In a large set of numbers that span a few orders of magnitude (example: the amounts of money in each account of a bank), the first digit of any random account is much more likely to be a 1, instead of evenly being distributed 1-9 like you might expect.
1 is the most likely, then 2, then 3, then 4, etc.
A range of numbers will include a 1 in the first digit before it includes a 2, and a 2 before it includes a 3. If you have numbers from 1-250, 1 and 2 are the only possible first digits for three digit numbers (100-250), while two digit numbers would have an even split of numbers 1-9. If you go into the thousands, you have to include 1000-1999 before 2000-2999, which comes before 3000-3999, etc. The result is that numbers with a 1 first digit will represent a larger % of possible numbers in the range.
Explaining it would be hard if you don't have a strong math background, you'd likely zone out, so instead I'll just ELI5 the outcome of Benford's law.
When you have large sets of organically generated numbers, they follow a pattern when analyzed (benford's law is that pattern)
That pattern is that when you look at the first digit of every number in that set [e.g. "547" has "5" as the first digit] and add up the totals, you expect to see ~30% of the numbers starting with 1, 17% of the numbers starting with 2, 12% starting with 3, etc.
When you look at the vote counts in precincts across the nation they all follow that distribution (Benfords Law). BUT when you look at vote counts in precincts that Joe cheated in (like Milwaukee WI) all of a sudden every candidates vote counts follow that distribution curve except Joe's which grossly violate it. (See graph here)
This is evidence that Joes numbers were man made and not naturally occurring. Because the sample sizes are so high it would be statistically improbable for this to happen on its own
ELI5: It's a bar chart that is used to indicate integrity of data.
In other words, it's a dual-bar chart that has a high-to-low curve along the graph. A dataset is then divided into leading digits (1-9) and put in the second bar on the graph for comparison to the first.
If they match, the data is likely kosher. If they are considerably off, then it may indicate fraud. It's especially utilized in determining accounting fraud within minutes and was used in determining reliability of Iran's election integrity.
Biden's national BLaw chart looks normal (as does Trump's), but his battleground cities he's now leading in show a horrible correlation. Trump's, meanwhile, is totally normal.
Fraud exists in those blue cities, and we know it. It's a pattern that keeps showing up.
In a large set of numbers that span a few orders of magnitude (example: the amounts of money in each account of a bank), the first digit of any random account is much more likely to be a 1, instead of evenly being distributed 1-9 like you might expect.
Adding to this:
1 is the most likely, then 2, then 3, then 4, etc.
A range of numbers will include a 1 in the first digit before it includes a 2, and a 2 before it includes a 3. If you have numbers from 1-250, 1 and 2 are the only possible first digits for three digit numbers (100-250), while two digit numbers would have an even split of numbers 1-9. If you go into the thousands, you have to include 1000-1999 before 2000-2999, which comes before 3000-3999, etc. The result is that numbers with a 1 first digit will represent a larger % of possible numbers in the range.
law regarding the probability of the appearances of leading digits (1-9) in non symbolic data sets.
Love that you replied-still don't understand.
Explaining it would be hard if you don't have a strong math background, you'd likely zone out, so instead I'll just ELI5 the outcome of Benford's law.
When you have large sets of organically generated numbers, they follow a pattern when analyzed (benford's law is that pattern)
That pattern is that when you look at the first digit of every number in that set [e.g. "547" has "5" as the first digit] and add up the totals, you expect to see ~30% of the numbers starting with 1, 17% of the numbers starting with 2, 12% starting with 3, etc.
When you look at the vote counts in precincts across the nation they all follow that distribution (Benfords Law). BUT when you look at vote counts in precincts that Joe cheated in (like Milwaukee WI) all of a sudden every candidates vote counts follow that distribution curve except Joe's which grossly violate it. (See graph here)
This is evidence that Joes numbers were man made and not naturally occurring. Because the sample sizes are so high it would be statistically improbable for this to happen on its own
Thank you!!
https://www.youtube.com/watch?v=7uhAn19V1EY
ELI5: It's a bar chart that is used to indicate integrity of data.
In other words, it's a dual-bar chart that has a high-to-low curve along the graph. A dataset is then divided into leading digits (1-9) and put in the second bar on the graph for comparison to the first.
If they match, the data is likely kosher. If they are considerably off, then it may indicate fraud. It's especially utilized in determining accounting fraud within minutes and was used in determining reliability of Iran's election integrity.
Biden's national BLaw chart looks normal (as does Trump's), but his battleground cities he's now leading in show a horrible correlation. Trump's, meanwhile, is totally normal.
Fraud exists in those blue cities, and we know it. It's a pattern that keeps showing up.
Thank you!