In my experience, this is how people react to pretty much any rng in any game. You always notice the bad rng more than the good rng so it always looks like the rng is skewed against you beyond what it claims the probability is.

And as always, the only way to be sure would be to test it hundreds of times and record all the results to see if it actually does consistently deviate from the expected results.
(or, in this case, somehow find the formula in the game's code to verify it)