And, indeed, perfection of the craft involves politics: it's not just understanding the technical space, it's about, eventually, understanding why other people see that space differently, what their goals are, how those goals overlap or don't, and how technical choices feed into that social layer.

Back in the day, Chrome was about a sandboxed subprocess architecture that made for a more stable browser. It was also about breaking the back of the Microsoft monopoly and advocating for why people should bother to care (remember the comic strip Google commissioned?). Nowadays, if it weren't about politics at all, Chrome would still be the best choice because it's still technically very good.

But there's more to the problem than simple technical competenece.

Those hours are sold off to someone else to fund the stuff that matters in my life, where the financial RoI is negative.

You can hone your skills while still maintaining a healthy detachment. You make your case at a thing, business decides to do something else that you think is dumb. You only "lose" if you were overly attached to the decision in the first place. Otherwise you simply get a chance to observe the outcome, see what went well/poorly, and reflect on whether/how you were totally right all along. Next time you have a clearer understanding and perhaps will be able to better articulate your position. You didn't lose. You gained experience and wisdom. You always win as long as you're open to do so. The business lost by listening to the wrong person.

If I need to dig into social engineering and extrovert masking to be an effective engineer I probably should also look for another job. I hate places where this borderline nepotism is the only way to get anything done.

Oh well, I'll just endure it until the job market relaxes a little.