+9 more
posted on Nov, 13 2020 @ 08:34 AM
There has been a
story floating around that the voting data from The New York Times shows that votes were switched from Trump to Biden. I decided
to download the
data for each state directly from NYT and write
my own script in node.js to confirm for myself whether this claim was true or not. I was looking for this data anyway because I wanted to check if the
large spike of Biden votes which occurred early in the morning was actually in the official data. Writing my own script would make it easier to see if
there were any bugs in the python script first used to find these anomalies in the data.
What I found is that Biden also loses a large number of votes, in fact more votes are taken from Biden than from Trump. Part of this mystery can be
explained by the way the data is formatted. We are provided with the total vote count along with two fractions which can be used to calculate the
number of votes for Trump and Biden. For example if Trump has 40% of the votes and Biden has 60% then the fractional value for Trump is 0.4 and it's
0.6 for Biden. You can multiply those fractions by the total vote count to get the number of votes for each candidate with reasonable accuracy.
The problem is each fraction has a maximum of 3 decimal places so it's not perfectly accurate and votes may be "lost" in some cases due to rounding
errors. However those rounding errors are far too small to explain most of these anomalies, in a few instances several hundred thousand votes suddenly
vanished. That means something besides rounding errors were involved, maybe glitches, typos, miscounts, etc. I decided the only real way to figure out
what was going on here was to visualize it by graphing the data. Here is my analysis for 6 of the swing states which had some of these large
anomalies.
As you can see from the first chart showing what happened in Virginia, Biden lost over 300 thousand votes, then gained around 400 thousand votes, then
lost over 300 thousand votes, then gained over half a million votes, resulting in a massive Biden lead, all in a fairly short period of time. Such a
high amount of volatility so late in the voting process would suggest to me something strange was happening in Virginia. It also turns out that the
Biden spikes do appear in this data, not only in Michigan and Wisconsin, but also in Georgia. It's rather coincidental that these anomalies favor
Biden so heavily.
Click on images to see them in full size and read my analysis of each state.
Virginia:
Michigan:
Wisconsin:
Georgia:
Pennsylvania:
Nevada: