dice.vs.skill

[Thezzaruz]

If something is designed to be random and that design is successful at it's function,
you are claiming that the successful results of that design are not random?

There are a few different definitions of "random" in the science world. If you use the more strict (and grammatically correct) as Klobber does then yes, it's impossible to create something "random". For some uses/definitions of "random" it is considered possible to create randomness as long as the procedure you use to create it follows certain rules and the result is distributed within other rules.

Basicly it's about using the word "random" (as its concept is more widely known) when in fact "unpredictable" would be more correct. For all uses on this site an "unpredictable" dice that keeps within the probability distribution is what we need, if it is truly random or not is actually irrelevant.

[oVo]

"Unpredictable" is an excellent word. I'm always surprised by how many times the attack/defender rolls can randomly come up the same. Six/five vs six/five, 3/1 vs 1/3 etc. and wonder if this is odds defying behavior or something that is to be expected.

Also... how rare is five dice snake eyes?

[kerntheconkerer]

Develop a pre-roll routine...it's perfect! I knock on wood three times, say a secret chant under my breath, and then roll the dice. Repeat for every button push.


--Andy

lol it doesn't seem like something you would do for clicking a button, but maybe actually rolling real dice

[Timminz]

Also... how rare is five dice snake eyes?

five 1's would appear .01286% of the time.

five of a kind, without caring about which number will appear .07716% of the time, which would be about once every minute and a half (based on Twill's 15 rolls/sec, mentioned in another thread)

[Thezzaruz]

Also... how rare is five dice snake eyes?

five 1's would appear .01286% of the time.

Or 1/7776 or one every 9 minutes or so if you're looking for that special case...

[RADAGA]

So, for 4 straight rounds I´ve been trying to conquer this neutral territory.

I put 3 armies at my land, and attack.

first day: 1;4;4 vs 5;5
second: 1;2;4 vs 5;6
third: 3;3;6 vs 6;6
fourth: 2;4;5 vs 6;6

so far, 8 armies wasted and none killed at the other side. I am not complaining about the dice. I am asking if I can borrow the defense ones to make my D&D characters.

[KLOBBER]

...say a secret chant under my breath, and then roll the dice. ...


--Andy

I've just GOT to find out what that secret chant is!

[RADAGA]

out of the blue, at random.org

You rolled 16 dice:

5 1 1 1 1 1 5 6 5 6 1 4 5 1 5 3

five ONES in a row.... 7 ones in 16 dice.. 43%

Timestamp: 2008-07-07 14:11:46 UTC

You rolled 16 dice:

6 1 6 1 2 1 1 4 1 6 3 5 1 3 5 4

6 ones in 16 dice ... 37%

Timestamp: 2008-07-07 14:12:36 UTC


You rolled 16 dice:

1 1 2 1 4 5 5 5 5 6 2 3 1 5 5 2

four 5 in a streak .. 6 fives in 16 dice 37%

Timestamp: 2008-07-07 14:12:50 UTC

You rolled 16 dice:

6 6 4 6 5 6 4 5 5 4 4 2 4 6 2 6

5 fours 6 sixes

not a single one, not a single three.

Timestamp: 2008-07-07 14:13:28 UTC



the problem are the streaks. since ties favor defense, every streak of numbers means defense will get it. and there are far too many streaks....

[RADAGA]

lets analyse:

5 1 1 1 1 1 5 6 5 6 1 1 4 5 1 5

5 1 1 1 1 1 attack, 1 defense
1 1 1 1 1 all defense
1 1 1 1 5 all defense
1 1 1 5 6 all defense
1 1 5 6 5 all defense
1 5 6 5 6 all defense
5 6 5 6 1 1 attack 1 defense
6 5 6 1 4 all attack
5 6 1 4 5 1 attack 1 defense
6 1 4 5 1 1 attack 1 defense
1 4 5 1 5 1 attack 1 defense
1 4 5 1 5 1 attack 1 defense

conclusion: STOP ATTACKING! lol

[RADAGA]

I just missed five 3x1 .... in a row.

Then, on the same round, missed four more.

Then, on the subsequent round, I missed six 3x1 in a row.

