CORRUPTION vs LEGENDARY POTENTIAL (PROPHECIES)?! | Last Epoch 1.0

[@FrozenSentinel]

Hey folks, I just want to clarify this video does not, and is not intended to, definitively prove that LP is not affected by corruption, the sample size needed to have sufficient confidence would be massive and as an individual I am not able to do that. This is just my own sample that is large enough to give an idea of expectations, and shows that even with a fairly large sample, small differences will be hard to notice by individuals

[@OneOnDruryLane]

EHG Dev here - this is amazing work! I'm such a numbers nerd too, I love this. I've shared it internally too =)

[@FrozenSentinel]

Hey Muffin, thank you!

[@Uli_Krosse]

Nerdy as I don't know what; cold, hard facts. Yes, you have earned that sub. Thank you for doing this AND for showing how it can be done.

[@nirvcoregaming]

I normally never like videos but the amount of effort you went through deserves it

[@OttoIpsen]

Ppl can say what they will about your video but the effort and the work you put in is amazing! Really good work sir!!

[@junk3996]

algorithm reward this man!

[@3lectr0o31]

Good effort bro, diving deep into games and collecting the knowledge gives me so much pleasure xD thanks for doing the work tho

[@doctort2853]

Thank you for the detailed analysis!

[@MaxxxMaxMaxx]

I was close to having a mental breakdown just now, realizing I might have wasted probably 1 million favor because many of my prophecies are in dungeons and im normally running corruption 1200 and dungeon mobs are not affected by that. Then i remembered i saw this video 4 days ago and went back to rewatch it. You're a legend. Thankt for taking your time with these kind of videos.

[@FrozenSentinel]

:D

[@SmthIcanNvrHave]

But dungeons don't give you extra favor?

[@MaxxxMaxMaxx]

@@SmthIcanNvrHave correct, I try not to run dungeons at all since the exp(favor) is terrible. But after a while I've collected a few and I do a run or two. My concern was about if the LP chance is lower and therefore making those prophecies even worse than I first thought, not just the reduced favor from mobs.

[@Bridgeburner56]

Awesome video, really appreciate the effort that goes into gathering this kind of statistical data (t-test my beloved)

[@GhostwarGWPTC]

Putting in the work 🙏 God tier Information

[@arnonw1513]

Thank you for this amazing thorough analysis!

[@XxNardiusxX]

I had same assumption cuz u can get t7 affixes from prophecies even doing campaign but to drop it from monster it needs to be at least 90 lvl, anyways great job, as always thx for your in depth calculations and time you spend making them :D

[@dcruzsteven]

Kudos on the epic empirical analysis

[@TenelenTheScrub]

Fantastic video with some impressive legwork put into it! I had been thinking myself about trying to do something similar, but I'm glad you got ahead of me!

[@malapias]

Great video as always. Keep up great work.

[@j6474]

Huge, thanks for the info. Fantastic video as per usual!

[@KokoMakesThings]

This is amazing. Absolutely fabtastic numbers work, gunna share it around with the playgroup, thank you!

[@mcjudilla774]

This is what my boss expects me to do everytime I tell him my performance is good.

[@FrozenSentinel]

omg performance review season

[@artemrusskikh2796]

I like your LE content, you have my subscription

[@paganizertube]

Once again. Amazing work.

[@eliel77]

Amazing work ! Thanks for sharing it. And another tumbs down for CoF players. The player perception of some mechanics not being rewarding enough or to have bad design being reinforced.

[@nergal2741]

Great Job! Also surprised I was this close to the actual number of dropped rings ;D

[@FrozenSentinel]

haha true! good guessing

[@tom-nl7fg]

WOW u farmed soo hard my respect for ur hard work .....

[@phantombyte00]

Super useful, thanks a lot!

[@pozytywniezakrecony151]

Thanks for commitment :D

[@CapitanBreakfast]

Good work man, please don't burn out though and keep up the good work!

[@FrozenSentinel]

I hope I don't but LE has me the most excited I've been to finish work and hop on a game in years, it's like being a kid again.

[@therissuishere]

I think I have spent roughly 700k on ring prophecies and still haven't got the Red Ring, and seeing you got it drop twice, gives me hope.

[@therissuishere]

I have kept all the 3lp rings. I've gotten 12 Tongues, 8 Thirds Eyes, 4 Sunwreaths, 3 Quicksilvers and 1 copy of Hollow, Liath and Heirloom......so far. I've done them all on ~600 corruption. If you got any pointers on how to increase my chanceson getting Red Ring, I'm all ears :D

[@lorzkotzros]

Got my from Rune of ascendsnce, i used around 40​@@therissuishere

[@FrozenSentinel]

In total I've dropped Red Ring 4 times (so 8 rings with Rank 10), small sample size but it seems like every 1 every 400 or so rings is a RR so far.

