1. RECORDING our votes

A paper record of each voter’s selections is necessary if anyone is ever going to be able to recount an election or check to make sure the computerized tabulators worked correctly. As a result, almost all Americans’ votes are now recorded on paper.

Each jurisdiction makes its voters use one of two methods to record their selections.

  1. Hand-marked paper ballots. In jurisdictions that use paper-ballot systems, voters use pens to create the permanent record of their votes. Those voters know, with 100% confidence, that their votes were accurately recorded. Any mistakes in recording their votes are their own fault; there’s no opportunity for interference by others.
  2. Computer-printed ballots. In some polling places–particularly early-voting locations in the larger cities– voters are provided access to a computer to record of their votes. Sometimes these computers print out individual paper ballots; in older machines, the votes are printed on a tape that the voter can view only through a little window.

There’s not much to say about hand-marked paper ballots–the ‘technology’ is obvious, old, and well-tested. If the pen is dry, the voter notices and asks for a different pen. If the voter accidentally marks a vote for a candidate they don’t like, they can ask for a replacement ballot. If the voter made marks the machine cannot read, the machine will spit the ballot out so it can be replaced with a correctly marked ballot. No one can grab the voter’s hand and make marks the voter doesn’t want.

Things get more complicated when voters use computers to mark the paper ballots. Computers can malfunction or be misprogrammed so that the printed ballots don’t accurately record the selections the voter touched on the monitor. Further, it’s entirely up to the voter to catch those misprints, because no one but the voter knows which votes the computer was supposed to print.

So voters must be vigilant. They cannot rely on the information displayed on the screen; a malfunctioning ballot printer could display one candidate’s name on the screen, but then print a ballot that fails to show a vote in that race; prints votes in a way the tabulator cannot read them, or record the wrong vote.

The most serious issues arise when the tabulators are designed to count only barcodes or QR codes. The printed ballots for those systems show human-readable text, which gives the voters the impression they can verify the printed votes, but that human-readable text will never be counted by any voting machine. The voting machine will count only the encoded votes, which the voter can neither read nor verify.

So poll workers need continuously to remind voters to check that the computer printed the ballots correctly before casting their votes. Poll workers must also take reports of errors seriously, and refrain from assuming they are all caused by voter error.  If the poll workers where you vote are not telling voters to verify the computer-printed PAPER ballots, contact your municipal clerk.

2. COUNTING our votes.

Some smaller Wisconsin jurisdictions still rely on hand counts, which routinely include a double-check validation count. Hand counts are never done by just one person, counting alone. Even preliminary election results are not reported until two or even more people have separately counted the votes and agree on the total.

But most of Wisconsin’s votes are counted by computers called tabulators. These machines count our votes only once, using one version of software that was written in proprietary secrecy by programmers employed by private companies. There is no double-check or verification. Our election officials hope and trust that those programmers are all authorized employees or contractors of the companies, but there’s no way for local officials to know if the voting-machine companies were hacked.  

A note on counting ballots versus counting votes: Tabulators count both. Wisconsin’s local elections officials are justifiably proud of their rigorous efforts to make sure the voting machines counted the correct number of ballots. This is a good thing; it means they are being alert for signs of what’s called ‘retail’ election fraud–voters casting additional ballots.

But that safeguard is powerless to detect or prevent ‘wholesale’ election fraud, in which the voting machines are misprogrammed to manipulate the vote totals for each candidate without altering the total number of ballots. Speak with your municipal clerk about the measures they use to verify that the tabulators counted the correct number of votes for each candidate. Chances are, they use no such measures, and choose instead to rely exclusively on hope and trust that nothing went wrong on Election Day.

Sources of electronic miscounts

Voting-machine miscounts are caused by the same things that can mess up any computer: 1) Malfunction; 2) Mistakes; and 3) Malicious interference.

Malfunction. Anyone who uses a computer knows that computers can do unexpected things because of power surges or outages, physical damage, or no apparent reason. In one case in New York City, tabulators overheated mid-day and lost calibration but kept counting ballots without counting any votes. Thousands of voters were disenfranchised, but the missing votes were not noticed until long after election results had been declared final. In a 2014 Wisconsin election in Stoughton (Dane County), dust bunnies inside a voting machine cast shadows on the ballots that the computer read as votes.

Mistakes. More than 600 voters in Medford, Wisconsin were disenfranchised in November 2004 when their voting-machine vendor neglected to program the machines to read straight-party ticket votes. The problem was not discovered until March 2005, when a political party consultant compiling data on registered voters noticed that the number of voters who cast ballots was much higher than the number of votes counted.
Thousands of Wisconsin absentee voters were disenfranchised in the November 2016 Presidential election when they marked their ballots with ink the machines could not detect, and poll workers failed to notice and fed the ballots into the tabulators anyway. The problem was particularly pronounced in Marinette and Racine Counties. Marinette officials corrected the miscount during the recount, but Racine officials did not acknowledge the problem even when observers at the recount pointed it out. Then, to the voters’ shock, the officials certified the same incorrect vote totals after the recount!

Malicious code. Wisconsin’s county and municipal clerks might maintain the tightest security they can, but the elections software security is not fully in their control. They are demonstrably ignorant about voting-system security. Private out-of-state corporations write and maintain the software that counts our votes. Wisconsin’s election clerks can do little more than trust that the vendors’ employees were all honest and competent, and that the vendors’ own security systems successfully resisted hackers. 

3. VERIFYING our election results. 

We wouldn’t have to worry about mistakes, malfunctions, or malicious code picking the winners of Wisconsin elections if our local election officials used our paper ballots to check the machines’ accuracy before declaring results final.

Election-Night results are only preliminary. Our local elections officials cannot declare the election results final (‘certify’ them) until they have waited for late-arriving absentee ballots and reviewed the records of the election to make sure everything is in order. Before they certify  the election results, they could verify the accuracy of the electronically counted results, and can correct any errors they discover. 

More than half the states as of 2018 require local election officials to check at least a few machines’ accuracy before they can declare election results final. Many states require audits of only a small percentage of the machines, which in seldom enough to make sure they are declaring the right winners. Even a small percentage, if randomly selected, might deter hacking. Other states go a little farther, wisely requiring local officials to expand the sample if they find problems in the first few machines.

Since 2014, however, the recognized best practice for election auditing is to conduct audits before certification that verify the machines identified the correct winners (not just spot-checks that confirm a few machines counted correctly.) These audits, sometimes called “risk-limiting audits”, or RLAs, use well-tested statistical formulas to enable election officials to make sure they’ve identified the right winners. Endorsed the the Presidential Commission on Elections Administration in 2014, and promoted by the federal Elections Assistance Commission and Department of Homeland Security, the method is gaining popularity. Colorado elections officials fully implemented the method in 2017, with Rhode Island and Virginia close behind, and other states moving ahead with plans to use it.

Wisconsin election officials do not attempt to verify Election-night vote totals. 

Go to any town, village, city, or county offices after any election. If you ask to observe the canvass (the procedure during which election officials review the results before they declare them final) you will notice that they check only that the computers counted the correct number of ballots, and that they just trust the vote totals.

Before they certify election results, officials in more than 25 other states routinely check the computers’ verdict against the votes marked on the paper ballots. But Wisconsin voters’ paper ballots remain sealed in plastic bags and stashed away in local storerooms, where their value in securing the integrity of Wisconsin’s elections will never be fulfilled.

Call your municipal and county clerk today and tell them “Verify before you certify!”

More information:

How IT experts tell us voting machines could be hacked, or compromised by corrupt insiders 

How Wisconsin’s officials could easily verify voting-machine output