I adopted a behaviour at work that if I am fairly convinced about X ending up being wrong, and I see that trying X is not too costly (esp compared to arguing about it), then I just let X eventually fail, and take it from there, already knowing why this happened.

People seem to learn better this way, and there is no better argument than reality itself. Of course it cannot be used everywhere, eg if trying X until it fails takes too long, if it involves buying an expensive machine that we will not be able to change etc, but there is a good portion of stuff it can actually reduce interpersonal friction on. And the process of changing from X to Z happens organically that sometimes I don't even have to explicitly say that "I knew all along" (though I must admit I derive an internal satisfaction that I knew all along).

It was a time when at work there was a widespread interpersonal tension between everyone, and reducing interpersonal friction was more important than spending more or less time on sth that would not work. I dont think arguing and discussing things are to be avoided per se, but in certain circumstances, if one knows that a team will eventually go down on path Z anyway due to necessity, it may not be worth arguing about at all.