[@FrozenSentinel]

Keep in mind ascendance seems to have reduced LP chance.

[@lorzkotzros]

@@FrozenSentinel Well yes but with the Chance of what is it, 6%?:D doesnt matter anyways lol you didnt find one with LP didyou

[@tetsero]

Didn't know grass was green, thanks!

[@Toby40]

Absolute legend, Devs should put you on the pay roll 🎉

[@boukimalice7909]

Wow.. amazing work.

[@Zhezyx]

Amazing work , thanks a lot.

[@AleryZz]

People playing ARPG are different haha

[@FrozenSentinel]

Wouldn't have it any other way :)

[@AleryZz]

@@FrozenSentinel Me neither, I love matching it (RegularSchmuk content is the amazing for that)

[@KnightArt708]

Nah this guy just built different 🤣

[@kunchen3278]

the ancient god in charge of spreadsheets

[@Hofodikfo]

Fantastic data .. youve earned my sub. Id be curious now on corruption and how it effects amounts of tier 7 loot.

[@MrDarkBoy93]

Sad outcome but great analysis. Also highlights how rare 4lp is. You got 1 or 2 out of 1k.

[@redisk]

Ok, this is effing awesome. Now I'm going to have to start clearing my stash so I can switch to CoF. I may have at least 60tabs full of uni gear with another 15 of lp... I'm the problem...

[@willowcompany]

Doing God's work 👍👍👍

[@RyouuGyn]

Nice work. I dont remember where a I read it but prophecies are not affected by corruption. You can do wathever prophecies you want (arena, monolite, etc) and the chances will be the same. Corruption only increase Rarity wich means more affixs (more chances for T7, double T6, etc) and drop chance for Uniques and after 1.0 slighty for LP too. Anyway, you always bring nice analys. Pliz, continue. Here's my like ↑↑↑

[@avndoon]

Insane work! Thank you :)

[@marcomeinhof5393]

Great work

[@adiozerg117]

this is great thank you!

[@Matthew-sj1zw]

great data!

[@skavgosu]

Would be interesting to check if something similar is happening with exalted. Their quality seems to scale much faster than LP with corruption, which would make the prophecies for them comparatively less valuable on high corruption.

[@FrozenSentinel]

The exalt prophecies are already pretty cost inefficient but yea I know what you mean.

[@rcglinski]

Fantastic.

[@preheet]

My theory has been for a long time that prophecies proc'd in high corruption has 0 effect. Best practice is to get a lot of favor at high corruption, buy a lot of prophecies for X item (ex. rings) and just let them proc for a better chance at getting high LP uniques since the more uniques you have drop, the higher chance at finding high LP. At 2k corruption I can confidently say that the only thing high corruption is good for it how fast you can farm favor/gold, leading to more porphecies per hour which indirectly affects your chance at getting high LP uniques. Wish there was some additional benefit for prophecies completed at high corruption :(

[@FrozenSentinel]

Yea that was pretty much my conclusion. I am also noticing the rank 6 does make a noticeable difference for boss drops, but still a grind for LP drops.

[@preheet]

@@FrozenSentinel Yeah it's quite crazy to imagine I haven't seen a 2LP Twisted Heard after probably 300+ kills at 2000 corruption haha, but lots of 1LPs. Praying for some tweaks in the upcoming cycles. Great vid and thorough work as always sir

[@str8totehtop]

good work

[@jjjstl1630]

Here's something I noticed. I was farming Exsanguinous and hoping to get a 3LP one. I spent around 700k favors on prophecies. Sometimes I would pick dungeon prophecies since they can be done very fast. And I do notice that I got tons of 0/1LP from dungeon prophecies comparing to the ones I got from 700ish corruption lvl monolith.

[@FrozenSentinel]

yeah I probably spent at least that amount too on exsanguinous for my warlock, same story, no 3LP

[@jjjstl1630]

@@FrozenSentinel I've also never seen any 2LP twisted heart/shackle/Last Steps of the Living after farming on 2 chars at 800 corruption lvl. CoF really needs some buff when it comes to farming these boss items.

[@harrybenson6731]

Hi, here's perhaps an interesting question: You know how sometimes you get a monolith map with tons and tons of rare monsters that give much more experience (favor) and many more exalted items than your everyday map? I always full clear these maps, they're great. But what timeline has the most of these blessed maps? Currently I'm farming fall of the outcasts timeline because I believe this timeline has an above average rate of blessed maps. I believe there is a significant difference in experience/item gains between the different timelines.

[@FrozenSentinel]

I've been wondering the same actually, not 100% sure if it's based on the map itself or any map can spawn like that, will need to do some testing.

[@SonicBoom474]

Thanks for the great effort! What about 100 corruption vs 0 corruption? The probabilities shown on LEtools correspond to no corruption?

