2024 US Election Analysis

 Clark County is the most populous county in Nevada according to the U.S. Census Bureau. It is home to approximately 70% of the state's population and includes world-famous Las Vegas.

Of the 2.3 million residents, about 1.5 million are registered to vote. This includes approximately 485,000 registered Democrats, around 393,000 registered Republicans, and a combined total of about 646,000 voters registered with third parties or with no party affiliation.

Using the Cast Vote Record (“CVR”) data for Clark County, the Election Truth Alliance (ETA) has identified voting pattern anomalies of significant concern. Our analysis uncovers unusual phenomena in the Early Voting results not present in Election Day voting or Mail-In Voting results.

• Drop-Off Difference: The term "drop-off votes" refers to the votes cast for a presidential candidate versus the votes cast for a down-ballot candidate of the same party. In Clark County, as was the case across the swing states in the 2024 U.S. Presidential Election, there is a significant difference between Trump’s drop-off rate (+10.54%) and Harris’s drop-off rate (+1.07%).

• Increased Volume of Votes Linked to Greater Discrepancies: The greater number of ballots cast and processed in Early Voting, the more Trump’s vote count increased while Harris’s vote count decreased. The pattern is more distinct (closer to 60% votes for Trump, closer to 40% votes for Harris) with more ballots processed by a given tabulator.

• Abnormal Clustering: In contrast to Election Day voting, Early Vote results display an unusual pattern: once approximately 250 ballots have been processed a visible shift is observed, resulting in a high degree of clustering and unusual uniformity. This is a departure from expected human voting behavior. 

According to Clark County CVR data, a total of 1,033,285 votes were cast in Clark County across the three types of voting: mail-in, early voting, and Election Day voting. Statewide, the total number of votes cast in Nevada was 1,484,840.

• Mail-In Voters (443,823 total): 
  • Harris received 61% of recorded presidential votes (271,455 votes) 
  • Trump received 36% of recorded presidential votes (160,824 vote)
• Early Voters (395,438 total): 
  • Harris received 40% of recorded presidential votes (156,705 votes)
  • Trump received 59% of recorded presidential votes (234,231)
• Election Day Voters (194,024 total): 
  • Harris received 47% of recorded presidential votes (91,831 votes) 
  • Trump received 50% of recorded presidential votes (97,662 votes)

The analysis stems from CVR data Clark County posted on its website. This data is publicly available for download. Links and archived links are provided here, and listed in detail in the final section of this document.

CVR data is significant because it shows, on a ballot-by-ballot basis, how votes in Clark County were recorded. It notes whether ballots were cast as Mail-In Votes, Early Votes, or Election Day votes. (Provisional ballots are not included.) Notably, Clark County CVR data shows the numbers associated with a given tabulation machine, labeled ‘TabulatorNum’, including how many ballots were processed. All mail-in ballots were tabulated by 6 machines.

The term "drop-off votes" refer to the votes cast for a presidential candidate versus the votes cast for a down-ballot candidate of the same party. The term highlights the number of voters who voted for president but skipped voting in lower races, like the Senate or House races.

In our analysis of Clark County, the drop-off rate is measured between votes for the Presidential race and the Nevada Senate race, which had Jacky Rosen (D) running against Sam Brown (R). The drop-off rate is calculated by finding the difference between the number of votes cast for president and the votes cast for the Senate race, and then dividing this by the total number of presidential votes for each party.

While some drop-off between the presidential and down-ballot races is expected, SMART Elections – a non-partisan group focused on election security – notes that in the 2024 U.S. Presidential Election, drop-off rates were very different on the Republican versus the Democratic side. Drop-off rates for the Republican side were very high, whereas the drop-off rates for the Democratic side were quite low or even into the negative.

They also found that drop-off rates in swing states tend to be significantly higher compared to non-swing states. This aligns with findings from other independent analysts who examined voting patterns in similar regions.

There are several possible explanations for a difference in drop-off rates, including:

• Differential between popularity of candidates at the top of the ticket versus down-ballot candidates
• “Split ticket” voting, where a voter casts a ballot for candidates of multiple parties
• Targeted political messaging directed uniquely towards swing states.

To date, the Election Truth Alliance’s analysis has not focused heavily on drop-off rates. Like most swing states, Nevada shows a disparity between Trump and Harris’s drop-off rates: +9.87% for Trump, and +0.58% for Harris. 

The Clark County CVR data allows a closer examination of drop-off votes by voting type: Mail-In, Early Votes, and Election Day. It has enabled us to “lift the hood” and look more closely at: 

• How drop-off votes break down by voting type in that county
• Any insights into what potential causes of the drop-off disparity may stand up to closer inspection. 

The breakdown of Clark County drop-off votes by voting type is below:

 When broken down by voting type, we see that there is an especially narrow margin between votes for Harris vs Rosen in the Early Voting data – only 89 votes compared to thousands for the other voting types. 

This prompted us to look more closely at the potential differences between the different voting types through a variety of analytic lenses. 

There are notable differences in voting data patterns across the three types of voting data available in Clark County’s Cast Vote Record. 

To some extent, this voting pattern is expected: there are differences in the populations who tend to vote using these different methods. However, there are patterns in the Early Votes as recorded that deviate significantly from organic human voting behavior. 

Human behavior is messy – including when we vote.  This scatterplot represents the variation that is expected from a large population. In the Election Day voting results, we see an expected degree of human voting behavior reflected

• After approximately 250 early voting ballots were processed by a voting machine, a shift is observed in the reported voting patterns;
• Instead of a chaotic, expected distribution, the vote percentages start to shift more heavily in Trump’s favor; 
• This pattern is not found in Election Day votes. 

The ‘scatterplot’ above is useful in identifying this phenomenon; however, the format makes it difficult to see how the “occurrences” (the individual circles representing each voting machine) are distributed among the two presidential candidates – meaning: 

• Each voting machine processed ballots and calculated what percentage of those ballots a presidential candidate won. (The percentage range of cast/processed ballots is the horizontal x axis.)

• We show the number of tabulators (100, 200, 300). (This is the vertical y axis.)
  
• The colored bars represent how many voting machines produced a result; showing a candidate (blue or red) receiving an approximate percentage of total votes. 

Grouping the results this way, we would expect to see something close to a normal bell curve. Normal data that is not manipulated and follows natural variability tends form a bell curve, or normal distribution.

This is the result we observe for votes cast for presidential candidates on Election Day: