Dice?

[KraphtOne]

I know there is the standard thread of idiots complaining...

but on the real though...

You ARE aware there are better algorithms out there for dice right?

sure over 20,000 games my dice will even out... but it is borderline retarded for you to tell me that i can possibly lose 23 vs 1...

hell we lose 15s+ vs 2s or 3s all the time (statistically impossible) and just chalk it up to bad luck... but seriously that should NEVER happen, and if it does it should be posted in forums on how insane it was... But we say "eh it happens, man i lose 15vs2 next game the other guy does the same thing"... two wrongs does not make it right... just makes it wrong both times...

c'mon boys let's fix this thing finally... hell my first game on here my buddy says to me he says "good site but the dice are crazy streaky"

yeah over 20,000 games my dice stats are even... but what that says to ME is that, i got hosed on a 23 vs 1 (statistically impossible) which means someone playing me will somehow lose 23 vs 1 (statistically impossible)...

equal bullshit does not mean the dice program works my friends

[KraphtOne]

Also just now when i looked to see the exact verb usage the guy used 7 years ago to tell me this... it didn't allow me to get my whole game history...

[Razorvich]

1 troop is a pushover...2 troops can take out an army

[rhp 1]

I know there is the standard thread of idiots complaining...

but on the real though...

You ARE aware there are better algorithms out there for dice right?

sure over 20,000 games my dice will even out... but it is borderline retarded for you to tell me that i can possibly lose 23 vs 1...

hell we lose 15s+ vs 2s or 3s all the time (statistically impossible) and just chalk it up to bad luck... but seriously that should NEVER happen, and if it does it should be posted in forums on how insane it was... But we say "eh it happens, man i lose 15vs2 next game the other guy does the same thing"... two wrongs does not make it right... just makes it wrong both times...

c'mon boys let's fix this thing finally... hell my first game on here my buddy says to me he says "good site but the dice are crazy streaky"

yeah over 20,000 games my dice stats are even... but what that says to ME is that, i got hosed on a 23 vs 1 (statistically impossible) which means someone playing me will somehow lose 23 vs 1 (statistically impossible)...

equal bullshit does not mean the dice program works my friends

anyone who knows me knows I don't bitch about dice other than the occasional "well, that was fucked up"... that said? Krapht has a point here... just because the dice program is fucked up for everyone, doesn't make it a good thing.. if there's something better (which Krapht would know better than I) get the damn thing...

[Agent 86]

I know there is the standard thread of idiots complaining...

but on the real though...

You ARE aware there are better algorithms out there for dice right?

sure over 20,000 games my dice will even out... but it is borderline retarded for you to tell me that i can possibly lose 23 vs 1...

hell we lose 15s+ vs 2s or 3s all the time (statistically impossible) and just chalk it up to bad luck... but seriously that should NEVER happen, and if it does it should be posted in forums on how insane it was... But we say "eh it happens, man i lose 15vs2 next game the other guy does the same thing"... two wrongs does not make it right... just makes it wrong both times...

c'mon boys let's fix this thing finally... hell my first game on here my buddy says to me he says "good site but the dice are crazy streaky"

yeah over 20,000 games my dice stats are even... but what that says to ME is that, i got hosed on a 23 vs 1 (statistically impossible) which means someone playing me will somehow lose 23 vs 1 (statistically impossible)...

equal bullshit does not mean the dice program works my friends

anyone who knows me knows I don't bitch about dice other than the occasional "well, that was fucked up"... that said? Krapht has a point here... just because the dice program is fucked up for everyone, doesn't make it a good thing.. if there's something better (which Krapht would know better than I) get the damn thing...

+1 Atmospheric noise from random.org is the biggest joke

[loutil]

I know there is the standard thread of idiots complaining...

but on the real though...

