When you give ranges, is it straight RNG within that range (so the expected value would be the midpoint)?
To add to this, there is no guild that has a perfect 33,333 cost spread. When deciding whether you, personally want to spend more time into the game grinding out gold beyond what you expect your guild to match, you could be shouldering a larger percentage of those legendary tasks. If you are a consistently higher donator, over time, you consistently pay more than 33,333 per LT, on average. The people that pay in less or get them for free are irrelevant because resources are non-transferrable (the exception of a very tiny amount of gold which may mitigate the average price of a LT by a percentage point). This puts a soft cap on how much you can grind toward any given goal with respect to also requiring others to grind with you if you want to complete it at a reasonable rate. For the relevant purposes of this discussion, I have to assume that you have already outpaced your guild and are looking to grind extra gold and put extra gold into gold keys rather than legendary tasks at a point where only x people are matching your donation (considered over an extended period of time, not locally) and your cost per task is 1mil/x. Having everyone pull their donations from LTs is folly, as just at a glance 30 LTs will almost invariably get you at least 11 gem keys and 100 glory keys or equivalent, which is the same purchasing power as the gems and glory gained from the gold keys each person can purchase for the same amount, not even counting the raw mythic drop chance from the LT.
On to that. A 1 in 124 chance of mythic in LT is much higher than I was expecting. However, this is still a random mythic, while VIP keys spent appropriately are always the mythic you want. This would, accordingly, have a āpersonal worthā of (missing mythics)/(total mythics)/124. If you need three like I do, thats a ~1/703. If you need one, thats a 1/2108. The chance worsens every month both because you might fill a spot and another mythic enters the drop table. If you have them all at the current moment, this mythic has zero value because you are simply building up resources so you donāt miss the next one.
I still contend that LTs are going to be better when the actual raw resources are going to be better, for almost everyone. However, from an economic standpoint for the people that are already grinding above and beyond a baseline and wish to grind even more to try and further their collections, there may be a cutoff when gold keys are actually still better than LTs for the purposes of collection. But at the same time, I also contend that it is still a bad idea for another reason.
Keep in mind that if we can earn 250k/hr, which is over 4100 gold per minute, you have now done 4 hours of speed grinding PVP to get about 3333 keys. This gets you about 50 gems and 2k glory. Counting the gems, you have only increased your purchasing power by 0.01~0.011 mythics at vip 5, and 0.0055 otherwise. Adding in glory from gold keys gets us to 0.02~0.021 for VIP 5 and 0.0155 otherwise, and an estimation of glory earned from PvP of 300/hr gets us to about 0.026-0.027 mythics at VIP 5 and 0.0215 otherwise. For 4 hours of your life. I think people need to really sit back and see the magnitude of a 150-154 hour grind with VIP and 186hrs otherwise with no intermediary short term step goals and whether or not they could even maintain interest in the game that long. And when you are done, you donāt generally bounce back. Iāve had zero-intermediary grinds of even just a few hours kill a game for me entirely.
1 Like
Thatās my assumption, but as the actual drops are done server side I canāt check that, and am only reporting the data as provided by the game.
1 Like
Good points about not everybody donating equally, and I noted that assumption in my earlier post. However, with LTās being over 25x better, you donāt need the whole guild to be ākeeping paceā. As long as there is at least one other person donating at your level, then your overall mythic expectation will still be slightly higher than with gold chests (and continuing to increase as more than 2 people are doing higher donations).
Yeah, it surprised me too that it was so high, and my guild hasnāt had that sort of luck. In the last 6 months, weāve done almost 700 LTās, but only pulled 1 mythic. Iām wondering if there might be something different actually happening in the back-end then what the drop tables and dev comments indicate. So for completeness/clarity, what I found is that the actual drop rate for mythics is 0.5% (half a percent, so 1 in 200). However, as the devs have indicated that selected items wonāt be repeated, once that is accounted for, the drop rate should increase to just over 0.8%. So despite trusting the math, I acknowledge that the expectation doesnāt line up with the results Iāve seen or heard reported, so I am a bit puzzled.
2 Likes
I think I might be able to answer why the 0.8% estimation at least may not be correct, then. The whole ātable 1, 2, 3ā thing is a misnomer based on the way they were stated to do the pulls. Sirrian specifically states here that they pull from the last table first when determining rewards. LTs have gone through a couple reward shifts since then with the runics being added and now with gems and minors being removed, this should still hold true. Therefore, a mythic at 0.5% on the last table would equal a 0.5% chance to get a mythic in the task, period. Your sample would then still be an outlier, but not nearly as big of one.
If you have some data on what the other tables might look like, specifically gem keys, event keys, glory keys and glory, Iād love to see them. I am particularly interested in what ~3.9 LTs will, on average, get us in mythic hunting resources, as this equates to the amount of gold we would have to put in to LTs after finishing Guild Tasks each week to equate to 25% more gold being put into tasks total (the amount of total more gold reportedly entering the economy that the 20% Guild Task reduction was allegedly done to compensate for). If you happen to have the old LT values saved somewhere for comparison purposes, that would be helpful as well.
