The standard argument for simulation boils down to "in the future, we'll get access to insane computational ability, and someone is bound to run ancestor simulations just because we can; once this becomes possible, the amount of time consumed by such simulations will dwarf the time of history." In other words, the argument is basically "it's not impossible, so it will eventually happen, and over a long enough time period, it doesn't matter how long it takes to happen."
(As you might imagine, I am not persuaded by this argument.)
No, it's far more likely that we're the result of an emergency overload dump of a runaway thaumic reactor experiment run in the squash court of a university for wizards on a much more sensible flat world. There's far less hand-waving in that narrative.
(As you might imagine, I am not persuaded by this argument.)