[whatshisface]

You have to watch out for the "water dowser" effect. If you're in front of a group of people who have no ability to check your answers, it's very easy to convince them that you're always right. If a trusted person with good social skills came in and said, "don't work with him he's not earnest," how would the room ever find out whether he was or not? They'd be unlikely to trust him enough to end up in a situation where they'd be able to see for themselves.

[noonespecial]

It's also impossible to know if she's "set right".

It's not enough to never pick a "bad" team if you're also turning away 9 out of every 10 good ones.

[michaelmior]

That all depends on the return you get from the 1/10 good ones. If you do great with that one good team, it could be totally worth it. I could definitely imagine a scenario where picking no bad teams has higher value than picking more both good and bad.