Ask HN: how to tell when an idea is terrible or great?

I've been ruminating for quite some time on this issue and can't seem to get anywhere with it, so I'm hoping you guys might throw in some interesting points of view I hadn't considered.

My peers and I have spent some time debating the worth of an idea a few of them had, and we're essentially in two different camps: some believe the idea is too complex, too vast and too risky for a team our size and with our expertise, others think the idea is the future and we should pursue it at all costs. According to this camp, even if we don't have the qualified manpower to execute on the idea, it's ok because once VCs take the bait, we'll have vast resources and hire a bunch of experts who will fill the vast gaps we currently have. Making smaller MVPs is also according to them impossible because it won't truly shine without all of the pieces in place, which is likely the work of several years.

This is the problem, on one hand you know that something like 90% of startups tank because they've bitten way more than they could chew, and were working on massive assumptions that didn't even bother validating. They're basically rolling the dice, and the odds are bad to say the least. On the other hand, you don't want to be the grumpy conservative dude who doesn't want to take risks and cannot see a brilliant revolutionary idea when it's right in front of him. Yes, you have experience, you've seen projects fail over and over, you want to bootstrap and grow in predictable steps rather than making big leaps of faith. At the same time you suspect that if everybody around you was like you, nothing new would ever be done.

At this point I basically don't know how to address the issue. It seems like in these situations you have to rely on your gut and risk being wrong, or never know if that concept would have taken off. Even the trick of "asking 20 random people to see if all 20 are super excited about the idea before pursuing it" doesn't really help you if you're creating something that wasn't here before. The iPad was deemed foolish by most, only to become an unbelievably successful product. Also as I mentioned, if the team believes that Lean-style innovation accounting will be not applicable in your scenario, then you have one less way of answering questions early on.

Is there really no way to tell if an idea is brilliant or completely irresponsible? Any suggestions?