[@FrozenSentinel]

I didn't really see the point in testing this as empowered starts at 100. Also non empowered monos are less than lvl 100 areas which also factors into the LP formula.

[@Oroo420]

One of the things I have noticed is I did not start getting t6 and t7 +skill on specifically relics until I raised corruption. 3 weeks without a single Frost Claw affix drop let alone any other +skill. Soon as I raised Last Ruin's corruption by just a little bit (200), they started dropping. I dont know if I was just unlucky or if it is locked behind higher corruption. Originally I didn't want to push until I had everything fully functional but turns out that may have been a big mistake.

[@FrozenSentinel]

I don't think affixes are "locked" it just seems they drop more frequently as the "rarity" bonus increases. That affix specifically (and other similar ones) are extremely rare, even at 1k+ corruption I've only seen T6/T7 +FrostClaw 3 or 4 times.

[@Freedom-kl8rt]

I don't know, yesterday I got plus 3 for frost clow in normal monoliths and several plus on

[@Oroo420]

@@Freedom-kl8rt I guess I'm just unlucky. :(

[@oFiGuReZo]

So basically you don't get rewarded for you're hard work by doing the only endgame option. Higher LP chance are a joke right now and unfairly balanced for many items.

[@FrozenSentinel]

You get significantly more favor, but otherwise yea it doesn't seem like a lot of things scale.

[@junk3996]

so 1 in 331.89 billion red rings will be a 4 LP. that would take, what, several life times for all LE players?? i want to see someone have one

[@FrozenSentinel]

I'm sure some folks in offline have them ;)

[@flutta_]

Someone on reddit got one

[@FrozenSentinel]

offline though? xd@@flutta_

[@brunowalkowiak3432]

how you liking that frostclaw build in 1000+?

[@FrozenSentinel]

Great, having a lot of fun with it. Can't wait to push further as I continue to gear up

[@brunowalkowiak3432]

@FrozenSentinel how high you think you can take it?

[@FrozenSentinel]

However far I can go I guess lol, 2k is the goal for now, hard to push that far as CoF but should be doable.@@brunowalkowiak3432

[@alexossan2829]

What about Prophecies that require higher corruption? Do those tend to give higher rarity (i.e. more LP) than Prophecies that require no corruption?

[@FrozenSentinel]

Doesn't seem like it

[@onavelzy5537]

total repect!

[@TheRealTrucido]

So you're saying Corruption is bugged.

[@FrozenSentinel]

At least for prophecies it doesn't seem like corruption matters, but that doesn't mean the same is true for normal non prophecy drops.

[@TheRealTrucido]

@@FrozenSentinel Ah true. I was hoping the 1000% increase in rarity would help get rare items but it would make sense that prophecies don't increase rarity wherever theyre found. I did notice a ton of 3 LP low level rings when I was farming for a red ring via prophecies though.

[@FrozenSentinel]

Yea I am starting to realize corruption synergizes extremely well with Rank 6 CoF that doubles LP chance. Rarity means the monsters seem to drop uniques constantly, then almost always they have at least 1LP. @@TheRealTrucido

[@Hatchet-w6x]

Whaddya got there? ..numbers?

[@GerdionXT]

You are working as analyst/IT ?

[@FrozenSentinel]

I do work in Software yes

[@Phearless7485]

Why not just add up the LP for every ring in your sample size. So if 300 rings at 100 corruption gives you 350 LP, and 300 rings at 1000 corruption gives you 390 LP, we can at least say, on average you get more LP but it isn’t game breaking. Seems to me that if the roll is doubled on LP1 for tier 6 league bonus, then that 20% for 1000% corruption is probably additive to that 100%, then seemingly you’d have diminished returns for each LP roll. Seems to me it would make less of a difference for LP2-4 but on average gives a higher outcome. Cutting out all of what you considered “outliers” is probably what gave you the results you got. Just add up the total LP and let the averages added up over 300 rolls give the answer.

[@FrozenSentinel]

Because all the rolls aren't weighted the same. a 2LP + 0LP is much luckier than 1LP + 1LP despite adding up to the same number. Also, a 1LP on a 60LP level is much luckier than a 1LP on a 20LP level. This would give us very skewed data. If I didn't cut out the "outliers" all the "expected weightings" for something that is unexpected to actually happen given my sample size, would skew the result a lot. For example, 3LP Arboreal Circuit 1/3700 chance, but also a 3700 weighting if it drops, so the impact it provides the to the expected weighting is the same as 1LP or 2LP, however the distribution of actual outcomes will actually be way higher depending on whether or not I get it. Another example is that I did actually drop a 4LP Third Eye in my 100 Corruption sample which is a 1/1150, I could include that the weighting would be over 3x my actual sample size, not statistically significant.

