in a row on roulette, I will always bet red the next and still lose 50% of the time. (More than 50% actually - I'm sure it's fixed).

Nice wee gamble today I had too (not). But a bit of money on a pretty safe bet and then remembered I'd forgotten to stick a bit on zero too like always. Of course, a big dirty zero spins up! Life really is strange!Just a good job it wasn't too much money! :lol:

As for the matter in hand, I'm sure the dice are fair enough and I'm surprised that the mods haven't been in to correct this (or perhaps they're just watching on with amusement! :evil:

). Anyway, I believe my story has a meaning - life very often gives you what you least expect

TipTop · Post by **TipTop** » Mon Jun 18, 2007 8:18 pm

maniacmath17 yeh your right, I realised this myself after I posted Doh!

papadopo · Post by **papadopo** » Thu Jun 21, 2007 12:31 pm

All i am asking for is to have exaclty the same chance of getting ANY dice combination AT THE TIME I PRESS THE ATTACK BUTTON.

I dont care about long tern or short term. I want at the time i press the ATTACK buton to know that i have EXACTLY the same chance of getting all possible combinations.

Do i have that?

Molacole · Post by **Molacole** » Thu Jun 21, 2007 1:26 pm

well I am glad it's not only me who seems to think territories might have something to do with luck/dice fator!

I can honestly say that I've been in many games were it just seemed impossible to take a certain territory. After conquering that territory I have also noticed others failed trying to get it back at great costs to the both of us. Yeah it could definitely be the dice playing games on us, but it's happened a few too many times.

I actually use that as part of my strategy now. If I find a spot that is difficult to conquer I tend to go back to it and fortify it lightly in hopes others will challenge it with the same luck I had.

One territory that had given me tons of problems to the extent I shy away from it is columbia on the world map and I always leave it lightly fortified if I get the bonus. I've had great luck defending it thus far in most the games I have played!

It's more than likely just the dice, but it definitely makes you think twice about attacking certain spots that have given you trouble on many occassions!

driskmaster · Post by **driskmaster** » Thu Jun 21, 2007 4:30 pm

I agree moleacue. Dont you ALL find it really tough trying to beat neutrels. I now i never attack neutrels coz its not the first time i have lost 10-15 men to them!!!

n8freeman · Post by **n8freeman** » Thu Jun 21, 2007 4:47 pm

how likely is it to be attacked and lose 42 to 24 me being attacked and i had the 42)

i mean come on
thats pretty insane

kha1i1 · Post by **kha1i1** » Thu Jun 21, 2007 5:26 pm

tiptop was on the right track. randomness cannot be measured in short term i.e. omg he rolled a 6,6,6,4,5,6!!! that can absolutely be random. random means that the numbers before don't matter so that it's impossible for the dice to 'know' that the player in question has rolled 6's lots of times or to be biased one way or the other. that being said, i have noticed an seemingly 'house bias' on the neutral territories, but that can be related to the fact that i would be apt to remember unlikely losses than normal wins.

p.s. as a test of true randomness, get a friend, 2 dice and throw them over and over again. record the results of each throw. u will notice that you can spot patterns (imaginary of course) even though you KNOW that there cant be any bias one way or the other. streakiness is a part of the game.

p.p.s i just went to random.org to see this dice flipper thing and rolled 2 dice over and over again (maybe 20 times). in that time , there was a period where it rolled (D1,D2): (4,1),(3,1),(2,1),(2,1),(2,1). there you have it. random at work.

zarvinny · Post by **zarvinny** » Thu Jun 21, 2007 8:31 pm

In the long run, the dice are "random" or at least they balance out. However, it is my opinion that they are More Streaky than real life dice. For instance, in the span of 2 games I lost when I attacked 6 on 1. completely lost. Chance of that, less than 1 in 100, yet in happened twice. Things like that happen, I also killed 1 army and lost 7 when attacking 10 on 3.

