I was telling a friend about the multiverse explanation for fine-tuning. He asked me a question that I had never thought about: Why assume that the conditions in different universes would be the same? Maybe it's all the same, and so the multiverse does not help with fine-tuning.
In fact, it seems the point can be strengthened. The constants in the laws of nature appear to be the same on earth, on the moon, in M 110 and around distant quasars. By induction we should assume they are the same everywhere. Granted, on some theories other island universes are not connected to ours (though on other theories, there is a containing de Sitter space, and on some theories the other island universes are just very far away). But while that may weaken the induction, it does not destroy it. Even before Europeans heard about Australia and Australians heard about Europe, each group had reason to suppose that apparently basic constants in the laws of nature would be the same in the other place, even though the two places are not landwise connected. Granted, however, the judgment whether some constant is basic is defeasible—thus, if one mistakenly takes the local gravitational acceleration to be a basic constant, one will mistakenly think it is the same on a high mountain as in a valley. But while a judgment of basicality is defeasible, it can still be reasonable.
Now, some multiverse theories grow out of a particular physical theory that implies a variation of constants, say because there is given some universe-generating process. So the point does not damage all multiverse-based explanations of fine-tuning. But it does raise the evidential bar: for, the defeasible presumption is that if there are other universes, they are very much like ours.