There is absolutely a scientific way to test that.

You find 100 people who have COVID who are vaccinated, and 100 people who are unvaccinated. Then you record which people get worse.

If the unvaccinated group suffers from more severe COVID symptoms than does the vaccinated group, then you can safely say that the vaccine appears to reduce the severity of symptoms in infected individuals, and therefore, based on these results, "it would have been worse if you were unvaccinated."

That's how you'd test it scientifically. The hypothesis is either correct or incorrect, and through observation and data, you can reach a conclusion.

Is there any data comparing the severity of COVID infection between vaccinated and unvaccinated populations?

Saying you don't trust the tests or the data that exist is not the same thing as it being impossible to scientifically test your hypothesis. And thank goodness, since a testable, falsifiable hypothesis is the only kind that can be considered researchable.

There is absolutely a scientific way to test that.

You find ten people who have COVID who are vaccinated, and ten people who are unvaccinated. Then you record which people get worse.

If the unvaccinated group suffers from more severe COVID symptoms than does the vaccinated group, then you can safely say that the vaccine appears to reduce the severity of symptoms in infected individuals, and therefore, "it would have been worse if you were unvaccinated."

That's how you'd test it scientifically. The hypothesis is either correct or incorrect, and through observation and data, you can reach a conclusion.

Is there any data comparing the severity of COVID infection between vaccinated and unvaccinated populations?

Saying you don't trust the tests or the data that exist is not the same thing as it being impossible to scientifically test your hypothesis. And thank goodness, since a testable, falsifiable hypothesis is the only kind that can be considered researchable.