Speaking of RANDOM STUPID LUCK, today in the span of 2 FLAT RATE GAMES, I got 8 dam red cards in a row. then i joined another one and got 3 more out of the first 4. Complete stupidity.

Sgt. Drake · Post by **Sgt. Drake** » Fri Jun 22, 2007 1:02 am

zarvinny wrote:

In the long run, the dice are "random" or at least they balance out. However, it is my opinion that they are More Streaky than real life dice. For instance, in the span of 2 games I lost when I attacked 6 on 1. completely lost. Chance of that, less than 1 in 100, yet in happened twice. Things like that happen, I also killed 1 army and lost 7 when attacking 10 on 3.

Speaking of RANDOM STUPID LUCK, today in the span of 2 FLAT RATE GAMES, I got 8 dam red cards in a row. then i joined another one and got 3 more out of the first 4. Complete stupidity.

Thats exactly my opinion on the dice. In the long run they are "random" but in the short term they are very streaky. I had wondered if it might have been because of the "atmosphere" that generated the random numbers, as the random.org site uses as their generator. Maybe there were times where the atmosphere would throw out similar dice or something. However some things brought up in this thread also raise questions.

RobinJ · Post by **RobinJ** » Fri Jun 22, 2007 3:26 am

n8freeman wrote:how likely is it to be attacked and lose 42 to 24 me being attacked and i had the 42)

i mean come on
thats pretty insane

Random means that anything can happen remember. Besides, winning 24v42 isn't all that big - a few lucky rolls and you're landed.

TipTop · Post by **TipTop** » Fri Jun 22, 2007 11:32 am

a little over 6% of the time it will happen

papadopo · Post by **papadopo** » Mon Jun 25, 2007 5:28 am

papadopo wrote:All i am asking for is to have exaclty the same chance of getting ANY dice combination AT THE TIME I PRESS THE ATTACK BUTTON.

I dont care about long tern or short term. I want at the time i press the ATTACK buton to know that i have EXACTLY the same chance of getting all possible combinations.

Do i have that?

well... do I?

RobinJ · Post by **RobinJ** » Mon Jun 25, 2007 8:59 am

zarvinny wrote:

In the long run, the dice are "random" or at least they balance out. However, it is my opinion that they are More Streaky than real life dice. For instance, in the span of 2 games I lost when I attacked 6 on 1. completely lost. Chance of that, less than 1 in 100, yet in happened twice. Things like that happen, I also killed 1 army and lost 7 when attacking 10 on 3.

Speaking of RANDOM STUPID LUCK, today in the span of 2 FLAT RATE GAMES, I got 8 dam red cards in a row. then i joined another one and got 3 more out of the first 4. Complete stupidity.

Those are nice stats! You roll a lot of 6s and your opponent rolls relatively few. Me, both me and my opponent seem to roll a lot of 4s.

jaseleo · Post by **jaseleo** » Tue Jun 26, 2007 6:52 am

If the dice are random how is everybodys stats either the same or around the same??? i havent seen anyone with a massive spike on dice analyser statistics surely if this was this was the case we would see some people have a variance of 20% to 100% in different stats???????

jaseleo · Post by **jaseleo** » Wed Jun 27, 2007 7:20 am

Can anyone tell me why there is no difference??

gingis khan · Post by **gingis khan** » Wed Jun 27, 2007 11:01 am

papadopo wrote:you think we are related GreecePwns?

lol, I dont have any relatives from Greece... I have some that are Olimpiakos fans though, all Cypriot...

cyprus is greak land!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! .....stolen by turks in a most shamefull manner (blitzkrig)

papadopo · Post by **papadopo** » Thu Jun 28, 2007 3:22 am

gingis khan wrote:
papadopo wrote:you think we are related GreecePwns?

lol, I dont have any relatives from Greece... I have some that are Olimpiakos fans though, all Cypriot...

cyprus is greak land!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! .....stolen by turks in a most shamefull manner (blitzkrig)

Half of the island is occupied by turkey... They keep massive armies here but i dont worry: We WILL get our land back.