All we know from the study is that the false positive rate was about 89%
Exactly. We don't know the false positive rate in the cases where the evidence was used. You can't claim that my speculation is not valid, yet yours is valid.
This isn't a case of a random subset of a larger population. In every case, there was a decision made whether or not to use the hair evidence. That decision was based on the evidence available. Thus, you can't assume that you have a random subset of the larger population. Thus your projection of an 11% false positive rate on the subset isn't valid.
Finally, a lack of other evidence would suggest a higher likelihood of innocence, which would imply a higher liklihood that the hair evidence was false.