Elder/shaped maps corruption outcome is bugged/unfavorable

A while back, elder maps could corrupt into various outcomes:

- unidentified

- 8 mods

- reroll to 4-6 mods

- nothing happens

- transform the map to another elder map adjacent to it - THIS was later removed since it was deemed pointless

In consequence, unidentified roll had something like 20% chance. It was empirically proven by corrupting hundreds of maps.

The problem is that unidentified corruption after the change still has 20% chance, which means that the chance that was used for transformation wasn't uniformly distributed to the outcomes that remained, but rather went to the chance of bricking, chance of nothing happening or chance of getting 8 mods. I haven't counted 8 mods since I specifically monitored only the unidentified outcome and out of 200 maps, approximately only 20% of them turned unidentified, rather than the desired 25%.

I'm not sure if this is a bug. If this was a design decision, please move it to feedback forum and I'll edit it accordingly to highlight that we want 25% chance rather than 20%.
Last edited by demonstorm on Feb 17, 2019, 5:26:06 AM
Last bumped on Apr 8, 2019, 6:02:53 PM
This thread has been automatically archived. Replies are disabled.
Actually, on all maps, it is empirically proven that corruption has:

25% chance unidentified

25% 8 mods

25% nothing happens

25% reroll with a 50% chance to gain one tier (with further 50% chance to be a vaal temple if you corrupted a non shaped? T15), since a T16 cannot gain a tier, it's ignored.

Far from your numbers.
Don't bump your post; if nobody confirmed you or posted something similar, you're probably wrong...
"
Mmimich wrote:
Actually, on all maps, it is empirically proven that corruption has:

25% chance unidentified

25% 8 mods

25% nothing happens

25% reroll with a 50% chance to gain one tier (with further 50% chance to be a vaal temple if you corrupted a non shaped? T15), since a T16 cannot gain a tier, it's ignored.

Far from your numbers.
Don't bump your post; if nobody confirmed you or posted something similar, you're probably wrong...


So by your odds, if last case gets ignored, then I should still have 25% chance to get an unidentified map, right? Which is definitely not the case. A lot of people over reddit confirmed that getting unidentified bonus occurs only in 20% of cases, not 25%. I collected data from 1000 maps since I posted this and I can confirm, based on this empirical data, that the chance is 20%.

I'm going to bump my post however I like until a moderator states otherwise, thank you :)
Hi!

I did some calculations for you! Corrupting 960 Siege maps and 960 Elder Siege maps each results in rolling unidentified 25% of the time, give or take 2% for variance.

Everything seems to be in working order. Good luck with your maps!

Report Forum Post

Report Account:

Report Type

Additional Info