Michigan will go red: Benford's law proves unilateral fraud in Detroit beyond a reasonable doubt
(media.patriots.win)
🛑 STOP THE STEAL 🛑
Comments (50)
sorted by:
DEFAULT COMMUNITIES
•
All
General
AskWin
Funny
Technology
Animals
Sports
Gaming
DIY
Health
Positive
Privacy
It’s a counter intuitive property of large data sets of random numbers. Consider this thought experiment: start counting up from 1 and pay attention to how many numbers you have counted start with a 1.
Count to 10, two of the ten numbers start with 1. Now double is and count to 20, about 50% start with 1. Count to 100 and the percentage starts dropping. Now to 200, shit it’s going up again. Slowly drops as you approach 1000, and then spikes up again to 2000. This cycle repeats infinitely. Most of the time 1 has a higher likely hood of showing up.
Mathematicians have analyzed this property and concluded that in a random data set you can expect 1/3 of your data points to start with 1.
To add:
The "distance" between 1 and 2 is "1". If you look how much of the initial number the distance is it is 100%. (1 out of 1 is 100%).
Now if you look at the distance between 2 and 3. It's also "1". But how much of the initial number (2) is this 1? It's 50% (1 out of 2 is 50%).
This goes up to the distance between 9 and 10. Again the distance is "1", but it's now only 1/9 of the initial number, so roughly 11%.
So you unintuitively need more "steps" to cross the distance from 1 to 2 compared to 9 and 10. This results in the observed behaviour that follows the law.