LordLundar said:
And you would be wrong because of the gambler's fallacy. In this case it's a confusion between the percent chance and the "a in b" odds, in this case 3 in 100. With the latter example it's a guarantee that within the selected number of draws you will get the result you're after, so a failed result only increases your chance of a subsequent successful one. The percent chance however is not affected by the previous draws so whether it's the first or one hundredth draw you still have the same percent chance.
Yes, I know that, however, if 1000 3% tries do not even yield a single success[footnote]let alone the 30 there should have been on average[footnote]Or even the 60 that there should have actually been, as it was a 6% chance[/footnote][/footnote] then it doesn't seem like the chance really was 3% or the RNG could be really bad. If you're hitting such a long "dry" stretch, it's a good indication that the RNG be broken.
This isn't
actually the case - there are certain properties that pRNGs need to conform to and we
can know if they are wrong or not. It's hard, true, as they might malfunction on, like, the 10th million iteration of spewing random numbers but we can still observe and determine that.
LordLundar said:
There is also the issue that the percent chance listed is an overall average of results, not a guarantee.
No, that is incorrect. Unless you have a different source of information, the
chance listed was at 3% and for that day it was, in fact, doubled. This
is the guarantee. Or at least what you're promised - whether or not the actual chances were 3% (or 6%) is not easy to know - it is entirely possible that the chances were 2% and were unchanged by the promotion.
LordLundar said:
So while his results were much lower than the average
Much, much, much,
much,
much lower than average, yes. One could say, almost unrealistically so. Yes, it is possible for that to happen, however, the odds are so extremely low that it warrants suspicion. Especially in this context - the company explicitly profits from players losing on the 3% (or 6%) chance. If the chances were fudged in some way, then that's sort of criminal in some places in the world.
LordLundar said:
other people were far more fortunate and got it first time, maintaining the average.
Prove it, then.