A better understanding of subprime MBS loss

HousingWire carried the story about another rating agency revising its loss projections for subprime mortgage-backed securities. These firms have fallen under a bit of a cloud because some think that they should have been in the best position to see the subprime crisis before it hit. While they've done a great job of working quickly to downgrade hundreds of securities after the fact, they are still in the process of trying to fully understand how the value of these securities is likely to change in the future.

I've been talking to executives inside two firms that do this sort of work for a living and there is a lot of background noise audible on their sides of the calls. I'm not saying it sounds like panic, but there is definitely a lot of activity going on inside these firms.

I understand the whole "black box" idea, and how proprietary models are the only asset these companies have to deploy in the marketplace, but I think it's time now for somebody to step up and explain why we should accept their predications. Or admit that historical analysis no longer provides any insight into the events of the last six months and that without new assumptions based on something someone pulled out of a hat, there is no way to know what the next six months will hold. But then, this isn't really compatible withe the concept of investor confidence, is it?