I liked this essay and I'm motived to write something about the subject. In the meantime here is some of what I think:

1. The simulation argument was also voiced in a rather famous short story — Vernor VINGE, The cookie monster, Analog 123(10):8–40, 2003.10 — and a well known novel — Robert WILSON, Darwinia, New York: Tor, 1998.

2. Much of what inspires the simulation hypothesis is metaphysical disquiet regarding the foundations of quantum mechanics. If you can imagine that measurement is necessary for existence, then you might be comfortable with the nonlocal (quantum blob) or equivalently participatory nature of the universe at the smallest level. After all, both John LOCKE and James HUTTON argued interaction and measured difference, and also well know, for example Satosi WATANABE (Knowing and guessing, New York: Wiley, 1969), are primary to existence. What is not measured at all induces no dynamic on any dial, and therefore doesn't do anything, so what does it mean for it to exist? At the smallest scale, where also large energies by E_0 = hv, if the measuring device doesn't measure they system, then nothing measures it, and we get counterintuitive behavior, as that situation never comes up in life in the classical macroscale world.
   Some people begin thinking of simulations, because the future determining the past aspects (made famous by John WHEELER in his lectures about a measurement today determining the state of a star in the past) seems very odd. Unless you view the world as an set of intersecting experimental setup each consisting of measuring apparati distributed over space and time, and we are simply observing nonlocalities. (Basil Hiley prefers to discuss quantum blobs and not participations. He showed that the average of just of a few quanta is often entirely deterministic as a unit.)
   Roger PENROSE and David FINKELSTEIN conjectured at one point or another that all bosons are statistics of fermions and that you have the classical world emerging from graphs of prespinor measurements that form linked even tuples of spinors. John WHEELER famously argued over forty years that everything was a statistic of bits, etc., etc. You get the same supermanifolds with Grassmann operators as in string theory, but this develops statistically into a Clifford algebra at the scale of larger particles and then into a Minkowski metric on the largest scales, which is unexpected yet desired. In that case, perturbations of quantum events are particles, and fields are monoidal categories with a few basic types of particles/operators and functors between each pair of categories. Etc, etc. Bob COECKE and Aleks KISSINGER (Picturing quantum processes, Cambridge: University Press, 2017), for example, are following up on that. Stephen WOLFRAM (A new kind of science, Champaign, Wolfram, 2002) has his own thoughts which he developed in his book.
   Israel GELFAND, Heinz Foerster, and Gordon PASK originally suggested, and it has been periodically repeated, that much of the indeterminism is because we don't have enough accurate knowledge of the past of particles, that past matters more than we admit for predictions. Giacomo M. D'ARIANO, G. CHIRIBELLA, P. PERINOTI (Quantum theory from first principles, Cambridge, University Press, 2017) recently had a fundamental book out.
   And other theories, such as the fact that if we admit general relativity, we get indeterminism for free, from any perspective, because a distributed computing system passing messages with time required for messages to travel, has, for example, computable but unbounded counts with a random number in the Wolfram sense that any statistical analysis would have it appear random, which do not exist in Turing machine models, which are deterministic or nondeterministic, random, but never compute unbounded counts.
   Many, many conjectures that remove metaphysical discomfort exist. They are areas of active research. Foundations of physics and Philosophical transaction of the royal society A, for example, are doing quite comfortably as journals.
   Yet all current proposal are all highly mathematical. There was a recent Royal Society paper that quite possibly there is nothing really wrong with quantum mechanics or relativity, other than philosophical dislike of some of the implications. The simulation hypothesis becomes appealing, because it explains nonlocality and quantum indeterminism and nondeterminism as generation of details on demand, the world as lazily computed, i.e., only what observers bother measure is computed. That is plausible considering how we program video games and other applications.

3. One issue with the simulation hypothesis is that it's unnecessary, for existing physical theories have no problem dealing with the philosophical issues in quantum mechanics. Noncommutative aspects are necessary anyways for measurement and sufficient difference to be possible, and moreover, to buffer measurements, so that measurements are meaningful. For a system where all velocities and group generators are commutative has singular commutators. What is the problem with that? The map null commutators to nonnull commutators does not preserve isomorphism, which means small errors in measurement can result in nonisomorphic topological skeletons for the universe, due only to measurement noise. Measurement are unstable in that case. Both in the sense that we cannot infer a different measurement result is due to the system and not due noise in the measuring device, and in the information theoretic sense related to existence of primary particle, and such a world will not long exist. If the universe were a simulation, and there was a larger computer outside it, simulating it, noncomputable results would occur more frequently than random (guess and check works), and we do not statistically observe that. Quantum thermodynamics is basically valid. (Hiroomi UMEZAWA, Advanced field theory micro, macro, and thermal physics, New York: American Institute of Physics).

4. The simulation hypothesis is plausible but very improbable. Yet most things that don't happen are such not because they are impossible, but because they are sufficiently improbable. That is the basis of thermodynamics.

. . . And did I just write another far too long comment? Looks like it!