<<As evidence of this I cite the myriad posts on iVillage which proclaimed that 9902b is a slam dunk based on the outcomes of 9901/9902a.>>
Example of another type of bias, this one certainly not trivial. Certainly you don't take anonymous MB posts at face value.
<< I would bet that fewer than 10% of DNDN longs have ever even considered program-survival bias.>>
On whether the market considers these biases, I think they do, implicitly if not explicitly by name. Again, I think very few if anyone truly believes that 9902b is a certain success. The stock price reflects that.