[@Phearless7485]

@@FrozenSentinel It has been a while since I sat through statistics class, it would seem you are doing what is technically the correct procedure. Course I always thought outliers were things outside the norm of possibility, where these occurrences, though rare, are designed to be that way, and thus not really an outlier in the way I think of an outlier. Sort of like winning the lottery could be considered an outlier, however, it is designed to produce very rare winners, so if you cast out the outliers, then the summation would be that not a single person ever won, which isn’t true. I know you didn’t cast the outliers out, but rather downgraded them and did other things to weight them it, I’d just find it more interesting to give a quantitative number. If you add the total LP of each ring then divide by the number of rings, you should get a probability score that is representative. I dare say though that without the CoF rank 6 ability, the 1000 corruption bonus would likely be more impactful, but that is just a guess.

[@FrozenSentinel]

Going with your lottery example, let's imagine there is a 1 in 1M chance to win 10M$, and there are 1M people playing. "On average" you would get 10$ per play, but the sample size needed to approach anywhere near that would be massive. If we were to both test this lottery system, and each perform 1K tests, you may actually win the 10M$ and conclude that "on average" you get 10K$ per play. Then I may run the test and not win the 10M$ and conclude that "on average" you get 0$ per play. You just can't accurately represent these occurrences in your results. That's why I excluded them from both the results I actually got AND the expected outcomes. Just averaging all the LPs as you suggest would not control for the different LP levels of different rings, which adds an unnecessary additional variable to your results. If for example we both drop 100 rings, and I get 100 Putresence's which have a 1 in 4.8 chance for 1LP and a 1 in 57.8 chance of 2LP, and you get 100 Tongues which have a 1 in 2.8 for 1LP and 1 in 5.9 for 2LP, you will get a much higher average purely because of the additional variable you aren't controlling for. In this case, the additional variables add MORE variance to your result than the actual difference you are trying to test for. That said, I also don't have a stats background, I am just a programmer who often has to deal with data. @@Phearless7485​

[@AnoBananoSRO]

ive invested 620k favour into unique rings and not a single red ring :/

[@MrWitz90]

What's strange is that I've had 2 drop, but neither were fulfillment of prophecies, so they didn't duplicate.

[@FrozenSentinel]

Yea it's rough. I think in total I got 4 RR drops, that each got duplicated, also including outside this test. Probably 2M ish favor total on rings.

[@AyeKFortySeven]

I have over 5 but I use runes of ascendance as well and I don't even need the ring but I want an LP one

[@FrozenSentinel]

Runes of Ascendance have reduced LP chance for uniques given, so getting LP Red Ring with that would be pretty difficult.@@AyeKFortySeven

[@AyeKFortySeven]

oh yeah I know I have 2000 hours in LE I'm just saying I used runes to get some Red Rings that way these guys could maybe get one. @@FrozenSentinel

[@sylm5374]

So this means it’s pointless to grind high corruption

[@FrozenSentinel]

That is definitely not the case; the favor you get scales with monster XP, at 1k+ corruption you can get several thousand favor per echo easily just from monster kills. The LP chance boost does seem to also apply to boss drops, etc. which has helped me get many LP Twisted Hearts.

[@edwinhoekstra8988]

Great Work, Can you start off future video's with the conclusion - highlights, and then explaining your method.

[@FrozenSentinel]

To be completely honest, it takes me many hours or sometimes days to collect the data and assemble these videos, taking place in evenings after my full time day job. If I put all the answers in the beginning, it will tank the total view time since there is no reason to continue watching, which will mean my content doesn't get recommended, etc. Most creators do not provide all the answers at the beginning for this reason.

[@MurdersFPV]

Nerd.

[@FrozenSentinel]

tru

[@MurdersFPV]

@@FrozenSentinel I'm glad you took that as a compliment

[@Ludak021]

Bro, if you are doing 1000 corruption, you don't need more gear. Collecting gear for another character? Waste of time. New cycle comes, you do everything all over again.

[@FrozenSentinel]

I'm just having fun pushing a character, there wasn't really any point pushing to 1000 either but it's a video game and that's what I enjoy doing :)

[@miguelayati7706]

this is quiet bad... trade league so much better

[@FrozenSentinel]

It is unfortunate, for the "top" players who can play the game / market optimally, MG will probably offer better results in end game. For players who don't want to deal with the economy / market, CoF still offers a lot of really nice QoL vs pre 1.0

[@Freedom-kl8rt]

depends for whom..I like to find and craft my own gear..

[@eliel77]

​@@FrozenSentinelI agree that there are some good buffs and QoL, but still some bad design to CoF compared to pre 1.0. Exiled Mages drops rates (both mad alchemist and exalted experimental affixes) are trash. Boss drops are about the same insane grind trying to find something good.

[@thedevolis]

Amazing work! Thank you