As i have figured if your myst/arms/science you should be crafting under those items with those labels... tho the whole system is kinda stupid as you can do arms under myst... but that's not the point.

As i play with the system i notice that the success rate calculation is wrong. I'll have an 85% chance of success and fail more then i succeed. Anything under 90% fails as if its at a 10% success rate and I've been testing this for awhile. And yes i know it goes up as i level and collect but my point being the success rate system is flawed and inaccurate. Its also silly that i'm finding myst and arms in science stuff etc etc.

I hope i did not waste my time typing this.

You will need to post data (number of tries/number of successes/success rate according to UI) for anyone to take such a claim seriously. If you don't have anything resembling data, you do not have the basis to make such a claim.

Anecdotally, I have not observed fusion success rate to be significantly different than what the UI claims, although I have never bothered to formally gather data. I've gotten streaks of good and bad luck, of course, but nothing exceptional.

I did fuse a few thousand mods (albeit alot of that was 1-2 and 100% for me) and I have yet to see any discrepancy, at least not on a level so drastic as you describe. I've done at least a few hundred between 3 and 4s, On occasion I might get 2-3 failures in a row.

But claiming you're getting a 10% success rate on anything below 90% (as in 89% and below) - I call bs. Then again, I've also seen the OP's other posts... so no surprise there.

"Hey, guys, I tried fusing and modding a couple of times and I failed. Clearly the system lies and this has to be fixed."

Yeah, people are great with percentages and actual maths :p

While, I think the OP is off his rocker.

Ill come to his defense a tiny bit by saying I have noticed that the random number generator tends to be a little streaky here.

Often times, I wont simply just get a costume drop. But a two or three of the same costume all within several drops of each other. I once had 3 werewolf feet drop from 3 separate mobs all within a few minutes of each other.

I don't often get a single failed fuse, but 2 or 3 successive fails. Even with a high percent chance of success.

Ill admit, my info is anecdotal. But i've seen it happen enough times to myself to notice it as a pattern.

Yes, but a properly-working RNG SHOULD be streaky, because true randomness is streaky. The absence of streaks would be a far stronger indicator of non-random-ness than their presence.

Suppose I flip a coin ten times. Which of these outcomes looks plausibly random, and which one looks faked? (H = heads, T = tails)
HTHTHTHTHT
HHTHTTTTHH

Both outcomes have 5 heads and 5 tails, which is the most likely total for 10 flips of a fair coin, but the first one is a perfect pattern and the second is jumbled (and has streaks). And we know randomness shouldn't generate perfectly orderly patterns.

Now, if the game's RNG generates streaks disproportionately often, that could be a problem, but that kind of thing is EXTREMELY difficult to spot without rigorously collecting data, because the human mind is spectacularly bad at recognizing such things instinctively. Confirmation bias, pattern recognition in random noise, etc.

When people hear "random" what they think in their mind is "even distribution of results per unit time". They don't expect 5 heads in a row, (though it is perfectly plausible) - they want to see an even amount of heads and tails - not some spotty grouped together stuff. That's one of the reasons CoX implemented a streak breaker - because missing a bunch of times in a row is not fun, (even if the determining roll is genuinely random).

If the drop rate of the FF slot from the grab bag was 1%, then people would expect that buying 100 bags would be a guarantee of getting one. Of course, that isn't necessarily the case, and you could end up getting 100 or none...

Very interesting and a true mathematician would explain you that the 1st one is fake because the odds are too linear.

That's all perfectly understandable. but does little to explain getting streaks of the same type of drop. Its one thing to get a streak of drops, in the same way its normal to get a streak of no drops.

But to get a streak of the same costume, repeatedly and in some cases sequentially, seems really odd. However, in thinking about it. It probably has less to do with the RNG and more to do with the drop table being borked or some such,

There's something to be said for a streakbreaker mechanic. Losing a fight/failing to craft an item/etc solely because the RNG decided to hand you a long string of terrible numbers completely beyond your control is frustrating.

But yes, people in general are terrible at grasping probability. Human intuition in this area tends to be drastically wrong. Personally, my gaming roots lie in tabletop RPGs and wargames, which means lots of dice, so my intuition has been forcibly reshaped by direct exposure (still far from perfect, though). There's only a 1 in 20 chance of getting a 1 ("critical failure" in many rule systems) on a 20-sided die... which means each player is almost guaranteed to critically fail not just once, but multiple times in a single gaming session. You have 50/50 odds of rolling a 4 or higher on a 6-sided die, but if you roll 20 such dice, having exactly 10 of them (rather than 9, or 11, or 15, or etc) come up 4 or higher is so uncommon as to be a noteworthy occurrence (among my social circle, the preferred exclamation for this was "how statistical!") I can scarcely tell you how many times my 3+ armor saves (2/3 chance of success) repelled ten hits in a row, or failed six at once, although of course most results fall somewhere between the two extremes.
...unfortunately, too many tabletop gamers take this too far in the other direction, and conclude that dice are incomprehensible and fickle gaming spirits that must be appeased, concocting rituals and superstitions that would make draw a raised eyebrow from the most inveterate gambler.

Edit:
No, the same principle applies.

I mean, if you can collect data and show that disproportionately many players are getting such streaks disproportionately often, sure. But the simple existence of streaks is in fact decent evidence that the RNG is working properly.