I think the reason so many people don't consider the simulation argument compelling is that it is usually presented inaccurately in the mainstream media. The original argument never makes reference to computer games (and it shouldn't because people in an unrealistic simulation would not know it was unrealistic). It was a trilemma proposed by Nick Bostrom in the form:
1. The human species is very likely to go extinct before reaching a "post-human" stage.
2. Any post-human civilization is extremely unlikely to run a significant number of simulations of their evolutionary history (or variations thereof).
3. We are almost certainly living in a computer simulation.
The author constructed the argument such that at least one of the three scenarios must be true. It can’t be escaped. (One can deduce that from rudimentary logic). We are then left to choose which one we consider to be true. Choosing either 1 or 2 will require a lot of presumption on our part. 1 especially as there is no compelling indicator for the impending extinction of the human race (Nuclear war, global warming, geological events, and all other common Earth-based disasters are probably not enough. It would probably require something of astronomical origin). The uncertainty of 2 becomes apparent once you read about estimates of the computational power that might be available to an interstellar or galactic scale human civilization. It is estimated that if you converted the whole of Jupiter into a computer, you could simulate the entire human history in less than a second. This means that a teenager 1,500 years from now might playfully simulate the whole of human history on a boring afternoon. This makes it not so unlikely that a future civilization might want to run ancestor simulations just for fun. (If that is not compelling enough, think of the requirement that of all the trillions of humans that will be born in the future, not one of them will decide to do this). As 1 and 2 are now doubtful, you are left with 3. 3 becomes compelling once you realize why the jump from 2 to 3 is so sudden: if a post-human civilization decides to run ancestor simulations, (given they can simulate all of history in a fraction of a second on just one extremely powerful computer), simulations would vastly outnumber real people and by sheer probability, we are more likely to be one of the trillions of possible simulations than that exact one reality. If you reject 3 based on incredulity or some axiom then you are back to 1 and 2, which is basically square one.
I agree with your first conclusion however and that’s generally my attitude towards the topic.