You ARE aware there are better algorithms out there for dice right?

sure over 20,000 games my dice will even out... but it is borderline retarded for you to tell me that i can possibly lose 23 vs 1...

hell we lose 15s+ vs 2s or 3s all the time (statistically impossible) and just chalk it up to bad luck... but seriously that should NEVER happen, and if it does it should be posted in forums on how insane it was... But we say "eh it happens, man i lose 15vs2 next game the other guy does the same thing"... two wrongs does not make it right... just makes it wrong both times...

c'mon boys let's fix this thing finally... hell my first game on here my buddy says to me he says "good site but the dice are crazy streaky"

yeah over 20,000 games my dice stats are even... but what that says to ME is that, i got hosed on a 23 vs 1 (statistically impossible) which means someone playing me will somehow lose 23 vs 1 (statistically impossible)...

equal bullshit does not mean the dice program works my friends

The dice clearly have some issues on this site. However, 23 v 1 is not "statistically impossible". It is Statistically unlikely but in a large enough sample it will eventually happen. If you flip a coin 1 million times you will likely get a few streaks of 23 heads or tails in a row.

[Dukasaur]

sure over 20,000 games my dice will even out... but it is borderline retarded for you to tell me that i can possibly lose 23 vs 1...

Analysis of 23 v 1. Okay, so for your first 20 roll you were rolling 3 dice to 1, then the 21st roll you rolled 2 dice to 1, and then on your last roll you rolled 1 dice to 1.
Probability of losing any of the first 20 = 0.3403
Probability of losing on the 21st roll = 0.4213
Probability of losing on the last roll actually goes to the defender = 0.5833

Okay, so your aggregate probability of losing that is 0.3403^20 * 0.4213 * 0.5833, and taking the inverse of that to express it as odds we get 1 in 9,381,614,482

1 in 9 billion and change. Pretty rare result, I'll grant you that. Not "statistically impossible" obviously, but pretty rare. Still, you've rolled the dice two and a half million times just since they started keeping track, and I-don't-know how many times before that, so if anyone was to come up against some of the freakier results, I would expect it to be you.

hell we lose 15s+ vs 2s or 3s all the time (statistically impossible) and just chalk it up to bad luck... but seriously that should NEVER happen, and if it does it should be posted in forums on how insane it was... B

Unlike your 23v1 example, 15v2 is not nearly as rare.

The math is quite complex, but luckily 15v2 is common enough that it appears on Risk website tables, like this one:
http://www.datagenetics.com/blog/november22011/
2 defenders, 15 attackers, the attacker wins 99.975% of the time. That means 3999 times out of 4000, 15v2 will win. One time out of 4000, the defender will win. Again, you're a guy who's rolled the dice two and a half million times. How many of those were 15v2? Impossible to say, but I'm guessing there's at least ten thousand or so instances of that, so I would expect you to have been in that situation a few times at least.

15v3 is, of course, even less unusual. The attacker will win 99.867% of the time, so the defender will win 0.133% of the time, which is once in every 752 instances. Not common, but not extraordinarily rare, either. Again, for a guy who's rolled the dice two and a half million times...

[KraphtOne]

3 dice vs 1 die is not the same thing as flipping a coin my friend...

to give you an idea the odds of you rolling 6/6 three times in a row are 1 in 45,000...

the probability of you losing 23 vs 1 is 0.9999999998935425

see how far over that little 8 is...

that's a 99.99% chance of winning (that 9 keeps going)... if you convert that to odds it is 0:1 odds... statistically impossible... ( i understand that you will say that since it isn't 100% then that means there is a chance, but that would mean you're a troll )

I am curious (i've done zero research) if we use the same random dice generator that hasbro used for its CD version... because it got complaints and was subsequently changed in later issues of the game...

and for Duky, what are the odds of losing 4vs1 7 times in a row... or losing 10 vs 2 twice in one game...

math is fun

my point is that yes these things happen, but HERE these things happen in STREAKS... your dice are never "good" your opponents' are just "randomly" bad... the fact that i lose 10vs2 or several 4vs1 or 6vs1 battles isn't crazy... but that it happens to at least one person in EVERY game where they have statistically terrible dice... makes no sense

[degaston]

sure over 20,000 games my dice will even out... but it is borderline retarded for you to tell me that i can possibly lose 23 vs 1...

