The unfairness of mog bonanza towards PS2 playersFollow

PAGE CLAI--- DAMNIT

The answer's still "do it yourself woman".

Yeah, I agree.

I don't really give a **** about the walls unless they are breaking apart (or, as a nod to FF4 and Star Wars, closing in on you). In which case, paint isn't really what you want to solve the problem.

Hint: Consider E's and O's as binary digits and count.

This represents all possible digit patterns:

OOOOO
OOOOE
OOOEO
OOOEE
OOEOO
OOEOE
OOEEO
OOEEE
OEOOO
OEOOE
OEOEO <- PS2 RNG pattern
OEOEE
OEEOO
OEEOE
OEEEO
OEEEE
EOOOO
EOOOE
EOOEO
EOOEE
EOEOO
EOEOE <- PS2 RNG pattern
EOEEO
EOEEE
EEOOO
EEOOE
EEOEO
EEOEE
EEEOO
EEEOE
EEEEO
EEEEE

2 / 32 = 0.0625

This means that with the PS2/PS3 number generator you can reach only 6.25% of all (100000) possible five digit numbers.

PS2 makes it insanely easy (albeit lengthy) to get ebisu rod while on pC you have no chance. PS2 is the only possible way other than running a script you will ever ever claim anything against competition. PS2 makes it to where they cant put a visable cape or a bow on a characters back. Even if it was rigged that makes us even.

That's the point. In order to calculate the probability that a single, truly random value falls into the pattern of that generated by the PS2/PS3 (EOEOE or OEOEO) you have to consider the whole field of possible numbers.

Strange as it may seem, restricting the numbers given in a "random" selection to PS2 players does not affect their odds in any category as long as the draw itself was random.

Here's one way to look at it. Consider the rank 5 number (match all 5 numbers). A truly random generation has 10x10x10x10x10 combinations = 100,000. The EOEOE or OEOEO numbers are related and have 10x5x5x5x5 combinations (once the first digit is assigned which can be any of the 10, the remaining digits are causally related) = 6250 combinations.

If the number is drawn at random, it has a probability of (100000-6250)/100000 =.9375 that it will be in the range not covered by the PS2 numbers. It has a probability of 6250/100000 = .0625 that it will be in the PS2 range.

If you have a PS2 number, your odds of winning are the percent chance of it being in your range * the percent chance that your number is the correct one in that range. That is .0625 *1/6250 = .00001, or 1 in 100,000. Same as for a truly randomly generated pick.

You can do the same for each other prize rank.

I think that is why SE updated the info to say that the numbers were randomly generated - that is all that matters for a "fair" lottery.

They still screwed up though. This error created a huge skew in the population distribution with all of the PS2 numbers stacked up in that 6.25% range. For sake of argument, lets say there are equal PS2 users and other users (don't know if that is true or not - this is just an example). Lets also say there were 1 M "tickets" sold (again, not a real number). Out of 1M tickets in a random distribution of numbers sold, there would be 10 grand prize winners.

In our example, though, the distribution is not random. 531,250 of the tickets would have been in that .0625 window (all of the PS2 tickets and .0625 of the remaining 500K tickets). Each of those has a 1 in 6250 chance of hitting the jackpot. There would have been 85 winners.

Now, this would eventually even out if the lottery were held a statistically significant number of times. For a once off event though, it could have been very ugly.

Did SE realize their mistake and check the distribution of winning against intended statisics? If they had drawn the rank 5 number in the .0625 zone, would they have nulled the result and allowed everyone to re-pick due to the error?

In a one off lottery, they should have nulled the result, admitted their mistake and allowed everyone to rechoose using a truly random number. The fact that having a "wrong number" picked for the big prizes would have resulted in a big exposure for SE raising the possibility of impropriety. In a RL lottery, this would have resulted in huge lawsuits.

There's one thing that keeps popping up in my head.

This is a pretty obvious pattern. EOEOE and OEOEO as the *only* combination would stand out to quite a few people.

What I want to know (seriously, I want a response) is WHY this was only brought to light now, AFTER nothing can be done about it.

It's very obviously not because no one noticed it. It's waaaaaay to convenient that a day or two after the drawing that someone went through and saw a pattern. So if someone noticed the pattern and didn't say anything about it, why was nothing brought to the attention of the community at large, or SE itself until after the numbers were drawn?

Did you think that *maybe* you found a glitch in the system that could work out to your favor...that instead of a 1/100,000 chance, you actually had a 1/10,000 (or w/e it is with the pattern) chance so you just let it slide?

What if the winning numbers had followed the EOEOE or OEOEO pattern in this same instance, where only the PS2 base followed that number generator pattern? Would this same response have popped up, or would the people who found this pattern just "let it slide"?

Having said all this, I completely understand why some people are upset over this, and I understand where some people see how SE could have hand-picked the number to avoid flooding the market with gil. I don't believe that it happened, but I understand where the point is coming from.

Like I said, it just seems WAY too convenient that only now after the winning numbers were announced...literally less than a day between that and this thread being posted...was this problem brought to the attention of the community at large. So while thinking about this, is this really a legitimate complaint or is it butthurt that a glitch that you thought would work out in your favor didn't?

TOTALLY

In my case, I bought the 10 marbles per character expecting them to be random then put them in storage and the closest I got to looking at them was to think "Wow I can't wait for the drawing so I can get rid of these." It's like buying a quick pick lotto ticket, you don't look at the numbers until the drawing because until the drawing they aren't important. Did you look at your mog marbles every day from the time you bought them until the time of the drawing?

I was reading alla while I was looking at the numbers and came upon the OP. If I had not, I wouldn't have seen the pattern because I was looking at whether the numbers matched the drawing numbers and only focusing on the relevant digits rather than the numbers as a whole.

I agree, there would have been a wrong if the PS2 patterns had been chosen as winning numbers because then the gil they would have won would have been worth slightly less as there would have been more gil introduced into the worlds.

Until someone can show that the winning numbers were not chosen randomly and were more likely to not favor EOEOE or OEOEO, the only wrong is the one I've listed (which is certainly a wrong).

I don't think they hand-picked any numbers, but I am almost positive that they checked the numbers they generated to see what kind of distribution the winners would have. This seems to be the only prudent thing to do. Imagine a scenario where SE randomly generated a winning number (not using a PS2 lets hope) and one of the winners was a level 5 taru named liausbflf, with all 10 marbles having the same number, and an IP address in China - would be a disaster. In fact, I am sure they checked to make sure none of the Rank 1 winners won more than once or twice.

Ok, taking this one step further, assume SE generates a winning number that follows the alleged "PS2 pattern", this number would be discarded because a disproportionately large number PS2 players would be winners. At least this seems like a very plausible explanation to me.