2* Farming

Quebbster

shardwick said:

Daredevil217 said:

aesthetocyst said:

smkspy said:

... is a 3 star farm worth the investment ...

Asked and answered several times on this forum.

The higher the tier, the better the numbers.

But, the higher the tier, the longer the timescale.

Double but, since neither iso nor HP can earn interest, or be borrowed at interest, they have no discount rate, no timevalue to consider.

Therefore, it is worth farming all characters that offer champ rewards, no matter how long the cycle, unless it's going to require roster slots, you'd have to use real world money to acquire the slots, and you can't afford to set money on fire.

If you can afford the slots from in-game earnings, or don't mind burning cash, farm'em all.

I think the net iso loss after all is said and done is only like 7K for flipping a 2*.  A DDQ worth of iso is worth it for all the rewards.

What is the net iso loss for a three?

When you take into account the iso you get from the champ rewards for that 2* it's not really a loss though. I think it ends up being a net positive of around 4500 iso. It really only feels like a loss since the price to champ the new one is more than the price to sell the maxed old one.

It is still a slight loss compared to just selling the 50 two-star covers outright. The ISO champion rewards exactly match the cost of the champion fee and the ISO you would have made selling the covers.
The loss is minimal, but farming will set you back ISO-wise ever so slightly. The other rewards more than make up for it though.

aesthetocyst

Jaedenkaal said:

smkspy said:

I think iso restriction is a bigger negative towards 3 star farming than slots.

The difference between selling your max-champed 3* and leveling a 13-cover 3* to 166 and champing them is only 22,149 ISO...

Unless you are "post-rank", you will get 50-75% of that 'loss' back in XP, depending on what rank you are. And XP is worth more in terms of iso at lower ranks.

The other consideration is that players are more likely to overvalue their maxchamp 3s and hold on to them way too long. Even 5* players keep them in a deathgrip. This delays the receipt of 105K iso, 500hp, and the freeing up of a roster slot.

Dormammu

shardwick said:
When you take into account the iso you get from the champ rewards for that 2* it's not really a loss though. I think it ends up being a net positive of around 4500 iso. It really only feels like a loss since the price to champ the new one is more than the price to sell the maxed old one.

It's a loss. You're not calculating the ISO loss for the 63 covers it takes to max-champ a 2-star (13 to max cover + another 50 to max-champ).

Nick441234

Quebbster said:

Nick441234 said:

sinnerjfl said:

farlus said:

The  most prohibitive thing about these solutions is that building a dupe for each 2* before the main is done requires another 1k HP, which isn't realistic for most people who are 3* to 4* transitioners or even for the 4* to 5* transitioners.

Waiting the 2 weeks after hitting 144 to get 13 covers is more viable, but as of late even that hasn't worked. In fact, with 2* Bullseye, in 2 weeks after hitting 144 I was only able to get ONE purple cover. I sold him and rostered at 1/5/5 and hoped that before the remaining green and black covers expired I would get the additional two purples I needed. I ended up having to roster a second 0/5/5 dupe so that the covers wouldn't expire. Absolutely atrocious. I should've just done the former method at that rate.

If you've got the HP, yeah, roster a dupe before maxing. (This is what I do for 3* because they are much less farmable.) The latter method is less cost prohibitive, but you're always taking a risk of covers expiring.

From the champ rewards and selling off the champ 2*'s, you stand to make around 4500 HP. Its really worth it and yes it will take some roster space but in the end, you are better off doing it.

I also much prefer getting a 2* to 144, building the dupe, sell the 144 and re-champ immediatly. Much less tempting to spend the iso on something else that way.

Exactly my thinking. If I leave the sold ISO in my pile too long, I'll put it elsewhere and end up having to find more ISO to champ the 2*. I wouldn't enjoy having to find ISO just to champ a 2*. 

It's only 65K ISO to champ a two-star, that's like two day's Worth of ISO if you play a lot.

65K towards champing another 4*. 

aesthetocyst

Nick441234 said:

65K towards champing another 4*. 

Sigh.

 The 65K comes back out the other end of the cycle. Look at it like ....

1. It's the same iso, cycling round and round, only costing you a couple thousand each cycle, OR ...

2. The iso for your 4*s is being funneled through 2* holding corporations. Or if you rather, certificates of deposit, or Christmas Club accounts. While the money sits there, you get paid dividends. When the dividends run out, you get almost all your iso back. At which time you can slap it into another 2* holding entity, or your current higher rarity leveling project.


Unless I have something pressing, I am always spending my iso on the low end of the bench and working my way up. Everything builds from the bottom up.