Analysis of 23 v 1. Okay, so for your first 20 roll you were rolling 3 dice to 1, then the 21st roll you rolled 2 dice to 1, and then on your last roll you rolled 1 dice to 1...

KraphtOne, was the situation as Dukasaur described, or did you both have larger stacks, so you made 12 attacks of 3v2 dice, and over that series you lost 23 and the defender lost one?

[jdw35]

So what are our other options for dice programs? Because I agree, the one we use now is extremely streaky, and I know that everyone has had their fair share of games where they lose several 90%+ rolls, and then the next game its their opponents turn to have the short end of the stick.

[demonfork]

I know there is the standard thread of idiots complaining...

but on the real though...

You ARE aware there are better algorithms out there for dice right?

sure over 20,000 games my dice will even out... but it is borderline retarded for you to tell me that i can possibly lose 23 vs 1...

hell we lose 15s+ vs 2s or 3s all the time (statistically impossible) and just chalk it up to bad luck... but seriously that should NEVER happen, and if it does it should be posted in forums on how insane it was... But we say "eh it happens, man i lose 15vs2 next game the other guy does the same thing"... two wrongs does not make it right... just makes it wrong both times...

c'mon boys let's fix this thing finally... hell my first game on here my buddy says to me he says "good site but the dice are crazy streaky"

yeah over 20,000 games my dice stats are even... but what that says to ME is that, i got hosed on a 23 vs 1 (statistically impossible) which means someone playing me will somehow lose 23 vs 1 (statistically impossible)...

equal bullshit does not mean the dice program works my friends

The dice clearly have some issues on this site. However, 23 v 1 is not "statistically impossible". It is Statistically unlikely but in a large enough sample it will eventually happen. If you flip a coin 1 million times you will likely get a few streaks of 23 heads or tails in a row.

I hate to break it to you but you're wrong! Losing 23v1 is statistically impossible.

[Dukasaur]

I know there is the standard thread of idiots complaining...

but on the real though...

You ARE aware there are better algorithms out there for dice right?

sure over 20,000 games my dice will even out... but it is borderline retarded for you to tell me that i can possibly lose 23 vs 1...

hell we lose 15s+ vs 2s or 3s all the time (statistically impossible) and just chalk it up to bad luck... but seriously that should NEVER happen, and if it does it should be posted in forums on how insane it was... But we say "eh it happens, man i lose 15vs2 next game the other guy does the same thing"... two wrongs does not make it right... just makes it wrong both times...

c'mon boys let's fix this thing finally... hell my first game on here my buddy says to me he says "good site but the dice are crazy streaky"

yeah over 20,000 games my dice stats are even... but what that says to ME is that, i got hosed on a 23 vs 1 (statistically impossible) which means someone playing me will somehow lose 23 vs 1 (statistically impossible)...

equal bullshit does not mean the dice program works my friends

The dice clearly have some issues on this site. However, 23 v 1 is not "statistically impossible". It is Statistically unlikely but in a large enough sample it will eventually happen. If you flip a coin 1 million times you will likely get a few streaks of 23 heads or tails in a row.

I hate to break it to you but you're wrong! Losing 23v1 is statistically impossible.

Nothing is statistically impossible. Rare, but certainly not impossible. I demonstrated the math up above.

Analysis of 23 v 1. Okay, so for your first 20 roll you were rolling 3 dice to 1, then the 21st roll you rolled 2 dice to 1, and then on your last roll you rolled 1 dice to 1.
Probability of losing any of the first 20 = 0.3403
Probability of losing on the 21st roll = 0.4213
Probability of losing on the last roll actually goes to the defender = 0.5833

Okay, so your aggregate probability of losing that is 0.3403^20 * 0.4213 * 0.5833, and taking the inverse of that to express it as odds we get 1 in 9,381,614,482

1 in 9 billion and change. Pretty rare result, I'll grant you that. Not "statistically impossible" obviously, but pretty rare.

[Gweeedo]

Has the attacking player ever won a 1v1 in this game?

[Trevor33]

I know there is the standard thread of idiots complaining...

but on the real though...