1 Like
Thanks for that link, helps a lot. I think that thread is from right around when I started playing, so I donāt think I saw it.
Yeah, I have the new drop tables, as well as a copy from 3.0.5. Iāll crunch some numbers and post everything tomorrow (itās almost 3am here).
1 Like
Old LTās (3.0.5):
Type | Table 1 | Table 2 | Table 3
- | - | - | -
Gems | 20% for 25 | 10% for 35 | 10% for 60
Glory | 10% for 100 | 10% for 300 | 10% for 500
GloryKeys | 20% for 10 | 20% for 15 | 10% for 25
GemKeys | 20% for 2 | 20% for 4 | 30% for 6
EventKeys | 10% for 2 | 10% for 2 | 20% for 4
MinorRune | 5% for 20 | 10% for 30 | n/a
RunicRune | 5% for 5 | 10% for 8 | n/a
ArcaneRune | 5% for 2 | 5% for 4 | 10% for 6
LegendaryTroop | n/a | n/a | 9.5% for 1
MythicTroop | n/a | n/a | 0.5% for 1
New LTās (3.1.0):
Type | Table 1 | Table 2 | Table 3
- | - | - | -
Glory | 10% for 100 | 10% for 100 | 10% for 750
GloryKeys | 30% for 4 | 30% for 4 | 10% for 50
GemKeys | 40% for 1 | 40% for 1 | 30% for 8
EventKeys | 10% for 1 | 10% for 1 | 20% for 4
RunicRune | 5% for 5 | 5% for 5 | n/a
ArcaneRune | 5% for 1 | 5% for 1 | 20% for 6
LegendaryTroop | n/a | n/a | 9.5% for 1
MythicTroop | n/a | n/a | 0.5% for 1
As for what you would expect from 3.9 LTās on average (not quite sure what the 3.9 is or how it was calculated?), it would be:
Type | Expectation
- | -:
EventKeys | 4.15
GemKeys | 11.45
RunicRune | 3.11
GloryKeys | 29.27
LegendaryTroop | 0.37
MythicTroop | 0.02
Glory | 402.23
ArcaneRune | 5.20
And for completeness, hereās a corrected analysis of my earlier post/calculations with the last table selected first:
With LTās the expectation is a mythic every 200 times. In a full guild of 30 members, that translates to 6.67 million gold each. Spent on gold chests, that would open 22,222 chests, for an expectation of about 333 gems. Used on VIP chests, which have the best mythic drop rate, it would open almost 7 chests giving you a less than 7% likelihood of pulling a mythic. So just on mythic probability, LTās are about 15x better (at the cost of not being able to target a specific/new one).
8 Likes
Forgot to post this, but when I was analyzing, I found it interesting to compare the old/new average expectation for a single LT. So here it is:
Type | 3.05 | 3.10 | Diff
- | -: | -: | -:
Gems | 15.15 | 0.00 | -15.15
Glory | 92.64 | 103.14 | 10.50
GloryKeys | 7.48 | 7.50 | 0.02
GemKeys | 2.73 | 2.94 | 0.21
EventKeys | 1.42 | 1.06 | -0.36
MinorRune | 4.72 | 0.00 | -4.72
RunicRune | 1.24 | 0.80 | -0.44
ArcaneRune | 0.93 | 1.33 | 0.40
LegendaryTroop | 0.10 | 0.10 | 0.00
MythicTroop | 0.01 | 0.01 | 0.00
3 Likes
The recent reductions in the overall economy were stated to have happened due to an overall increase in the influx in gold of āabout 20-25%ā. A guild completing all their Guild Tasks would have put in 15,660,000 gold weekly. The same guild, now getting 25% more gold, is theoretically capable of putting in 19,575,000 with an equal amount of effort. Therefore, they are now capable of doing 3.915 legendary tasks per week. Iām asserting that a 20% reduction in rewards in order to ābalanceā with a 25% inflation is not mathematically sound reasoning when you have sharp diminishing returns for how much gold you have put in throughout. It paints a picture to the casual observer that the you are intended to get the same rewards for the same amount of effort on average (or for a not mathematically proficient observer, that you might even be getting more rewards for the same amount of effort). I just want to see what the same āeffortā is expected to get you post-patch versus pre-patch, and how much additional āeffortā you are expected to exert to get back to a similar level. I also want to see if I can calculate what āvalueā they would need to have assigned to the relative rewards in Legendary Tasks in order for the assertion of a 16% value reduction at legendary task level to be true. Iāll run some comparisons later.
1 Like
Another thing to note is that in the course of those 200 tasks, the mythic isnāt the only reward. Youād also end up with around 500 gem keys, which have their own decent chance for drawing a Mythic. Youād also get ~1k glory keys and ~175 Event keys, which have a chance as well, albeit much smaller.
