wolfpack0530 wrote:I guess i am admitting to your first point now bakc in the other thread. took a while to see what you meant, but i finally realized it. Yes, that is what i want mets. Rig the dice, change the formula translate the properties of perfect randomness to a small sample size. guarantee me that i will only see 30 lose to a 2 times or less in a calendar year. if i have been unlucky enough to see it a third time, then elinate that specific possibility from happening through, code, manipulation or anything. make occurance of super rare events over 10000 rolls match their perfect probabilities of 10,000,000,000,000,000,000,000,000 rolls.
I want lack to consider all possibiliies, and even possibly reject the notion that randomness = fairness. i realize i have created a straw man above, and if i screwed up translating the jist of your arguments then i apologize for that. the point here was not to discredit you or even attempt to. feel free to continue to point out the flaws in my reasoning, as it helps my understanding of the issues. just know that i may continue to keep complaining that blue is actually red as long as i keep seeing the red happening every day lol.
There are two separate issues being discussed here. On the one hand, you have the argument made by many individuals that the dice are
not actually random. That is, they "regularly" see some event such as losing a 30 v 5, and from that conclude that the dice cannot be random, because randomness implies that such an event should happen proportionately infrequently. The flaw in this argument, as you seem to understand, is that this does not actually prove that the event is happening too frequently. That is, the dice rolls have to be considered over a large set of samples, not just a small set from, say, a period of a few games. From a purely rational point of view, I believe that it must be the case that if you considered the rolls over a large sample, then you would see 30 v 5 winning at roughly its expected rate, since I trust the randomness of random.org numbers. I submitted an eTicket a while ago for lack to release some of the old lists for analysis, just to confirm that; the admin who responded said that this would happen sometime in the near future (doesn't seem to have occurred yet). Still, unless lack is lying about where the data comes from, I fully expect that the dice are random.
The issue that you raise here is quite a different one. There are two problems I see. The first is on the level on which you are trying to engage me. You say that it is more fun when you know that a ridiculously unlikely event cannot actually happen. I'm not convinced that this is true. I'm a frequent chess player, and what I love about that game is the pure determinism - at a given time, there are a finite number of moves that can be made, and you know exactly what happens when a given move is made. When I come to Conquer Club, though, I'm not looking for pure determinism. I'm looking for the thrill of rolling the die. I love it when I roll 30 v 20 and win basically every roll, or when I roll 5 v 7 and manage to grab the territory. It's a great feeling. I also feel lame when I roll 30 v 20 and it ends up being 3 v 12. But that
thrill, the not knowing what's coming, is what I think is valuable about Risk, and that makes it different. It has every right to be called a pure strategy game, and yet you cannot avoid taking some risks to win. You have to
hope that your luck turns out. Sometimes it doesn't work out for you, and sometimes it does. But if you take away the losing 30 v 5, you also take away the winning 20 v 40, and I don't want that to be taken away from me! ;P
The logical flaw with your suggestion is that rolls are not considered as a set - that is, when you go 30 v 20, you're doing several rolls sequentially. Even if you click auto-assault, the game engine processes each roll separately. You can't directly program a result such as "win 30 vs 20." You might say, "Well, in that case, just make it so that if I click auto-assault in this case, it just skips the dice process and lets me win." The problem then is - how many troops do you have left over afterward? Do you have the most probable result every time? If this is the case, then you have to take all of the randomness out of the RNG, because there's no non-arbitrary way to determine when your "hack" will be applied and when it won't be. For example, let's say you have borders in question - on one border, you have 30 troops and the opponent has 5, and on the other you have 29 and the opponent has 5. If you lose the 30 vs 5, your suggestion would imply that you should then have a guaranteed win for your next 30 vs. 5. But you could still lose the 29 vs. 5 attack, because your rule didn't cover that situation. However, it wouldn't
feel any different - you'd still feel like you got terrible luck both times. There's just no feasible way to implement such a code!