Jaedenkaal

Nick441234 said:

65K towards champing another 4*. 

Sure, but putting only 65,000 ISO into a 4* over level 150 or something barely gets you anywhere. The last 10 levels practically cost more than 65,000. Putting 65,000 into a 2* in one day (plus only a little bit more) gets you something meaningful. Putting it into an under-leveled 4* without getting that 4* to 270 gets you next to nothing. 

The only time you should save that ISO is if you've got that 14th 4* cover waiting, and it would expire before you could use it if you spent that 65,000 now. If you find that happening to you all the time... maybe slow down on opening legendary tokens a little. They'll wait. The value you're losing by not having that 2* champed today is gone forever.

carrion_pigeons

I start my dupes at 140 and I find that even if I have to hold onto some extra covers on the vine for a while, I have yet to need to actually sell one off.

[Deleted User]

Ive never bothered with dupes in 2*.  Once I get that first cover past 144 I sell up and re-roster.

Ive yet to have a problem this way.  2* nodes in PVE are easy, Ive always got characters good enough for DDQ...

Starfury

Nick441234 said:

Theres got to be a point where it stops being random though. Theres no way I should still have 7 at 144 at the same time as each other after as long as I've been rotating them. 

People always confuse randomness with even distribution.

Let's assume you're doing 4 pulls from a token with 4 characters with 25% chance each:

Your odds of pulling AAAA are exactly the same as pulling ABCD. (Both about 1 in 256) Neither is more likely or "random" than the other.

The trend towards an even distribution comes from the fact that among the 256 different outcomes, only one will have exactly 4 A, while there's 24 different sequences offering exactly 1 of each.

Nick441234

Starfury said:

Nick441234 said:

Theres got to be a point where it stops being random though. Theres no way I should still have 7 at 144 at the same time as each other after as long as I've been rotating them. 

People always confuse randomness with even distribution.

Let's assume you're doing 4 pulls from a token with 4 characters with 25% chance each:

Your odds of pulling AAAA are exactly the same as pulling ABCD. (Both about 1 in 256) Neither is more likely or "random" than the other.

The trend towards an even distribution comes from the fact that among the 256 different outcomes, only one will have exactly 4 A, while there's 24 different sequences offering exactly 1 of each.

Then why does AAAA come a hell of a lot more often than ABCD? 

Quebbster

Nick441234 said:

Starfury said:

Nick441234 said:

Theres got to be a point where it stops being random though. Theres no way I should still have 7 at 144 at the same time as each other after as long as I've been rotating them. 

People always confuse randomness with even distribution.

Let's assume you're doing 4 pulls from a token with 4 characters with 25% chance each:

Your odds of pulling AAAA are exactly the same as pulling ABCD. (Both about 1 in 256) Neither is more likely or "random" than the other.

The trend towards an even distribution comes from the fact that among the 256 different outcomes, only one will have exactly 4 A, while there's 24 different sequences offering exactly 1 of each.

Then why does AAAA come a hell of a lot more often than ABCD? 

Because you only notice it when you keep getting A's you don't need.

If you are serious about this, start listing all the covers you get, even the ones you apply as champion levels. If you still get lots of AAAA sequences... still doesn't prove anything, but at least you have a stronger case that something is up with the RNG.

Straycat

Quebbster said:

Nick441234 said:

Starfury said:

Nick441234 said:

Theres got to be a point where it stops being random though. Theres no way I should still have 7 at 144 at the same time as each other after as long as I've been rotating them. 

People always confuse randomness with even distribution.

Let's assume you're doing 4 pulls from a token with 4 characters with 25% chance each:

Your odds of pulling AAAA are exactly the same as pulling ABCD. (Both about 1 in 256) Neither is more likely or "random" than the other.

The trend towards an even distribution comes from the fact that among the 256 different outcomes, only one will have exactly 4 A, while there's 24 different sequences offering exactly 1 of each.

Then why does AAAA come a hell of a lot more often than ABCD? 

Because you only notice it when you keep getting A's you don't need.

If you are serious about this, start listing all the covers you get, even the ones you apply as champion levels. If you still get lots of AAAA sequences... still doesn't prove anything, but at least you have a stronger case that something is up with the RNG.

A watched pot never boils. You aren't noticing the champ levels, the 50 it takes to go from 94-144 for 14 2*s. I recently had to build a dupe 2* Hawkeye because I was only pulling his purple cover. If he was champed those would all be applied without a thought, but since I was rebuilding him it stuck out.