Very true, and I mentioned that in my earlier post. Iām simply focusing on that for simplicity, since even if you received nothing other than the mythic chance, the LTās would be superior to gold chests. So effectively all that other stuff and the chances they give you can almost be seen as a bonus.
Right, but bonus with additional not-insignificant Mythic chance.
1 Like
Wow, thatās actually surprising! Everyone is losing their minds over the removal of Gems but the rewards are better on average across the board and no more minors. All in all, not as bad as it first seemed.
One compromise could be to use the two āsmall rewardā tables for gems and replace the runics. It would probably only be a couple gems, but that seems like as much as could be expected.
Not sure thatās how Iād interpret the data. While the traitstones have slightly improved, the Slight increase in gem keys is still only about half the loss in event keys, which makes the (most valuable) gems basically a straight loss. Personally, Iād rather have baby traitstones back in the table if it meant I could have the gems again.
2 Likes
Well the Glory and Arcanes also have value to consider. On the whole it seems like everything balances out with the exception of the elimination of Gems. It also seems like their % reduction in value wasnāt far off given so much of the value is in the Legendary and especially the Mythic. Though obviously the 15% they hit was very popular.
I think replacing Runics in the smaller tables with a couple gems is a good compromise personally. Itās not a huge buff but it could still mean a couple hundred gems per week for the super high LT completers.
This is how I feel as well, but I wanted something to compare. I decided to look back at my guildās LT history over the previous month for some averages. For the sake of math I converted Gems to Gem Keys at 10:1.
Date - Average Event keys per task - Average Gem keys per task - Average Arcanes per task
8/14 - 1.41 - 3.71 - 0.35
8/21 - 1.93 - 3.21 - 0.71
8/28 - 1.90 - 3.50 - 1.40
9/4 - 1.33 - 2.09 - 1.82
Obviously the current week isnāt over so the 9/4 numbers are subject to change (though folks who tend to dump gold have already done so). What is immediately obvious to me is that Gem keys went from a 3-week average of 3.47 per task down to 2.09, which is a 40% reduction. Arcanes seem to have received a slight buff, but I have no idea what the Gem value of an Arcane is since you can farm as many Arcanes as you want from explorations and you canāt farm unlimited Gem Keys.
Iām also not accounting for Minor/Runic traitstones or Glory keys. These things are all infinitely farmable, so I donāt know how to value them. If someone has suggestions so I can get a better idea, Iām all ears.
1 Like
So, lets make some projections about just collecting mythics. Legendaries and Runics and Arcanes are goals that are likely to be completed before collecting all mythics if you started post 2.1 (or now post 3.1) because you get them incidentally due to the gigantic amount of keys you have openend and time you have played to achieve this goal if you need to collect retroactively. This, arcanes and runics are effectively zeros toward this goal, as are legendary troops.
To do this, I need to define Mythic Purchasing Power (MPP). I define this as the overall contribution or chance toward getting a new mythic.
From the table, for relevant mythic hunting items per LT, we have about:
7.5 glory keys - 0.00075 MPP
2.94 gem keys - 0.00294
1.06 event keys - 0.00106 MPP
103.14 glory - 0.000573 MPP (converted to event keys)
0.01 mythic troops - ~0-0.005 MPP (varies depending on how many you need to collect and how many are in the game)
For a total of 0.005323 - 0.010323 MPP per LT. As you can see, the relative value of a legendary task is going to vary a lot dependent on how many Mythics you currently have, since the ability to get a duplicate from the legendary task will severely diminish its overall value. In a perfectly balanced guild, 1 mil gold put toward legendary tasks while everyone else also puts in 1 mil gold nets ~0.15-0.30 MPP depending on your VIP level and level of progression, but those arenāt realworld conditions. However, at minimum, you need to be shouldering at least 250k of the cost of every single legendary task in your guild before you consider pulling further donations and them toward guild keys. 250k of each task, total, not just the ones you personally donated to, but an average of all of them over an extended period of time. And you need to be donating the full 1 mil, by yourself, for gold keys to be better by a very thin margin if you are non VIP and donāt have any mythics.
I feel I canāt stress these last point enough. I canāt think of any real-world scenario where gold keys would be āworth itā above legendary tasks, even post nerf.
Edit: revised for half a percent Mythic drop chance rather than 1%. This does not affect my lower bounds worst case scenario calculations, which assumed all current mythics collected and therefore ignored this drop entirely.
3 Likes
To all of you who have done amazing, wonderful things with your math skills:
Thank you!
This thread has turned into so much more than I ever could have imagined! 
Iām going to slowly close the door now and leave quietlyā¦
Your Mathematical Wizardy scares me! 
3 Likes
That would be a slap in the face of they adjust the gem drops in gold chests.
1 Like
After all the recent slaps, whatās the harm in receiving another, eh?
2 Likes