Yes, you must draw conclusions from empirical data. However, there is a likelihood, and there is a prior. The emdrive has no theoretical basis, so the prior i

Yes, you must draw conclusions from empirical data. However, there is a likelihood, and there is a prior. The emdrive has no theoretical basis, so the prior is very low. The likelihood is small but not very small, because the results are just about at the detection limit. They claim that thrust is linear in the 40-80 watt range at 1.2mN/kW. Rather than go to the press, why don't they just turn up the power until the effect can't be denied? The fact that they did not publish that is also information.