How poor statistics led to conviction for murder: the case of Lucia de Berk

I thought this might be of interest to some of you. Some time ago in my country, The Netherlands, a nurse was wrongly convicted for murder of several babies in her ward. The assertion was that the number of children who died on her watch was too high to be coincidence.

Statistical arguments played a role in her conviction. In the end the judges ruled that this evidence was not admissible (I think), but it's hard to believe that they were not in the least influenced by the statistician's argument.

After some time there was a movement for her release, led by philosopher of science Ton Derksen. The statistical argument was pretty much demolished. See here for a general treatment, and here for the scientific paper.

After spending a few years in prison, she was eventually retried and cleared of any wrongdoing (and compensated by a large sum). It's a tragic example of poor statistics leading to disastrous consequences for the people involved. At least, that's my admittedly non-expert opinion. What do you think?