People can make correct predictions, but there are tons of nuances around them. You could have predicted the Note 7 disaster, but Samsung's stock is up 27% from its pre-incident price. Maybe you knew about VW's problems years ago. Oh, too bad, their stock is up over 50% since 2011. Oh, you knew about the Toyota mat problems? That entire incident amounted to a nearly invisible blip on the stock price.
Betting nearly a billion dollars of other peoples' money on one single thing is absolutely insane. There are a million ways to be completely right and still lose everything.
Not disagreeing with your points about the million ways you can be right (and be unable to execute on it), though. Just that there are about a billion ways to be wrong, and a lot of those come from not taking a broad enough perspective about the impact. Burry did the research, saw the systemic complexity and dominoes that would fall, which perhaps gave him a little more confidence than a CNBC blowhard just looking at some new high in home buying and calling a market top. But he absolutely could have been wrong if he executed poorly.