You ARE aware there are better algorithms out there for dice right?

sure over 20,000 games my dice will even out... but it is borderline retarded for you to tell me that i can possibly lose 23 vs 1...

hell we lose 15s+ vs 2s or 3s all the time (statistically impossible) and just chalk it up to bad luck... but seriously that should NEVER happen, and if it does it should be posted in forums on how insane it was... But we say "eh it happens, man i lose 15vs2 next game the other guy does the same thing"... two wrongs does not make it right... just makes it wrong both times...

c'mon boys let's fix this thing finally... hell my first game on here my buddy says to me he says "good site but the dice are crazy streaky"

yeah over 20,000 games my dice stats are even... but what that says to ME is that, i got hosed on a 23 vs 1 (statistically impossible) which means someone playing me will somehow lose 23 vs 1 (statistically impossible)...

equal bullshit does not mean the dice program works my friends

The dice clearly have some issues on this site. However, 23 v 1 is not "statistically impossible". It is Statistically unlikely but in a large enough sample it will eventually happen. If you flip a coin 1 million times you will likely get a few streaks of 23 heads or tails in a row.

You can't compare flipping a 50/50 coin with rolling 3 dice against 1. You could roll 3 dice against 1 a billion times and not get 23 wins for the 1 in a row. Do you realize the math involved for that to happen?

23 50/50 cion flips i a row is almost 8.4 million to one. Meaning statistically you would have to flip the coin 8.4 million times before getting 23 in a row. Convert that into 3vs1 dice... impossible.

[loutil]

I know there is the standard thread of idiots complaining...

but on the real though...

You ARE aware there are better algorithms out there for dice right?

sure over 20,000 games my dice will even out... but it is borderline retarded for you to tell me that i can possibly lose 23 vs 1...

hell we lose 15s+ vs 2s or 3s all the time (statistically impossible) and just chalk it up to bad luck... but seriously that should NEVER happen, and if it does it should be posted in forums on how insane it was... But we say "eh it happens, man i lose 15vs2 next game the other guy does the same thing"... two wrongs does not make it right... just makes it wrong both times...

c'mon boys let's fix this thing finally... hell my first game on here my buddy says to me he says "good site but the dice are crazy streaky"

yeah over 20,000 games my dice stats are even... but what that says to ME is that, i got hosed on a 23 vs 1 (statistically impossible) which means someone playing me will somehow lose 23 vs 1 (statistically impossible)...

equal bullshit does not mean the dice program works my friends

The dice clearly have some issues on this site. However, 23 v 1 is not "statistically impossible". It is Statistically unlikely but in a large enough sample it will eventually happen. If you flip a coin 1 million times you will likely get a few streaks of 23 heads or tails in a row.

You can't compare flipping a 50/50 coin with rolling 3 dice against 1. You could roll 3 dice against 1 a billion times and not get 23 wins for the 1 in a row. Do you realize the math involved for that to happen?

23 50/50 cion flips i a row is almost 8.4 million to one. Meaning statistically you would have to flip the coin 8.4 million times before getting 23 in a row. Convert that into 3vs1 dice... impossible.

People in this thread keep throwing around the word "impossible" which is just hyperbole. Low probability is not the same as impossible. 1 billion to 1 is not impossible. Further, I suspect that the OP may have been exaggerating his claim but cannot say for sure. I will accept the premise that the dice are extremely streak oriented on this site and I have seen more than a few extreme dice variances.

[rhp 1]

I know there is the standard thread of idiots complaining...

but on the real though...

You ARE aware there are better algorithms out there for dice right?

sure over 20,000 games my dice will even out... but it is borderline retarded for you to tell me that i can possibly lose 23 vs 1...

hell we lose 15s+ vs 2s or 3s all the time (statistically impossible) and just chalk it up to bad luck... but seriously that should NEVER happen, and if it does it should be posted in forums on how insane it was... But we say "eh it happens, man i lose 15vs2 next game the other guy does the same thing"... two wrongs does not make it right... just makes it wrong both times...

c'mon boys let's fix this thing finally... hell my first game on here my buddy says to me he says "good site but the dice are crazy streaky"

yeah over 20,000 games my dice stats are even... but what that says to ME is that, i got hosed on a 23 vs 1 (statistically impossible) which means someone playing me will somehow lose 23 vs 1 (statistically impossible)...

equal bullshit does not mean the dice program works my friends

The dice clearly have some issues on this site. However, 23 v 1 is not "statistically impossible". It is Statistically unlikely but in a large enough sample it will eventually happen. If you flip a coin 1 million times you will likely get a few streaks of 23 heads or tails in a row.

