It seems to me that there are three possible sources of error within the report presented in Frankham's post.
1. Printer's error.
2. Mechanical error in the "punching" process.
3. Voter error.
A fourth possible error source, where most of the attention seems to be focused, is in the mechanical reader process.
For conspiracy buffs, yet a fifth possible error source could be some nefarious insider deliberately producing mis-aligned ballots and then carefully sneaking them into a "targeted precinct" and surrepetitiously mixing them in with the proper ballots before the voting process began.
Assessing the relative merit of these error sources doesn't seem to have been done. Instead, partisan interests have attacked that part of the overall process within which they perceive the greatest possible benefit for their interest.
That partisanship has created a situation in which the "smoke and mirrors" may well obscure, subvert, or even prevent, a serious effort to determine what actually happened.
The interesting thing is the manner in which people presently infused and enthused with partisanship now seem to find it so easy to dismiss thoughts from persons who don't share their particular brand of partisanship. It seems to be a form of what a professor acquaintance-turned-politician once called the "struggle for advantage", the root of most forms of prejudice and apartheide in the world.