The Many-Worlds interpretation is possible when the event at issue has a 50-50 probability. It becomes suspect when the probability of an event is not 50-50. The prime example is experiments with entangled particles with correlation results violating the Bell inequalities. For example, at the maximum positive difference there is an approximately 71% correlation between entangled particles (showing lack of locality) in comparison to the classical outcome of 50% correlation.
The Many-Worlds interpretation includes the assumption that the laws of physics (and the Schrödinger equation) are the same in each world. That means that in each of these worlds the 71% correlation applies (assuming the experimental parameters are set correctly).
For the MWI to be correct, though, the events happening here must have opposite outcomes in other worlds. Our world having a 71% correlation at the optimal settings means that there must be a corresponding total 71% non-correlation in other worlds. The problem is that each of those other worlds must have its own 71% correlation, leading again to a total 71% non-correlation in other worlds. Presumably decoherence has been happening since roughly the time the Universe began. Doesn’t this mean that it’s extremely unlikely the MWI is correct, even though this does not prove that the MWI is wrong?