Drop rates changed on 10/15 as they have in the past.
The chances of a "fluke" or "luck" are 1 in 15 billion.
I was hypothesizing from personal experience that drop rates are modified in the days before or after patch. This may be intentional, or the result of some other reset/function. However, based on the below, there is no possible way this is the result of RNG. My sample size, while seemingly small, with 2 individuals, is actually quite large. It consists of ~500 monsters killed an hour, over 2 hours daily, over 120 days, for 2 separate independent accounts. Any efforts to refute the math below will only solidify my claim, as I was exceptionally lenient in rounding and incorporating outlying probabilities.
-My bf and I run A3 together around 2 hours daily. He's military, and I'm on the East coast, so we chat and slay demons. It's lame & cute.
-Neither of us MF and we're both around 50 paragon. I swap to mf on elite kills (my equipped gear).
-The day before and after patch 1.04 we received a drastic spike in legendaries (around 20x higher than normal rate. The day of patch we each received 7-8 legends, as opposed to the 1 between the two of us we generally average. At that time we were both mf swapping on elites, but paragon MF was not a factor.
-On 10/15 we logged on and received FIFTEEN legendaries within 3 hours. They were high quality legendaries dropping every 10-15 minutes. We even relogged a few times, but the drops continued to come. I immediately made a post predicting that 1.05 would be released today, but it was taken down.
-Later that evening I made THIS post
-Hours after that, 1.05 was announced, proving my prediction based on drops correct.
-All our legendaries came from 'white mobs', MF was a constant 200 or less each w/ NV.
-Lets assume a legendary drops from a white mob with 200mf one time out of 3000 (1/3000).
-To simplify, lets say we received 14 legendaries in 3 hrs, and not 15 (7 ea) and
-We kill exactly 500 monsters an hour. x3hrs= 1500A and 1500B
-Since there were 2 players, we must consider each environment individually.
To simplify this to 'basic' probability, please consider these analogous problems:
(A)my has a deck of cards with 3000 cards.
One of these 3000 cards is orange. The rest are white, blue, or yellow.
Daily, Mary shuffles her deck of cards and draws one card, then replaces it and reshuffles.
She repeats this task 1499 more times, a total of 1500 times.
Mary draws the orange card exactly SEVEN times out of her 1500.
What is the likelihood of this happening?
1day / ~74,000 days
(B)oyfriend has a bag of marbles.
One of these 3000 marbles is orange. The rest are w, b, or y.
Daily, B shuffles his bag of marbles and draws one marbles, then replaces it and reshuffles.
He repeats this task 1499 more times, a total of 1500 times.
He draws the orange card exactly SEVEN times out of her 1500.
What is the likelihood of this happening?
1day /~74,000 days
What is the liklihood of A Drawing the Orange card EXACTLY 7 TIMES AND
B drawing the Orange Marble EXACTLY 7 times?
1day / ~427,000 days
(This is based on the assumption that droprates being modified in the day before and after 1.04 and the day prior to 1.05)Finally, A and B draw items once daily for 120 days. 3 of these days are the day before or after patch. 117 are not. What is the chance that A and B will draw Exactly 7 orange items on EACH of the 3 patch days?
What is the chance that A and B will draw Exactly 7 orange items ONLY on EACH of the 3 patch days and NONE of the other 117 Days?
Please consider the above math does not include the chance that we would have gotten more than 15 (as in 16-3000 legendaries out of 3000 rolls) drops. In this case, the question would have been "Mary draws the orange card AT LEAST SEVEN times out of her 1500. What is the likelihood of this happening?"
However, to counteract the likelihood of all of these occurrences, I lowered my actual droprate (15 out of 3000 rolls) to 14 out of 3000 rolls. This more than accounts for this exclusion, for the chances of getting over 15 legendaries out of 3000 decrease at an exponential rate in a parabolic fashion towards 0. For example, if you had a deck of cards and shuffled then chose one card and replaced it 100 times, you would be thousands of times more likely to draw the Ace of spaces 1x,2x,3x,4x,or 5x vs 6x,7x,8x,9x,10x,11x,12x,13x,14x,or 15x out of the 100 draws.
Also, please during the time it would have taken us to get to 5 paragon with each new game, our MF would have been below the 200 I've stipulated above, further reducing the probability of these drops.