One thing that will skew the odds is that the game rolls a reward on table 3, and notes the result. Table 2 gets rolled, if it matches the result from table 3, reroll table 2, otherwise, note the result. Finally, table 1 gets rolled, if it matches either of the previous results, reroll table 1.
Not allowing duplicate rewards from the different tables will skew the observable probabilities of table 2, and table 1 even more so.
You’re comparing apples and oranges here. The old system took approximately 85k gold for a full cycle of gems-tokens-keys, which would yield 12 glory keys and 10 gems. This was a total guild-wide donation, so it should be compared to the guild-wide 1 mil, or it should be split 30 ways to be compared to the 33k/member average.
Not as strong of a RoI as the old system, but still better than the 750k final tasks in each color.
Yup I get that restricting the drops from table 2 and 3 based on what dropped before will skew the results. I still expected the probs to be closer though. Maybe it’s just me …
For the moment, I don’t know if it’s a good idea to continue because of the seals: when devs will put them back normally the weights will be changed and we will have to start again… Is there news about the seals @Sirrian?
Else, I confirmed that in Table 2 Event keys are dropped x2 OR x3. It’s strange…
Finally, I updated the stats with Marthos Guardians 22 Legendary tasks ;-). If someone need our stats, I can give it here.
We just completed a legendary task on Xbox One and received 2 arcane stones, 6 gem keys, and 3 event keys. Has the Table 2 event key reward been adjusted on PC/mobile as well, or was this a console-specific thing? FWIW, we’ve completed a number of legendary tasks each week, but this is the first time I recall getting 3 event keys.