It's really a shame. Some fields are undergoing reproducibility crises at the moment, and I'd expect there are innumerable questions that you could survey those

It's really a shame. Some fields are undergoing reproducibility crises at the moment, and I'd expect there are innumerable questions that you could survey those communities about, only to return with a strong but irreproducible consensus. Peer review is hard enough without challenging the status quo, and peer review has demonstrated itself to be an incomplete mechanism for science (short term, at least, and we need climate action in the short term). With that in mind, red teams sound great. However, when doing anything statistical, there are questions you can debate before and after you collect data, and that distinction is incredibly relevant here. Having a red team push back against methodology while no-one has yet seen the data sounds great. But if you have a red team supporting your methodology before you collect data, only to challenge it based on the conclusions, you're basically endorsing using your "researcher degrees of freedom" to push in a particular direction, and it's terrible. It's p-hacking with an agenda. It's oxymoronic to ask people to do science that leads to specific conclusions, and that feels like where this is going.