Page 2 of 2

Posted: Mon Feb 19, 2007 4:48 am
by nzld

All I am trying to say here is that the dice rolls are not random. I am not critcizing the program, in fact I repeatedly have said how much I enjoy it. All I am saying is that the dice rolls are not random.

The computer code means nothing to me, I am not a programmer. Hearing that it is a windows function answers alot.

Todays' game, I controlled Venezuela, Peru and Argentina. Brazil had no extra troops on it. I attacked it with 20 from Argentina. Somehow I lost all of them without gaining the territory. I then attacked with 15 from Peru. Again, I lost them all without gaining the territory. Lastly, I attacked with everything I had left on Vensuela, 16 of them. Again I lost them all.
To say that this sort of thing is bound to happen if you play enough, is like saying that if you play the lottery every week you are bound to have a streak where you win every week. Again, just not possible.

On another note I have started changing the numbers of competitors and armies etc, and am enjoying the game anew. The dice still frustrates me, but I am getting a lot of fun for my $20.

In Defense of Conquest Dice Rolls

Posted: Sat Mar 17, 2007 10:28 pm
by cklarner
I ran a number of experiments to assess the dice odds used in Conquest, the Risk computer game, on January 17, 2007. I created large piles of armies, then used ?automatic resolution? to resolve the battles. There were a total of 5,428 dice rolls in these battles. Over the five mass attack experiments, the attacker lost 4,923, and the defender lost 5,933 (defender lost 20.52% more, or the attacker wins 45.3% of the time). By my calculation of Risk rolls (3 v 2), the attacker loses 46 to the defender?s 54 on average. That matches what the computer yields pretty close. This doesn't address the issue of "small battles" or going against one defender, but it is instructive.

I play this game A LOT and have not seen any evidence of messed up die rolls. If you have a complaint with the dice rolls, do an experiment similar to the one I did above and keep records. Humans' ability to make subjective assessments of probability is EXTREMELY BAD.

One last thought: perhaps something is going on in the demonstration Conquest, versus the full blown Congquest?

Typo Above

Posted: Sun Mar 18, 2007 1:27 am
by cklarner
It should say the attacker LOST ~45% of the time, not won. My apologies.

Posted: Mon Mar 19, 2007 7:52 pm
by sid6.7
well if you've played the board game long enough you undertsand
that with dice rolls anything can happen, yes you are bound to
have a losing streak if you play long enough, yes you are bound to
have a winning streak if you play long enough...the computer version
is no different...

yes i have taken 20 VS 5 and lost
yes i have taken 5 VS 11 and won

how often this occurs i've never kept track could be LONG
periods of time between or back to back games...

its all randomness....

Posted: Sat Jun 23, 2007 1:32 am
by Empty Hat
*little OT

Have any of you read Darrel Huff's 'How to lie with statistics'?

Good intro book

It has an anecdote in it about some statistical test researchers who ordered a random number generator for their experiments; on their first test it generated six 1's out of six numbers.

Feeling somewhat edgy about this the researchers called up the company and asked them about this, the company agreed that this was possible but unlikely and agreed to check the machine.

When it was returned to the researchers they spent an entire day asking it to generate numbers and found that none of the results were six 1's out of six.

As it was statistically probable that this should have happened at least once during the day they again called up the company who admitted that they had found nothing wrong initially but had removed the capacity of the machine to roll six 1's.

Of course the researchers needed the possibility of six ones for their experiments so the machine had to be 'fixed' again.

I frequently balk at how the computer runs through my light defences with so few losses leaving me to re-conquer my former territories for far higher losses but I accept that this is just my own whineyness and that I probably only feel cheated.