Is anyone else here feels like some classes are rigged to fail, compared to others?
All my attempts at playing a wizard lead me ragequitting because my spells would constantly miss, while the enemy easily chipped away my party health with occasional crits. I get that the rng can screw both screw you over or make the playthrough a cakewalk, but here it feels like the game is bugged and displays wrong values or straight up lies.
I started the campaign as a wizard 3 times yet every single time, when I was shown that my spell has a 90% of hitting a target, my character would either miss or deal a minimum amount of damage. I had cases when my spells would miss up to 5 five times in a row. Not even a single time throughout those battles did I save scum in case of a miss.

To show you what 90% really means when you're rolling dice (in D&D but really any dice game):
As we know those percentages in BG3 do only concern attack rolls 'cause for skill checks they're simply not displayed. Which should honestly be the case for attacks as well, and here is why.

On an attack roll there is one outcome that will always miss, a natural 1 (NAT1). The probability to roll a NAT1 on a straight roll is 1/20 (or 5%), with advantage (roll twice take the higher result) 1/400 (or 0.75%) and with disadvantage (roll twice take the lower result) 39/400 (or 9.25%).

"Perfect rolls" scenario:
Let's say you roll 20d20 and you only get the worst rolls possible. On 20 straight rolls you will have seen 1 NAT1 (along every other number as they all have a probability of 1/20), with advantage you will have seen 1 NAT1, with disadvantage 20 NAT1s.

Let's translate 90% chances to hit to their rolls. Straight roll: 1, 2; disadvantage: 1; advantage: 1, 2, 3, 4, 5, 6
Huh! 90% to hit with advantage gives you actually the most rolls on which you can miss, because the likelihood of rolling those is very slim.

Now there's one oddity in the game that makes me believe there is something wrong. Not the rolls but rather the numbers being compared.
If you go into the beginning Imp fight with a +5 to hit you get hit chances of 95/100/90% (straight,Adv,DAdv) as they have an AC of 7, so you can only fail to hit on a NAT1. However if you miss the game will most of the time say "miss" even though there is a display for "critical miss" in BG3.

This is wildly incorrect. 50% is a *probability*. It is possible (with a very low probability) that all 100 tries will be successful (with an equal probability that all 100 tries will be unsuccessful).

Thank you for actually knowing this. 50 percent doesnt mean you will land half of your total number of hits. I cant understand why people dont get that.

-Leet

These guys should stay away from Las Vegas.

it's not really that the RNG i rigged, it's just that the % doesn't really translate what is happening when we make an attack roll, grizzledlone explained it

I ran a wiz and got what I thought was way too many crit hits yet my rogue has yet to successfully pass a pick pocket check of 6.

I think of the to hit chance more like a coin toss 50/50 you might get lucky and land hits for every shot or you might not

edit: also should post this to larian forums if it isnt already

I think it has more to do with advantage.
I did not know this before but if you roll to hit from a higher elevation with a ranged weapon you get advantage on all rolls, this is why Larian has frontloaded level 4 and 5 encounters with ranged enemies. Because the second they unlock multi-attack they only need to climb up a stepladder and they are death incarnate.


Note how many enemy-adjacent platforms start to show up at level 4.

That's what I genuinely hate about these kinds of hit chances. It's hard to enjoy the combat when you can't even rely on the simplest of skills, while the enemy just chips away at your party's health. Not to mention AI always an advantage in numbers.
I don't a single reason not to reload saves when the bad rng just starts wiping out your team without any blink of hope.

I have the same feeling. I know the successive attacks are independent in terms of probability, but still, the law of large numbers seem to indicate a low probability of compliance vs the expected (independent) probabilities.

Since they're using a pseudo-random generator, and not a true random generator, it's possible they implemented a rigged system, more by mistake than by design. Those are quite difficult to implement correctly.

There's only one way to go, collect enough values, and use an estimator to determine the chance it's rigged.

RNG stands for Rigged number generator.

Not that BG1-2-ToB didn't have RNG-problems.

There 50% ment maybe 33%, and if you got sorrounded by a bunch of enemies, they hit you every single time, while you barely landed any hit.