Hold up Roarke, I'm not sure that the probabilities actually calculate out favorably that way. Though I can't figure it out for myself... we never stayed long on probability calculation in my math class and the last time I did any work with probabilities was back in like grade 10 or 9 (so around 6-7 years ago). Anyway, here's my reasoning.
Cuz you have to both get the 1/10 chance of getting your replacement parent and the 1/12 chance of getting the perfect right? You have to hit two outcomes instead of one, and while both of those outcomes might be more likely than the one, getting both is less likely. Except I'm not sure if I'm putting this together right. Actually I'm pretty sure I'm not. It's the difference between rolling a pair of dice and trying to get snake-eyes and trying to roll a single die until you get a one twice. I'm not sure if there's a difference in the probability calculations between the two. The goal here is to figure out on average how many rolls it'd take to get two ones. The first example (using a pair of dice, rolling both simultaneously) definitely uses the simple multiplication method we used to arrive at the other probabilities (in this instance, 1/6 * 1/6 = 1/36). The second one though (which is similar to the breeding scenario in question), I'm pretty sure it doesn't? But I don't remember how you calculate the average number of roles it'd take in that scenario. Does it just add together? That would make it an average of 22 eggs for McGrrs method. But I'm not entirely certain that's the way it works.
Anyone know the math to figure this out? I want to make sure that we got the best method up there. Also, it'd just be good for me to finally get a grip on how to calculate probabilities again XD.
I didnt want to think too hard on this and I didn't take it too seriously to get out a piece of paper to make sure it was correct, but i think he was unclear in saying that he would use an everstone on the 31/x/x/31/31/31(call it parent A) with the newly bred 31/x/31/31/31/31 parent (call it parent B).
parent A i would assume had the original correct nature, while parent B does not, but since everstone negates that by putting everstone on parent A and destiny knot on parent B. So at this point he's corrrect in saying he has a 1/12 chance (1/6 * 1/2)
He then says once he has two 31/x/31/31/31/31 parents he has a 1/6 chance to make a new one.
I believe his calculations are correct UP UNTIL the point where you start calculating Gender rates, which actually do matter. It gets a bit more complicated when you start including gender and it would also depend on which gender you managed to find first, and i would need a pen and pencil and calculator to find out the exact rates.
The reason You are incorrect in using the simple multiplication method is because we do not have to multiply the 1/10 and 1/12 for EACH Scenario of the 10 (hard to explain in words at 2am) im saying there are 10 different scenarios that can happen, but if you get the first 9 unfavorable scenarios, you are not using those scenarios to start breeding for the next stage of breeding (1/12 probability scenarios)
you would wait for that 1/10 scenario to occur in which you can then start searching for the 1/12 scenario. A better estimate would be to add the 10 and 12 together and say your probability is 1/22 which is ABSOLUTELY INCORRECT, but it would be more accurate then multiplying 1/10 and 1/12 which is 1/120. I would need a pen and paper, and still probably get the correct probability incorrect, just cause it starts getting confusing.
I dunno if any of this made sense, im sure it wasnt formatted or worded well, because im tired and kind of rambling, but i hope it helps