Dormammu

If you flip a coin five times and it comes up as heads every time, your mind thinks, 'no way! what are the odds???' But really, the chances of that happening are the same as any other result. There is a 50% chance of heads every time you flip it, but our brains trick us into thinking that 50% chance is reduced with every sequential flip that comes up heads. It's called Gambler's Fallacy.

It's such a potent trick that our minds play on us the idea that the game is somehow rigged becomes more believable. Because obviously the developers made special code for me in specific to receive more of one color 2-star cover than the other two.

animaniactoo

I did track this at one point because I was just starting up my 2* farm, I'd emptied the DDQ vault AND then had 2 weeks worth of 2* covers for the characters I had not already champed.

So I did what any sane person would do. I made a spreadsheet for the purpose of figuring out how many of each I had in my 130+ covers on the vine, and therefore which ones made the most sense to champ vs cashing in the covers.

Absolutely there was an ebb and flow of which characters I was getting and to some extent even what colors I was getting for them.

I was able to draw some conclusions that I think make sense given game dynamics. 2*s that were upcoming as essential were more likely to be drawn. They were also more likely to be drawn for their "best" color. Second to that were 2*s that were upcoming as weekly boosteds.

Everything else was a relatively random distribution.

Since there isn't an exact rotation for essentials and boosteds (due to overlap issues, etc.) over time, I have found that it's impossible to accurately stagger champing and rebuilds. Therefore, I don't even bother.

I have 2 copies at all times. 1 is the fully champed because I hate the concept of playing with an underpowered character when I've achieved max power at some point, and I have the HP for roster slots so that's my personal bugaboo works best for me. 2nd is the rebuild. As soon as a rebuild is fully champed, I sell it off and put the ISO into whoever is ready to be champed. Then I re-roster the rebuild and start collecting covers for them. By the time I hit 5 covers in any of their colors, it usually works out so that I can get them fully covered and champed in the next 2 weeks after that. If I have to sell a cover or 2 now and then I don't much care. I'd rather do that than spend the time to track the progress of the 2nd copy to time doing the 3rd.

ETA: In a pinch, when I had to hurry up and champ a 4*, I've also raided the farm for the fully champed versions if their rebuilds were only a couple of covers away from max champ level. Makes for a good "loan" source.

Jaedenkaal

animaniactoo said:

I was able to draw some conclusions that I think make sense given game dynamics. 2*s that were upcoming as essential were more likely to be drawn. They were also more likely to be drawn for their "best" color. Second to that were 2*s that were upcoming as weekly boosteds.

Guaranteed, neither of these things are actually true, even if those patterns happen to fit your observations. 

Forcing random events to fit a pattern with personal meaning is yet another classic human logical failing.

animaniactoo

Jaedenkaal said:

animaniactoo said:

I was able to draw some conclusions that I think make sense given game dynamics. 2*s that were upcoming as essential were more likely to be drawn. They were also more likely to be drawn for their "best" color. Second to that were 2*s that were upcoming as weekly boosteds.

Guaranteed, neither of these things are actually true, even if those patterns happen to fit your observations. 

Forcing random events to fit a pattern with personal meaning is yet another classic human logical failing.

I'll take an opinion of unlikely, but not a statement of guaranteed unless you've had some confirmation from D3/Demi previously?

It would not be hard to temporarily increase the draw rate, and it would make sense to do from a game perspective. I've noticed it enough in other places that I think it goes beyond confirmation bias. I could be wrong, but I won't accept that as an absolute unless you've got some official confirmation on your end.

Dormammu

Enough people have cracked into the code to confirm it is entirely random.

sambrookjm

Another benefit of farming - I just sold 4 max champed 2-stars to prevent two 4-star covers from going to waste.  Now I've got below average 2-star Thor, Black Widow, Daken and Captain Marvel, but my Kate Bishop and 4* Rocket and Groot are now each at LVL 271.  Any extra ISO left over is going towards champing 3* Gambit, for whom I also have a cover in the queue.

rocks52

The only thing I wish (which I know will never happen) is that each 2 star feeds 3 3* characters.  There's what 47 3* characters and 13 2* (I know it's 14 but does anyone really count bagman since he's so hard to get.  So that means you can more easily attain champ levels for 13 3* characters while 34 are left out in the cold.  Although oddly enough I've sold off a few maxed out 3* characters and only a couple of those were actually ones being fed by a continuously farmed 2* character.  

sambrookjm

rocks52 said:

Although oddly enough I've sold off a few maxed out 3* characters and only a couple of those were actually ones being fed by a continuously farmed 2* character.  

The only maxed-out 3 star I've sold is Cap, but the next highest 3-star on my roster is, for whatever reason, Blade.  No idea why I get a lot of his covers, since he's not even my bonus hero.