Here is an analysis of the data from the 2024 Chaffee County election.
The first analysis examines batches of votes for Trump and Harris to determine whether any ballots were swapped to manipulate the results. Ballots are scanned in batches of up to 50, though some batches are much smaller. Instances where a batch of votes is heavily skewed in one direction occur only in smaller batches. Therefore, we find no evidence of ballot batches being exchanged.
Here, we analyze one county commissioner race, examining the total votes each candidate received and the vote distribution by precinct. McCracken suffered a significant loss in Salida, precincts 007, 008, and 009.
Here’s a table to identify the precincts:
Precinct 7041308001 is Buena Vista Ward 1
Precinct 7041308002 is Buena Vista Ward 2
Precinct 7041308003 is Buena Vista Ward 3
Precinct 7041308004 is Nathrop
Precinct 7041308005 is Salida addresses outside of the City
Precinct 7046008006 is Salida addresses outside of the City
Precinct 7041308007 is Salida Ward 1
Precinct 7041308008 is Salida Ward 2
Precinct 7041308009 is Salida Ward 3
Precinct 7041308010 is Nathrop
Precinct 7041308011 is Poncha Springs
Votes for McCracken and Armstrong:
McCracken Armstrong
6321 7540
Here we look at the correlation between Trump and McCracken votes and Harris and Armstrong votes. What follows is a correlation heat map.
Votes for McCracken Armstrong Trump Harris:
McCracken Armstrong Trump Harris
1 6321 7540 6034 7992
Trump voters who did not vote for McCracken: 633
Harris voters who did not vote for Armstrong: 1017
Armstrong voters who voted for Trump: 363
McCracken voters who voted for Harris: 613
Chi-squared test results: 95% confidence showing that the Trump/McCracken distributions are significantly different.
Chi-squared statistic: 8969.812320761532
p-value: 0.0
Degrees of freedom: 1
Expected frequencies:
[[4882.03444976 3713.96555024]
[3426.96555024 2607.03444976]]
The distributions are significantly different (reject null hypothesis).
Chi-squared test results: 95% confidence showing that the Harris/Armstrong distributions are significantly different.
Chi-squared statistic: 9003.550892893029
p-value: 0.0
Degrees of freedom: 1
Expected frequencies:
[[3216.91182502 3421.08817498]
[3873.08817498 4118.91182502]]
The distributions are significantly different (reject null hypothesis).
Finally, here’s an analysis of of the voter rolls between August and November 2024.
18,384 active and inactive voters were registered in August.
4,647 of those registered in August did not vote in November.
832 new voters were added to the rolls in November and voted.
14,569 did vote in November.
218 of the 832 new voters voted in person, not by mail.
I do not have addresses for 70 of the 832 new voters. Google could not find another 127 of the 832 addresses; however, Google is not very reliable.
Here’s a map of 635 of the newly registered voters: Newly Registared Voters in 2024 - Google My Maps
The new voters appear randomly scattered across the county.
Here’s a k-means cluster analysis. It shows precincts 2 and 4 strongly Republican and 7 through 9 strongly Democratic. The other precincts are neutral.
Trump Harris McCracken Armstrong Nyberg Lucrezi cluster
Precinct
7041308001 0.466429 0.490000 0.506429 0.439286 0.438571 0.469286 0
7041308002 0.533065 0.418631 0.553767 0.401380 0.518689 0.391029 2
7041308003 0.413208 0.528302 0.452830 0.486792 0.393396 0.474528 0
7041308004 0.539474 0.413158 0.571930 0.388596 0.509649 0.382456 2
7041308005 0.455731 0.514938 0.454644 0.491581 0.422053 0.474742 0
7041308007 0.277851 0.687003 0.293767 0.640584 0.264589 0.644562 1
7041308008 0.239517 0.709311 0.248045 0.698650 0.209666 0.683724 1
7041308009 0.317784 0.648688 0.325802 0.615889 0.279883 0.614431 1
7041308010 0.472249 0.494645 0.508277 0.450828 0.455696 0.452775 0
7041308011 0.410382 0.539667 0.407444 0.525955 0.367287 0.526934 0
7046008006 0.417937 0.552466 0.451121 0.510314 0.388341 0.522870 0
