Thread: [Event 32] Crucible of Combat vs The Demon Gluttony (Union Event)

Originally Posted by Coconut

I'm in a similar position, but it's for an alt that currently has 16k jewels, Gaia, and Titania. The way I see it is a bit like shopping for a computer - there's always something better coming. In this case, rolling for Seth gives you a shot at Seth, Cybele, Hraesvelg and Hastur, but rolling for Cybele only gives you a shot at 3/4 (but with an appearance bonus). And if your rolls are duds, or you get good rolls but for a different element, then you might've been misdirecting your farming.

So I'm a bit tempted to roll for Hastur, knowing that I'll have more jewels when Cybele releases. Is this wrong?

This is a mathematical problem with an optimal solution. More options for 'hits' doesn't necessarily equal better odds, as inbetween Cybele and Seth are kamihime of other elements, which can be considered 'misses'. So if the number of misses added is proportionally greater than the number of hits, you're better off rolling earlier.

You can actually calculate all this using the Gatcha calculator in my Toolbox.

Originally Posted by sanahtlig

This is a mathematical problem with an optimal solution. More options for 'hits' doesn't necessarily equal better odds, as inbetween Cybele and Seth are kamihime of other elements, which can be considered 'misses'. So if the number of misses added is proportionally greater than the number of hits, you're better off rolling earlier.

You can actually calculate all this using the Gatcha calculator in my Toolbox.

Okay, so if I used the correct data...

Desired - odds:
Hastur only - 0.047/chain
Hastur, Cybele, Hraesvelg - 0.057/chain
Seth, Cybele, Hastur, Hreaesvelg - 0.046/chain

Leventale would probably put Gaia on his desired lists and so would have different results.

Originally Posted by Coconut

Okay, so if I used the correct data...

Desired - odds:
Hastur only - 0.047/chain
Hastur, Cybele, Hraesvelg - 0.057/chain
Seth, Cybele, Hastur, Hreaesvelg - 0.046/chain

Leventale would probably put Gaia on his desired lists and so would have different results.

One wrinkle in this is that the calculator assumes that all characters in the same rarity+category tier (i.e., SSR eidolons) have the same base draw rate. With the +100% eidolons, this assumption probably does not hold true; the draw rate for these (assuming you've landed an SSR eidolon) is probably much lower than 1/(total number of SSR eidolons). The draw rate penalty may be as large as an additional 1/5 or 1/10. I currently have no reliable data to estimate the exact penalty, however.

In short, don't include +100% eidolons as a 'hit' or 'miss' if you want a valid result from the calculator. They can only be validly compared amongst themselves right now.