That's my point. The paper didn't predict bad things happening. It specifically said that banks were holding higher level tranches of CDOs. Because they're MTM

That's my point. The paper didn't predict bad things happening. It specifically said that banks were holding higher level tranches of CDOs. Because they're MTM accounted they could suddenly drop in value if default rates increase. This, in turn, could cause other banks to suspect their creditworthiness and pull their funding lines. In other words, it was very specific about what it thought a collapse mechanism was and it turned out to be spot on. I don't believe they were either lucky or prescient. I think that they were well informed and joined the dots in a way that few others had.