I have just read the three articles, and comments from other posters, and here are my thoughts:

The articles do a good job of outlining the problem, and making clear that there is information that must be inferred, or estimated. The articles also follow the issue very logically, and do a credible job of assessing other existing theories.

However, they do make some logical presumptions that fit their theory, but require either an acceptance on the part of the reader or deserve more detailed explanation. Being an article rather than a presentation to a jury, this is both common and understandable.

Obviously if one accepts their basic conclusion, than their is no criminal activity, but it is reasonable to question whether the error is significant enough, or negligent enough to justify dismissal.

Note that 2 million could seriously have effected the various outcomes, and that some persons could raise the issue of negligence in a lawsuit. The article did touch on this.

Finally, while I think the theory is quite reasonable, I think the evidence would be lacking to establish it in a court of law (even by a civil standard). Not that it couldn't be, access to hard records, and the ability to examine individuals under oath (especially during depositions), could allow for more evidence to more firmly establish the facts.

Either way a very interesting read.