Then two more.

then I got to random.org and rolled 16 dice

You rolled 16 dice:

3 2 5 4 2 6 3 3 3 3 5 5 3 4 2 5

Timestamp: 2008-07-16 13:31:25 UTC

four three in a row. that, for a game where a tie favours a side, is terrible. No matter how statistics show the average is ok. The spread is biased. in 1.000.000.000 rolls there might be even in statistics, and even then be pure BS.

for instance. If I roll 36 dice and say the results were:

1111111.2222222.333333.444444.555555.666666

In that sequence. That make them statistically perfect: six times each value, in 36 rolls, each with a possible outcome of 1 out of 6 numbers.

look at that other roll:

Die Roller

You rolled 16 dice:

4 5 4 6 4 6 4 4 4 6 4 5 3 4 5 1

Timestamp: 2008-07-16 13:35:22 UTC

eight fours. in 16 dice. including 3 in a row. pretty average, i´d say


next:

Die Roller

You rolled 16 dice:

4 6 4 1 6 6 6 2 5 1 3 3 3 4 4 4

Timestamp: 2008-07-16 13:38:33 UTC

6-6-6 then 3-3-3, and then 4-4-4 there are just WAY TOO MANY in-a-rows on those dice. But of course, this post will be either ignored, or mocked, so, in the end, why bother?

[Sludge_King]

Is it best then to use the auto attack option? I just lost like 15 times straight to a country with three armies. probibly lost me the game.

[Timminz]

You rolled 16 dice:

4 6 4 1 6 6 6 2 5 1 3 3 3 4 4 4

Timestamp: 2008-07-16 13:38:33 UTC

6-6-6 then 3-3-3, and then 4-4-4 there are just WAY TOO MANY in-a-rows on those dice. But of course, this post will be either ignored, or mocked, so, in the end, why bother?

This would be divided up into groups of 5 for CC. You'd have 4,6,4 v 1,6 (1 loss each) 6,6,2 v 5,1 (attacker wins 2) and 3,3,3 v 4,4 (defender wins 2)

There's only a single occurrence of 3 in a row there.

[FabledIntegral]

Is it just me or are the dice making this game less and less about skill im starting to lose 10vs1 then im completely out and its happenning every other game is the game becoming more about whoever has better dice?

Just so you know, the chance of losing 10 vs 1 is less than 0.0001%, since its 0.001% to lose 6 vs 1.

So that roll is once in a lifetime.

Lost it mroe than 5x, (and yes I know the statistics), which has lead to me having numerous rants in the forums. I've taken statistics, the dice are significantly more streaky than probability suggests.

[RADAGA]

You rolled 16 dice:

4 6 4 1 6 6 6 2 5 1 3 3 3 4 4 4

Timestamp: 2008-07-16 13:38:33 UTC

6-6-6 then 3-3-3, and then 4-4-4 there are just WAY TOO MANY in-a-rows on those dice. But of course, this post will be either ignored, or mocked, so, in the end, why bother?

This would be divided up into groups of 5 for CC. You'd have 4,6,4 v 1,6 (1 loss each) 6,6,2 v 5,1 (attacker wins 2) and 3,3,3 v 4,4 (defender wins 2)

There's only a single occurrence of 3 in a row there.

no, to be true with analysing this data, since you dont know what position would you start, you would have to make
464 16 AD
641 66 DD
166 62 AD
666 25 AA
662 51 AA
625 13 AA
251 33 DD
513 33 AD
133 34 DD
333 44 DD
334 44 DD

3 full attacks
3 mixed
5 full defence

Remember defence have a LOWER probability to win a 3x2

Most of the times you go to random.org and tell it to roll 16 dice, the possible streaks that would generate CC dice favours defence. Thats due to the unreal amounts of streaks you get.

My guess? static doesnt varies fast enough, so the readings, from random.org, due to who knows, lag(?), get the same "atmospheric static value" several times, generating those streaks

[Timminz]

The way CC has said they use a dice file, is to divide it into batches of five. If you only roll 3v1, the last one gets discarded, and the next batch of 5 is used for the next roll. So, in the situation you provided, it would go the way I said.

[RADAGA]

The way CC has said they use a dice file, is to divide it into batches of five. If you only roll 3v1, the last one gets discarded, and the next batch of 5 is used for the next roll. So, in the situation you provided, it would go the way I said.

Yes, BUT you dont know what position you would start to divide the batches. What if, at the moment CC gets their data file the random number generator is generating the, lets say, second "four" of that list:

4 6 4 1 6 6 6 2 5 1 3 3 3 4 4 4

then you would have to get the batches as 416 66 251 33 and 444 _ _

and what if the list is on the first six?

well, you got the picture.

So, to see how "good" a random.org roll is, you have to examine every possible batch of five in a given roll.

[RADAGA]

Game 2848683

I get 3 armies, and try to attack a neutral.

I´ve been doing that for 16 rounds so far

15 of those attacks were full defense victories.

1 of them, a mixed.

So far, I have lost 31 armies versus those 3.

But whatever, after all, it is extremely common, using random results, to loose 15 of 16 3x2