But you do trust mining groups not to get to 51% capacity, which has happened before. You do trust that ECDSA/secp256k1 and double SHA-256 don't have unknown sh

But you do trust mining groups not to get to 51% capacity, which has happened before. You do trust that ECDSA/secp256k1 and double SHA-256 don't have unknown shortcuts, even though that hasn't been proven.¹ You do trust the Bitcoin Core developers, and the developers of your wallet program, not to add a backdoor, and you do trust your own ability to keep your private keys confidential. In minimising trust, you're also centralising a single chain of it; the real world is resilient against misplaced trust (so long as many entities aren't secretly untrustworthy at once, which we call a conspiracy), but Bitcoin isn't. It's a small number of single points of failure. I'm glad you said “reduced trust”, rather than “trustless”, because otherwise this would've been a several-page rant. ¹: It's unlikely either of them will have significant shortcuts discovered any time soon, but that's not the point.