You can't compare flipping a 50/50 coin with rolling 3 dice against 1. You could roll 3 dice against 1 a billion times and not get 23 wins for the 1 in a row. Do you realize the math involved for that to happen?

23 50/50 cion flips i a row is almost 8.4 million to one. Meaning statistically you would have to flip the coin 8.4 million times before getting 23 in a row. Convert that into 3vs1 dice... impossible.

People in this thread keep throwing around the word "impossible" which is just hyperbole. Low probability is not the same as impossible. 1 billion to 1 is not impossible. Further, I suspect that the OP may have been exaggerating his claim but cannot say for sure. I will accept the premise that the dice are extremely streak oriented on this site and I have seen more than a few extreme dice variances.

you're right... "impossible" is "hyperbole".. but when applied to the scenarios that Krapht is discussing in this thread, 23v1 is, for all intents and purposes, not possible... and for it to happen so many times (I know I've lost more than 1 20 something v 1 since playing on here) the odds for it happen once, multiplied by the fact that it's happened numerous times? sorry.. it's impossible...

[Metsfanmax]

I will accept the premise that the dice are extremely streak oriented on this site and I have seen more than a few extreme dice variances.

Streak oriented compared to what?

[trinicardinal]

Damn, now I need more popcorn. These dice threads always have me feeling to eat popcorn while I watch the drama. I guess because something is statistically unlikely means that it can't ever happen to you. I assume that means that you don't play lottery, carry insurance or anything else that deals with those statistically unlikely events. Excuse me for a bit... I'm going to make more popcorn.

[degaston]

...
you're right... "impossible" is "hyperbole".. but when applied to the scenarios that Krapht is discussing in this thread, 23v1 is, for all intents and purposes, not possible... and for it to happen so many times (I know I've lost more than 1 20 something v 1 since playing on here) the odds for it happen once, multiplied by the fact that it's happened numerous times? sorry.. it's impossible...

OP never clarified whether he was talking about 23 3v1 losses in a row followed by 1 win, or a series of 12 3v2 battles where he lost 2 troops on 11 of them, and they each lost one on one of the battles. If he's talking about the former, then he should have taken a snapshot of his battle outcomes, because I see nothing in his recent history that looks out of the ordinary. If he's talking about the latter situation, then not only is it not impossible, it's something that should happen about twice a month on this site. Nothing to see here, people. Move along.

There is a problem with the dice which has been documented in several other threads, but as far as I can tell, the problem is not with the numbers from random.org, but with how CC is using them.

[demonfork]

I know there is the standard thread of idiots complaining...

but on the real though...

You ARE aware there are better algorithms out there for dice right?

sure over 20,000 games my dice will even out... but it is borderline retarded for you to tell me that i can possibly lose 23 vs 1...

hell we lose 15s+ vs 2s or 3s all the time (statistically impossible) and just chalk it up to bad luck... but seriously that should NEVER happen, and if it does it should be posted in forums on how insane it was... But we say "eh it happens, man i lose 15vs2 next game the other guy does the same thing"... two wrongs does not make it right... just makes it wrong both times...

c'mon boys let's fix this thing finally... hell my first game on here my buddy says to me he says "good site but the dice are crazy streaky"

yeah over 20,000 games my dice stats are even... but what that says to ME is that, i got hosed on a 23 vs 1 (statistically impossible) which means someone playing me will somehow lose 23 vs 1 (statistically impossible)...

equal bullshit does not mean the dice program works my friends

The dice clearly have some issues on this site. However, 23 v 1 is not "statistically impossible". It is Statistically unlikely but in a large enough sample it will eventually happen. If you flip a coin 1 million times you will likely get a few streaks of 23 heads or tails in a row.

