that’s not how statistics work. i think you are demonstrating that you don’t understand the math. you need a big enough sample size to establish statistical relevance, otherwise there’s no way to prove that your results mean anything. this is very basic statistics. 80% chance doesn’t mean you “win” 80% of the time… it means that you have an 80% chance.
do you play D&D? you roll a lot of 20-sided dice in that game, and you theoretically have a 95% chance of not rolling a 1, but it happens. sometimes multiple times in a row. the same is true for rolling a 20.
the chance doesn’t change no matter how many “failures” or “successes” you get. if you flip a coin, it’s a 50% chance for either side every time. it’s extremely unlikely that you’ll land on heads 10 times in a row, but it CAN and DOES happen.
Do you have records of:
How many times you crafted on EACH item.
What each item STARTED as. Affix/Instability
What each item ENDED as. Affix/Instability
What glyphs were used during EACH craft.
No. You don’t.
You have 1 number. “I started with 100 items. 57 failed at EXACTLY 86%”
I 100% don’t believe this, because every single item starts with different affix tiers, different instability, heck different affixes overall. Saying 57 failed at the exact same number is so unlikely, that you’ve hit the mathematical jackpot.
Without the information about the exact point every item failed, your claims have a ‘confirmation bias’. You believe they all failed at 86% because that’s the number you remember the most.
If you had all the information recorded, and could point us to a spreadsheet with every item listed, all the affixes, instability, and fracture points, then I would believe that all 57 failed at 86%.
If you started with 100 White items? This is the absolute worse way to craft in this game, and you will NEVER make a T5/5/5/5 starting from white items. I don’t believe it’s even possible, as the instability and affix failure chance mean you literally cannot get that far.
I agree with @darkdeal, uniques should change how skills/etc work, they just need more of them with interesting affixes.
It’s hard to “crowd out” other poison weapons when there’s only 1 other (Draalsting, no, I’m not counting Alchemist’s Ladle). Plague Bearer’s isn’t even that strong or interesting, it just has 100-220% poison chance (a t7 chance to poison affix caps out at 108%). I’d far rather it did something a bit more interesting that just have a lot of poison dps.
How would you want to see exalted items implemented? Apart from having level-appropriate bases drop.
LMAO this is hilarious. You have no understanding of math whatsoever. Probability is not deterministic. It’s like me complaining that I got heads two out of two times and complain that it’s not 50%. And I also call bullshit on your claim you had 56% success rate when it said 86%. Each craft attempt will have different success rates. When it says 86%, it means for that attempt you have an 86% chance to succeed and 14% chance to fail.
Saying 86% success probability is not the same as 86% of attempts were successful. The first part is chance of something happening. The second is what happened. Chance/probability does not equal results in a small sample size. And yes, 100 attempts is a small sample size. And like we said it’s bullshit you attempted a 100 tries with an item that shows exactly 86% success probability.
Yes, it’s an aRPG, PoE & D3 have similar amounts of RNG but they don’t let you choose what affixes you want on your gear.
It’s not crap, the devs have tested it, it’s decently close to what it should be:
That’s called pareidolia, seeing patterns where there aren’t any and given that even with RNG you can get patterns if you look into the details enough.
It is an 86% chance, if you took a sufficiently large sample, 86% of them would succeed, 14% of them wouldn’t.
Try PoE & D3 and see how much RNG they have in their loot drops…
I guess it’s hard to say it’s BIS. Theoretically if you get an item with T7 poison chance and damage it will outdo it. But getting the staff is so much easier than aiming for the stars.
Getting level appropriate drops alone would drastically improve loot. Moreso for exalted items since it’s so low you want to have it narrowed down as much as possible. Besides that I’d like to see the drop rate increased slightly. A larger improvement I want to see as a whole is a smart loot system. Meaning you don’t get crit chance and damage over time affixes on one item. For example, remove all DOT affixes from rolling on solarum bracers and silver amulet and the crit relic. I’m sure there are other examples of that. It doesn’t even need to be removed, just lower the possibility so other combinations are slightly more likely to drop.
Ok, so a max roll t7 poison chance (108% chance to poison) & a high roll t4 (up to 108%) or above poison damage prefix would outclass it, but if you only got a max-roll t5 chance to poison (76%) you’d need something like +189% poison damage (a min-roll t7) to equal plague bearer’s.
Anything close to that is probably better than the staff considering the hp drain. But if hp drain doesn’t matter much then I’ll take your word for it (can’t figure out the breaking point like you did).
Here is a great video that talks widely about VG design and RNG. It’s worth to watch it whole as it is quite instructive and enlighting.
But I’ll try to quote the part more specific to LE crafting/looting. And it is at 6:00 to 8:00
There it is
So it is explained how with OR we have our plan and strategy e.g. craft gloves with life and attack speed, and the game comes after our decision to ruin our plan with “fractures”
^This is certainly the case of POE that is almost uniquely based on OR, despite it being a PVE game it has some of the most stressed edgy community i’ve ever seen.
And players who avoided it (and for example are happy to play GD or anything else with a less inhumane rng tbh) describe POE as a skinner box
The rest half of the video explains how devs of various games solved issues/removed unnecessary and excessive OR and introduced as much IR as possible
Now before you come to dismiss the entire post, please understand that there is a lot of IR in LE and it is possible to have more of IR than OR as well.
The stab/guard runes are an example actually and tuning like odds at high tier crafting can solve a lot of multi layered RNG (imo).
Also other things present here or in other games like the gambler, kadala in D3 and target farming in GD are all great solutions than 99% randomness that kills player choices/strategies and leads to irritation
Also, it is why so many people (incorrectly) stick with their 1st pick in the Monty Hall Problem - we humans don’t understand probability very well and our cognitive bias takes over.
