f115r / Hand @ or Hand 3 ?

[rikforto]

Twelve lines? That is unlikely to be a big enough sample to be able to make a conclusive statistical analysis that the writer is someone different.

Without addressing the other points, LFD's argument is not statistical in nature and doesn't fall down there. She is arguing for qualitive differences to identify the hands by, and this is very normal methodology accepted by her peers. Essentially, if key markers shift for a passage, you can ascribe a different hand to it. It is certainly possible to disagree with her judgements---for my part I find her identifications extremely subtle at best, though I lack any training here---but a statistical test is not at issue.