chance to hit % inaccurate?

[irk]

Played first campain using rifles, 2nd campain using melee.

Anyone else feel that the chance to hit % above enemies might be inaccurate? I can't tell you how many times I attacked someone with a > 85% chance to hit only to miss. It almost feels like a 90% chance is more like 70% while 99% is going to hit 95% of the time.

Maybe I be crazy, but I find anything less than 90% chance to hit is a serious gamble in this game.

[Sugoi]

I was wondering this the other day. I think I've missed way more than 1% at 99% hit. At lower percentages I was willing to toss it off as human nature, but it does seem slightly fishy. Possible that their RNG is abnormally lopsided.

[irk]

O ya, I also started a summoning runner the other day and found it terrifying with a 14% chance for the spirit to become uncontrolled because it definitely feels like it gets loose more than that... Every run the spirit gets out of hand even with such a low chance to do so.

[DrMcCoy]

Naw, I'd say the RNG is probably fine and the chance to hit are accurate.

You can find this discussion for every game displaying to hit percentages (like XCOM, for example). Humans are pretty bad at estimating probabilities and randomness in general, and confirmation bias does the rest.

[Elysion]

Ive missed a 99% 4 times in a row. It was the same character attacking.

One thought I have is that maybe its something maybe with LoS calculations and the game telling you that you can attack, letting you attack, but its actually something you cant attack at all.

Regardless of if thats whats happening or not i feel like the kind of symptoms fit that sort of cause, because i dont feel like under normal situations the percentages are wrong, but its morelike in addition to standard situations there are certain situations where you just can not hit at all no matter how many times you try. Ive done experements like that where someone missess say 80% three times in a row, and ill haste them just to give them another 2 shots to see if they still miss, and they do. - And important to mention to people who just claim its human bias towards no streaks in randomness, i can not remember a time wherei intentionally hasted someone who repeatedly missed to see if they could hit, where they could hit that same target from their same position.

I would think that since i was doing an intentional experement that i would remember if it had ever resulted in a hit.

[Sugoi]

McCoy, I already anticipated that argument and said that there's clearly something else at play here. If you're missing 5% of 99% shots over a large sample, something is wrong.

[Infinite Probability]

I do believe that they stated earlier that not all NPC defensive bonuses were factored in for the hit percentage. For example, if you enemy was standing in the open, it would calculate the base hit percentage, account for no cover, but leave out quickness/dodge bonus, so your 99% would actually be 70% or something. The bottom line is the hit percentage was flawed in some way that they couldnt give you a true number, but the base number.

[Prairie Judge]

the percentages are wrong. i thought it was because i played on hard, maybe they didn't bother to upgrade the percentages. it could be gambler's fallacy, but a chance to hit 75% or more seems to miss at least 50% of the time.

[Fork_Q2]

Originally posted by Duggie Fresh:

the percentages are wrong. i thought it was because i played on hard, maybe they didn't bother to upgrade the percentages. it could be gambler's fallacy, but a chance to hit 75% or more seems to miss at least 50% of the time.

Was going to mention the gambler's fallacy. Missing four or five times in a row when the accuracy is 99% isn't that unusual, 100 shots and no hits - that's something.

[Dorok]

Originally posted by irk:

Played first campain using rifles, 2nd campain using melee.

Anyone else feel that the chance to hit % above enemies might be inaccurate? I can't tell you how many times I attacked someone with a > 85% chance to hit only to miss. It almost feels like a 90% chance is more like 70% while 99% is going to hit 95% of the time.

Maybe I be crazy, but I find anything less than 90% chance to hit is a serious gamble in this game.

At first you isolate a case and then change it in a general complain.

It's very difficult to analyze, for players saying they get/got 5 99% miss in a row, I wonder if they really quote well the events or if I have an insane luck to never have seen that. I never ever saw three 99% miss in a row, and I'm not sure for two 99%.

It's possible there's a bug triggered only for some players and generating a ToHit problem.

Now about aiming a target behind an enemy or a friend, I remember had the feeling in SRR that it was increasing the miss versus the chance shown. That feeling didn't came back at all with Dragonfall. In Dragonfall I also put a much higher focus to target ToHit higher than 90%. I still do attacks with lower ToHit but I constantly try improve the ToHit as a major goal.

[Elysion]

Im convinced there are certain situations where you will not hit regardless of what the % is telling you. To hit chance feels right most of the time, but ive done things like burn ap drugs or haste buffs in situations where i get streaky missess with high hit rates, and it almost always continues the miss streak

[Fork_Q2]

Originally posted by Bringer:

Im convinced there are certain situations where you will not hit regardless of what the % is telling you. To hit chance feels right most of the time, but ive done things like burn ap drugs or haste buffs in situations where i get streaky missess with high hit rates, and it almost always continues the miss streak

That's the gambler's fallacy right there; if you roll a die, and get four each time, the universe isn't going to compensate this by giving you a number other than four - it could just very well be four again.

You would have ignored the streaks of hits when you had far below 90%, but you pay attention to the times you noticed you did not hit with 99%. The only way we test the accuracy of the "to hit" modifier is to run this over a very large sample.

[Philip Marlowe]

It's simply the nature of randomness and probability which are counterintuitive to the constant pattern creating nature of ♥♥♥♥ sapiens brain wiring.

I've actually made 3-4 subsequent hits in a row on probabilities between 20-25% but I didn't mind those at all.


Addendum/correction: While it is true that our organic brains are wired to constantly find patterns, even ones that aren't there. It' important to note Actually I was only partly correct, as one can actually use conditional probability to check how strong the evidence is that the game mechanic's random generator is faulty.

[Sugoi]

I don't think you guys realize how improbable missing 5 99% chance shots in a row is, and to see similar behavior multiple times in your LIFETIME is quite remarkable. It's definitely wrong on some level.

[Fork_Q2]

Originally posted by Sugoi:

I don't think you guys realize how improbable missing 5 99% chance shots in a row is, and to see similar behavior multiple times in your LIFETIME is quite remarkable. It's definitely wrong on some level.

No, it really isn't that special. Each shot is probably independant of each other, so in theory the chance to hit re-rolls after every shot you make.

I would be suspicious if you took a large sample of different scenarios where you had 99% to hit, and found a "miss" figure significantly higher than 1%. Look up law of large numbers.

Not saying the accuracy modifier isn't bugged, but I haven't seen much evidence for it.

[Philip Marlowe]

I have never missed 5 99% probabilities (yet) but chances are someone will in the large pool of players. Fork_Q is correct it is the Law of Large Numbers. Each shot is independent of each other.


Addendum/correction: Actually I was mistaken, one can actually use conditional probability to check how strong the evidence is that the game mechanic's random generator is faulty.