Ran a little test.

[Giganx]

Missed a 90% to hit on my opening, so in the interest of testing whether or not the to hit in this game is accurate I decided to do some save scumming on my opener. I'm on Tactician and it's the fight against Anders; I have bless and Vow of Enmity up which gives me 90% to hit with advantage. I saved right before the swing and scummed it repeatedly.

So far the hit has missed 36 times out of 50. I could keep going but I'm pretty confident that my actual to hit is sub 50 and the to hit % displayed doesn't mean squat.

I only see this when I'm fighting things higher level than me; I'm almost positive this game doesn't show how level difference is factoring when displaying your to hit.

[Ashling]

I'm pretty sure reloading affects hit-chance or that the hit calculations are somehow run and stored before the hit but in a way which allows some variation.

[Glyph]

Missing a 90% chance to hit does not change the fact the odds are 90%.

[Zerostalgia]

Do you have karmic dice on?

[Zackerie]

sweet you only need to roll 9,050 more times to get the actual probability of a d20.

[Zerostalgia]

Originally posted by Glyph:

Missing a 90% chance to hit does not change the fact the odds are 90%.

This.
50 is a very small sample.

[zero]

Originally posted by Glyph:

Missing a 90% chance to hit does not change the fact the odds are 90%.

you did 50 attempts, it is not a meaningful sample size to show the case

do it thousands of more times and you'll start to see a pattern.

[dexter]

if you have karmic dice, it will affect that tremendously depending on your previous rolls.
Also if you just hover over the bottom right, it will tell you what goes into the calculation of the roll. It makes much more sense if you do this.

[Chrystal]

Advantage is weird when it comes to %s.

Even though you're rolling double the dice, the odds for each dice are the same and it's still possible to roll low on both.

But I have been suspicious about the dice system at times. I play light cleric a lot and the amount of times an enemy rolls a 17 or higher on their D20 attack die and I use warding flare to roll disadvantage on their die, they STILL roll a 17 or higher is ludicrous, like I have that happen at least 3-4 times PER fight.

and yes karmic dice is off.

[Ashling]

Okay, thinking about the test I ran it was probably karmic dice like others have said. (I ran one, once on a successful role, once on a failed roll, and thought it repeated a bunch, but it makes more sense if it's just karmic dice)

[Giganx]

Karmic dice is off.

I've save scummed with other fights to, and the results are generally fairly accurate when I'm the same level as what I'm fighting. I'm telling you, the game calculates level discrepancy but doesn't represent it in the displayed to hit.

[Zygfryd]

Originally posted by Giganx:

So far the hit has missed 36 times out of 50. I could keep going but I'm pretty confident that my actual to hit is sub 50 and the to hit % displayed doesn't mean squat.

If you get a screenshot of the roll tooltip, we can double check the math.

[zero]

Originally posted by Giganx:

Karmic dice is off.

I've save scummed with other fights to, and the results are generally fairly accurate when I'm the same level as what I'm fighting. I'm telling you, the game calculates level discrepancy but doesn't represent it in the displayed to hit.

there is no level resistance mechanics in this game.

it is simply armor class(or save) vs what you roll.

rng is rng.

[Giganx]

Originally posted by zero:

Originally posted by Giganx:

Karmic dice is off.

I've save scummed with other fights to, and the results are generally fairly accurate when I'm the same level as what I'm fighting. I'm telling you, the game calculates level discrepancy but doesn't represent it in the displayed to hit.

there is no level resistance mechanics in this game.

it is simply armor class(or save) vs what you roll.

rng is rng.

You say there's no level resistance mechanic in this game, but do you know?

I'm guessing you assume there isn't one because the displayed % to hit doesn't change based on level, but if you actually test higher level vs same level enemies you miss a lot more against higher level enemies with same AC.

RNG is only RNG when you're rolling an actual, physical dice. This is programmed RNG in a video game; it has layers of conditional code built around it. It's not really hard for me to believe this incredibly buggy game has buggy to hit.

[Runic Tunic]

As others said, if you want decent sample size to take any of this seriously, get ready to do this for another 950 times at least. 50 is far too small. It's extremely likely someone could do their own 50 trials and end up with completely different results, even if everything else was the same.

Not to mention we humans have a natural cognitive bias towards noticing negative outcomes, and having a bad tendency of confirmation bias as well. So the better objective data we have, the better to weed that out.

[zero]

Originally posted by Giganx:

Originally posted by zero:

there is no level resistance mechanics in this game.

it is simply armor class(or save) vs what you roll.

rng is rng.

You say there's no level resistance mechanic in this game, but do you know?

I'm guessing you assume there isn't one because the displayed % to hit doesn't change based on level, but if you actually test higher level vs same level enemies you miss a lot more against higher level enemies with same AC.

RNG is only RNG when you're rolling an actual, physical dice. This is programmed RNG in a video game; it has layers of conditional code built around it. It's not really hard for me to believe this incredibly buggy game doesn't have buggy to hit.

it is not my job to prove something doesn't exist when there is a combat log that says exactly what you rolled, and what modifiers are being applied.

if you are saying there is a hidden value that is being thrown around then YOU have to prove it.

you are falling for small sample bias, the odds of hitting a 90% hit on a lvl 20 vs a lvl 1 is 90%, every time.

your final argument of "oh but its not true random" also means nothing, the difference between true random and computer random is so infinitely small that you will functionally be unable to tell the difference, anyone with even a modicum of programming skill understands this.

but yes: prove your argument cause like, your trying to contradict the in game chatlog that states otherwise.