You can't compare flipping a 50/50 coin with rolling 3 dice against 1. You could roll 3 dice against 1 a billion times and not get 23 wins for the 1 in a row. Do you realize the math involved for that to happen?

23 50/50 cion flips i a row is almost 8.4 million to one. Meaning statistically you would have to flip the coin 8.4 million times before getting 23 in a row. Convert that into 3vs1 dice... impossible.

People in this thread keep throwing around the word "impossible" which is just hyperbole. Low probability is not the same as impossible. 1 billion to 1 is not impossible. Further, I suspect that the OP may have been exaggerating his claim but cannot say for sure. I will accept the premise that the dice are extremely streak oriented on this site and I have seen more than a few extreme dice variances.

Is losing Graham's number vs 1 impossible?

[bird45]

I have had a few insane battles none as bad as 23-1 but when it seems the dice program is broke I stop the attack. Seems to be the only option other than losing your whole stack to something that should be impossible. If I send 23 guys into battle against 1 unless its the terminator I should win hands down.

[Metsfanmax]

Is losing Graham's number vs 1 impossible?

Yes.

[loutil]

I know there is the standard thread of idiots complaining...

but on the real though...

You ARE aware there are better algorithms out there for dice right?

sure over 20,000 games my dice will even out... but it is borderline retarded for you to tell me that i can possibly lose 23 vs 1...

hell we lose 15s+ vs 2s or 3s all the time (statistically impossible) and just chalk it up to bad luck... but seriously that should NEVER happen, and if it does it should be posted in forums on how insane it was... But we say "eh it happens, man i lose 15vs2 next game the other guy does the same thing"... two wrongs does not make it right... just makes it wrong both times...

c'mon boys let's fix this thing finally... hell my first game on here my buddy says to me he says "good site but the dice are crazy streaky"

yeah over 20,000 games my dice stats are even... but what that says to ME is that, i got hosed on a 23 vs 1 (statistically impossible) which means someone playing me will somehow lose 23 vs 1 (statistically impossible)...

equal bullshit does not mean the dice program works my friends

The dice clearly have some issues on this site. However, 23 v 1 is not "statistically impossible". It is Statistically unlikely but in a large enough sample it will eventually happen. If you flip a coin 1 million times you will likely get a few streaks of 23 heads or tails in a row.

You can't compare flipping a 50/50 coin with rolling 3 dice against 1. You could roll 3 dice against 1 a billion times and not get 23 wins for the 1 in a row. Do you realize the math involved for that to happen?

23 50/50 cion flips i a row is almost 8.4 million to one. Meaning statistically you would have to flip the coin 8.4 million times before getting 23 in a row. Convert that into 3vs1 dice... impossible.

People in this thread keep throwing around the word "impossible" which is just hyperbole. Low probability is not the same as impossible. 1 billion to 1 is not impossible. Further, I suspect that the OP may have been exaggerating his claim but cannot say for sure. I will accept the premise that the dice are extremely streak oriented on this site and I have seen more than a few extreme dice variances.

you're right... "impossible" is "hyperbole".. but when applied to the scenarios that Krapht is discussing in this thread, 23v1 is, for all intents and purposes, not possible... and for it to happen so many times (I know I've lost more than 1 20 something v 1 since playing on here) the odds for it happen once, multiplied by the fact that it's happened numerous times? sorry.. it's impossible...

Color me skeptical that there are many instances of 23 or more vs 1 losing. I have over 800,000 rolls on this site and the worst I have ever experienced is losing 8 or 9 in a row against a 1. Something that should happen about once every 1,000 rolls but has only happened once or twice.

[demonfork]

Is losing Graham's number vs 1 impossible?

Yes.

What is the smallest number vs 1 that would be impossible to lose?

[Metsfanmax]

Is losing Graham's number vs 1 impossible?

Yes.

What is the smallest number vs 1 that would be impossible to lose?

Take the expected lifetime of Conquer Club and divide it by the number of rolls you could make in that time.