That assumes a high correlation between validity, acceptance and success. Peer review itself is conservative, reactionary and political. There is a strong "con

That assumes a high correlation between validity, acceptance and success. Peer review itself is conservative, reactionary and political. There is a strong "confirmation" bias in what gets published. This is good, I suppose, but: It takes a lot more than being right to get your results out there and accepted by your peers. And if your results contradict the existing consensus, god help you! You'd have to be very politically astute to be published at all, and hang on to your funding. Conversely, if your results appear to support the existing consensus, or can be doctored up to do so, then the "validity" requirement is greatly relaxed. Either way, political concerns will always play a role.