Tools and Estimators to Support Hand Counting

What if ballots are counted in a central counting location? How about absentee and early votes — there can be thousands of those to count?

These questions are very similar to the Findings presented in, “Works in Any County, Any Size” beginning on page 271 of the eManual. We provide an estimating tool to provide the same details as the Hand Count Estimator, except for single- or large-locations used for counting ballots.

The screenshot above is a portion of the full spreadsheet. It shows a Wyoming county (about 20,000 registered voters) with absentee and early absentee ballots. The results are based on counting these ballots in 4 hours with teams of 4 counters.

Looking at the column on the right, the estimates show that 188 counters are needed, separated into 47 teams. How much space will that take, this estimator includes an estimation of square footage needed.

Refer to page 264 of the eManual for a more detailed explanation of the Hand Count Estimator. And use the Library Annex below for more documentation about this estimator tool.

Link to Central Counting Location Estimator

Whether working on a cost comparison, planning for an official election, or putting together emergency supplies, it is essential to have an approximation of the supplies and forms needed for hand counting in the county.

Estimating forms and supplies is based on the same data used in other estimators: historical voter turnout. This historical data will help to project the anticipated voter turnout per polling place. The other projections needed are anticipated number of “choices” on the ballot (total candidates plus lines for write-ins), and whether there will be “questions” (Yes/No options).

This estimator will then estimate which forms are needed and in what quantities for each polling place. It will also estimate the supplies per polling place down to the number of binder clips and Sharpies.

The total row will provide the “shopping list,” while the individual polling place (precinct) rows will save the election official time in sorting the forms and supplies and packing for each location.

The screenshot above is a small snip of the full spreadsheet.

Refer to page 265 of the eManual for a more detailed explanation of the Forms and Supplies Estimator. And use the Library Annex below for more documentation about this estimator tool.

Link to Forms and Supplies Estimator

Yes, and creating visualization charts illustrating voter turnout is a great tool for presentations that include this topic.

While this tool is not an estimator, it uses the same data as the estimators: historical voter turnout. What this tool does is create visualizations.

Most election officials who object to hand counting probably do so because they feel they are already stretched thin for EVERY election. And they probably are because, consider this: they have to do all the same tasks for EVERY election no matter if the turnout is 10,000 voters or 100,000 voters. Why? Because using voting machines requires the same steps for EVERY election.

Machines are pulled out of storage, programmed, tested, delivered to polling places, recovered after the election, tested again, and returned to storage with 24/7 security. That’s just the basics. If a “glitch” is encountered anywhere along the way, reprogramming, issues with ballots, or myriad tasks could arise adding to the burden of voting machines and, possibly, requiring some ballots to be dropped in “Box 3” and counted at the clerk’s office.

Compare with hand counting: supplies are already being packed for polling places; some additional supplies will be required. There’s no programming, and no testing before or after the election. A “glitch” might be a shortage of supplies, but any ballot issues are taken care of at the polling place and everything is counted.

Remember, Presidential Elections Only Happen Every Four Years

The other reality is that election officials are putting in a lot of work for the elections between Presidential elections. The money and time spent for these ‘in-between’ elections is ridiculous and wasteful.

Look at the bar chart above. Cape Girardeau County in Missouri has about 55,000 registered voters. Each vertical bar is the voter turnout for a precinct/polling place. Three elections are displayed: a presidential held once every 4 years (dark blue), a municipal held yearly (red), and a mid-term primary held every 2 years (light blue).

The horizontal lines represent counting teams – in this example for every 600 ballots cast: purple is 600 ballots (recommend 1 counting team), gold is 1,200 ballots (recommend 2 counting teams), and green is 1,800 ballots (recommend 3 counting teams).

Cape voters should be aghast that they are paying for expensive voting equipment when only 1 precinct had more than 600 ballots cast for municipal elections (red), and 3 precincts were more than 600 ballots cast for in the primary (light blue).

What is remarkable, though, for the people of the Cape, is how few precincts need more than 1 or 2 counting teams for a presidential election (dark blue). With 34 precincts (voting locations), Cape needs 136 counters to put 1 counting team per location. But if Cape doubled that number, added another 136 just for the presidential election every 4 years, they would have enough people to HAND COUNT ballots and stop paying for voting machines. With 55,000 registered voters, are there 272 people in Cape Girardeau County willing to hand count their elections?

What About Visualizations for Larger Counties?

Look at the chart above for St. Charles County, Missouri, which has about 292,000 registered voters. The first visualization shows voter turnout from the 2020 (blue) and 2024 (red) presidential elections. So, this does not compare election types like we did with Cape Girardeau, it is showing turnout for presidential elections. You might also notice that there are less red vertical bars – this is because the county reallocated polling places and reduced the total by about 7 locations.

The horizontal lines show counting teams, but this chart is in increments of 400 ballots cast (which would equate to 8 hours).

For a presidential election that occurs once every 4 years, it appears most polling locations will benefit from having 2 counting teams. Perhaps about a third of locations would need 3 counting teams, but very few would need 4 or more counting teams. Again, this is based on counting all the ballots within 8 hours.

Now compare the chart above showing St. Charles voter turnout for municipal elections which occur every year. These elections are easily counted by 1 counting team per polling place for the majority with about a third benefiting from having 2 counting teams.

How many counters would St. Charles need for a municipal election? About 550, including counting the absentee ballots. That’s 0.2% of the registered voters in the county. And what about once every 4 years for a presidential election? Probably around 1,550 counters, or 0.5% of registered voters. That means that less than 1% of registered voters in St. Charles County could count all the ballots for an election in 8 hours.

Refer to page 271 of the eManual for a more detailed explanation of visualizing voter turnout. And use the Library Annex below for more documentation about this visualization tool.

Link to Visualization: Voter Turnout

Missouri has statutes about the final steps of an election before it is certified at the county level. That is the work done by a Verification Board.

Read more about Verification Boards beginning on page 278 of the eManual. It seems to have become a ‘lost art’ in Missouri, with election officials using election judges or a ‘bipartisan team’ to do the certification. Not to mention that the People do not have access to much of the election result data, having to rely on voting machine reports posted on county websites.

In response to that, this tool was developed for use by both verification boards and individuals to review and verify election results. Even though we cannot see ‘inside’ the machines, there is information that can be monitored for accuracy.

The screenshot above is one tab of the ERV tool. It is information from an official Verification Board for the 2024 presidential election. Note the red box drawn around three cells in the bottom-right. The instructions state that these three totals must match, but they do not. There is a discrepancy of 53 ballots.

Keep in mind this information was compiled after the election official and staff, all the elections judges, and the bipartisan team had completed their work and felt that everything was accurate. It is the job of the Verification Board to “check the math” and, because of this spreadsheet, the error was detected.

The resolution was that the election official realized the test deck of 53 ballots which had been run through a tabulator prior to election day had not been cleared out. So those numbers were being included in the election results. Once the correction was made, the results were reconciled and certified.

Link to ERV