You could, and that's a fair point, but you're more talking about philosophy of science and epistemology. My point is that Many Worlds seems to die to Occam's Razor, because of all the additional requirements to accept Many Worlds over something like wave function collapse.
Back when I read about this stuff I had the exact opposite reaction - Occam's Razor seemed to favour Many-Worlds.
Given direct experience of one universe, I think it's more ontologically conservative to introduce multiple universes than it is to introduce a whole new unprecedented category of cats that are simultaneously dead and alive.
Very loose analogy: if I see a mouse in front of me and simultaneously hear a squeak behind me, I'd probably assume two mice rather than one mouse which has learned ventriloquism.
Interesting. Occam's Razor says "don't multiply entities beyond need", not "don't multiply types of entities beyond need". Of course, you could argue that types are entities too. So many worlds minimizes entity types at the expense of total entity count.
"Number of entities" is probably best understood through Kolmogorov complexity. So two regular mice, or million regular mice aren't much more complex than one regular mouse. But having one regular mouse and one "special" mouse is more complex all of that.
To quote from a quote in David Wallace's The Emergent Multiverse
"Davies: so the parallel universes are cheap on assumptions but expensive on universes.
Deutsch: Exactly right. In physics we always try to make things cheap on assumptions.
(Interview between Paul Davies and David Deutsch)"
Something like Many Worlds is necessary to explain what it means for something like a wavefunction to collapse. Although it may multiply entities, that doesn't mean that it does so unecessarily. The trade-off is that new physical laws aren't introduced when doing so, and with other theories, you can say that introducing new laws is unnecessary, with respect to Many Worlds.
But is it even possible to view or otherwise observe these alternative universes? That's my biggest problem with the Many Worlds hypothesis: it's impossible to verify.