Yes, a keen understanding of business is more important to success than any other talent.
The flip side of the coin would be William Shockley. He received a Nobel Prize in physics, and then started a semiconductor company around the same time. He was brilliant on technical matters. But he was such a bad manager that eight of his best scientists jumped shipped and started their own (successful) company. His company slowly died after that.
Who is in a better position to determine whether an estimate is reasonable if not the people who are going to work on it? And even then, they might not even actually know how long it will take if it’s something that’s never been done.
As a manager, I don’t see my job as about second-guessing their estimates. It’s about making sure they are using a defensible approach for coming up with those estimates and managing expectations with the people funding the work.
In my experience programmers are always optimistic. I never, ever cut an estimate - I usually doubled it and added a week. (For a short time span.) This worked reasonably well.
The biggest issue I had with project management software (though I’m a bit out of date) is that it doesn’t handle rework well, and that is a constant for the projects I was involved with. The training example was putting on a play. Handling the case of finding you didn’t get the rights to the play and having to start over was never covered.
Bad estimates are not from people being incompetent, it is from the assumption that you work on a project around 40 hours a week. Oops, meeting. Oops, fire drill. Oops, another meeting. I’m sure you’re well aware of that problem.
Mention of estimates brings to mind the worst Department Head I was ever doomed to work for. He’d been a mechanical engineer in locomotive and railcar maintenance. Somehow, he was put in charge of all the aerospace, mechanical, and electrical engineers at a Navy aircraft depot (where we overhauled aircraft and engines.) Coming from what essentially was a steel industry, he decided that we were taking too long to work up estimates for overhauling the aircraft. All we needed to do ( 
) was take the total weight of the aircraft and multiply it by a standard dollar figure.
Generally, if your work is steel fabrication, that will give you a pretty accurate number - Item X weighs 5300#, so it should cost $XX,XXX.XX to build at a profit. But when you’re looking at something as complex as an aircraft made from a variety of materials and full of electronics, such a simplistic approach is insane, to put it mildly. Yeah, the radio you’re installing only weighs six pounds, but there’s wire runs and control cables and modified circuit breaker panels and mounting hardware and maybe things need to be moved around so it will fit where it needs to go, and there’s an antenna that needs to go on the outside, so you’re looking at structural work for that, plus weight and balance is critical in aircraft, especially those going on a carrier, and every single project is going to differ from the one before because stuff happens…
Anyway, last I heard (after I’d trransferred) - they moved that guy away from any engineering responsibilities and gave him a non-technical department to annoy. He may have been an ace for the railroad, but he was the bane of our existence. Or maybe he was just a moron?
I was say the same thing. IME, the problem isn’t engineers acting like Scotty from Star Trek, doubling estimates so that they can deliver when Kirk orders them to cut the time in half. It’s engineers THINKING they are Scotty and making major infrastructure repairs, on the fly, without testing, without mistakes, in a real-world high stress environment.
Or even minor infrastructure repairs. When I was at Bell Labs the biggest meltdown of the network was the result of a very small software change. The programmers didn’t notice that they didn’t terminate a switch statement in C properly, and it brought down telephone switch upon switch.
It’s possible he was a moron.
Modern locomotives are pretty complicated devices too. I don’t think you typically estimate repairs “by the pound” for anything like that.
I mean maybe you can tie estimates to aircraft by metrics like weight, number of engines, prop vs jet or whatever. But you also need to establish some baselines first to account for how an aircraft is different from a ship from a locomotive.
I’m pretty sure the entire reason my job even exists is because most leadership in corporate America are, in fact, morons.
There is a recent article in The Atlantic about Elon Musk’s [non-]management of Twitter. As background info, the article cites research that indicates that it’s a fallacy to assume good managers can manage without experience in what their business is. The article is written by James Surowiecki, who wrote The Wisdom of Crowds.
Article is semi-paywalled - one free article per month.
Not about Elon Musk and Twitter, but about the OP’s question: I think a lot depends on the team of people. If the team is sufficiently effective and good, and also if they’re willing to break in a new manager and give him/her their honest opinions, then it’s possible to manage the team and ramp up on the issues and challenges. An effective manager can deal with the higher ups, push back or give as needed, and protect the team and their work.