Part 2 of 3

I’m back again with another entry in the Secret Decoder Ring series that you can use to decode and discard pretty much any theory along the lines of “Election 2020 Stolen via Fancy Technology.”

In Part 1, I boiled down one whole family of theories to its essentials — a conspiracy of election officials (EOs) formed to steal the presidential election — and pointed out that such a conspiracy could do its work with less risk and a lot less hassle with basic dirty tricks that pre-date even the very first voting machines.

And I’ll add a one final point here before we move on.

Suppose you really are concerned about such a conspiracy, and you really think that the members of this evil fellowship really are intent on using detectable technology-based methods rather than undetectable paper ballot tampering. Well then, here’s a much more simple technology-based method that’s more believable than the complicated stories.

A bent EO does not need to mess with arcane voting system features and mathematical vote shifting and goodness knows what else has been imagined. All they need to do is tamper with the output of the voting system, which after all is not much more than a table of numbers, that ordinary tools like text editors and spreadsheets can tinker with to get the exact same results (i.e., some specific desired false vote totals) — with a lot less hassle.

So, please, forget about all the election management system software mumbo-jumbo — no sensible election fraud perpetrator would do it that way; and a really sensible crooked Election Official would go after the paper ballots that are in their custody, and not mess with the technology. (And again, in our experience “crooked election official” is an oxymoron.)

Now, let’s move on to another theory, in which it is not the election officials falsifying election results using digital tools, but rather shadowy technology people who have embedded some sneaky hidden code into voting machine software, and/or election management system (EMS) software to steal the election right under the noses of the honest hard working EOs.  Is this a more credible theory?

Slightly.

First, let’s be clear: there is plenty of evil software in the world, and plenty of evil-doers too: keyloggers in computer games that also steal your passwords; ransomware tools used by cyber criminals to extort money; the Solarwinds hack that put a “sniffer” on many thousands of sensitive networks, and more. Nearly every kind of software you use also has the possibility of containing malicious code; or if it doesn’t at first, it can have a bug that’s a security vulnerability that opens the door to inserting some malicious code into your copy of the software.

And of course, these basic facts of life in computing also apply to software in voting machines, ballot scanners, and EMSs. The scanners’ software could, for example, have hidden code that did get used to intentionally incorrectly record a vote for “candidate A” as a vote for “candidate B”. That’s as possible as any other software evil deed that you can imagine — it’s all software, it’s all possible, but that possibility doesn’t mean it actually happened (especially in this case of the 2020 election).

Part 2 of Using the Decoder Ring

You should not be persuaded by the “Evil Software” theory.

Here is why.  These basic facts of life in computing are not something new to elections. Quite the reverse!

Everybody in election technology land knows this.  That’s why we’ve been moving away from invisibly hackable all-computer paperless voting systems, to better systems with paper ballots, modern high speed scanners for rapid counts, and support for people to cross-check the computer’s work and detect when the software made mistakes (accidental or malicious, or anything else).

The fact that (in most of the US) elections are all-paper ballots, with computerized counts that can be cross-checked manually, makes the job of Evil Software pretty difficult, if the Evil Software’s developers want the software to do its evil work and not get caught.

It’s difficult because with honest election officials, the Evil Software needs to be able to “read minds” and “predict the future, in order to be effective.

I hope that broad statement convinces you that such election stealing Evil Software is unlikely, but let me explain what I mean.

  • Cheating software needs to know when to cheat and when not to cheat, in order to avoid detection — as the Volkswagen emissions testing software developers learned the hard way years ago.
  • In elections, every ballot counting machine is tested for errors, to ensure that during a controlled test with test ballots, the vote counts are exactly as expected from human counts of the same ballots.
  • Cheating software would have to know when the people operating them were testing or not — that is, reading the minds of the people using a voting system in test mode.
  • You might suppose that clever software could be more predictive, like cheating only when it can tell that it is Election Day; but the fact is that the machines are operated for a period of weeks, and both testing and real usage can happen on the same day.
  • And of course, the testing itself could include (for realism) setting the system clock to pretend to be Election Day.

So that’s what I meant by reading minds.  Now, here is why I said “predict the future.”

When ballot counting systems’ work is cross-checked by people, it’s done with a sample of real counted ballots that’s chosen randomly, after all the vote counting is complete.

  • If the post-tabulation selected ballots included some ballots that the Evil Software intentionally mis-counted, then that miscount would be detected when the auditors’ vote counts differed from the vote counts of the Evil Software.
  • So, the Evil Software would need to know ahead of time which ballots to not cheat the count of lest they be chosen in the post-tabulation selection.
  • Did I mention the ballot selection is particularly and intentionally random?  You see the issue.

Of course, a dedicated conspiracy theorist can always come up with additional theories to work around sensible explanations, so I am not going to venture down every possible rabbit hole.  Instead, I just want you to understand the point is to use common sense to see that beyond the claim of “vote stealing software” are details that are very hard for Evil Software developers to address.

However, I’ll tackle one more: the variant theory where it is not the tabulators with the Evil Software, but instead, the central brains of a voting system, what we call the “election management system” (EMS).

  • Sure, EMS software could be evil in its critical function of taking the ballot-counters’ tallies and adding them all up.
  • EMS Evil Software could just intentionally add up the tallies wrongly, to favor one candidate.

But remember the point about not getting caught?

  • The vote tally data itself is public, whether paper tapes posted publicly in voting places, or data posted during publicly observed central counting of ballots — all before an EMS swings to action to stack up all the vote tallies.
  • Anybody who wants to, can take the public raw data and add up the tallies themselves to get the true vote totals.

That’s a simple fact that makes all the Evil EMS theories look pretty silly. 

Yet, if you didn’t know the facts about how ballot counting is conducted for public transparency, then the stories sound possible.

The same is true when you don’t know about the very detailed activities in the “canvass” process that states do after all ballots are counted, but before certifying election results — a lot of cross checks to catch possible errors by election officials or election technology.

So, like I explained in Part 1, you don’t need to know a bunch of techie stuff to be skeptical of these election-stolen-by-technology stories; you just need to learn a few facts and apply your common sense.

A bit of new information, and your common sense:

The Two Pillars of the not-so-secret decoder ring.