so based on publicly available data on websites run by state governments on the votes being tallied and the same date reported on by the associated press, ABC and other sources there are 5 things that happened in the election that are a virtual statistical impossibility other people have done the math and posted these graphs but I wanted to get it all in one place.
the first is batches of mail in ballots reported in Wisconsin and Michigan at around 4 am on Nov 4th which you can see in these graphs which are based on
ABC news' reporting on the ballots as tracked by 538
the count was reported by the associated press for michigan
in these two states batches of mail-in votes were counted one being around 140,000 votes and the other being around 50,000 votes, virtually 100% of these went for biden. Not only did trump basically get zero neither did 3rd party and write-ins get any either. If you're counting votes and therefor any batch of ballots you begin counting is randomly selected this is basically impossible. It's not just the way that the graph has a sharp 90 degree turn but it's that only biden made gains for that amount of votes leftist are trying to use other states graphs which show the count jumping up in a similar way as a large batch of ballots are counted, and then reported in a short span of time, but those examples all show trump making gains at the same point in time with a comparable trajectory to the ones biden gets. A batch of ballots in the 10s of thousands should not all go for 1 candidate as a function of probability
as an example let's assume we're counting votes that come in from a county of a city where biden generally enjoys a margin of 75% voter support and for the sake of simplicity disregard options other than biden and trump;
- Picking 2 ballots at random there is a 56.25% chance neither are for trump.
- Picking 3 at random it's a 42.18% chance.
- with 10 it's a 5% chance
- with 50 it's a 5.663216564269376e-7% chance
- with 100 it's a 3.207202185381504e-13% chance.
With 10,000 not to mention 50 or 100 thousand a 100% margin for one candidate is impossible. The math holds up even if we assume the ballots are coming in from somewhere with 90% or more support for biden with this many ballots unless they were coming from somewhere where literally 100% of voters support biden and only biden.
second is the data coming in for biden violates bedford's law. This is a law of math which states that in many naturally occurring collections of numbers, the leading digit is likely to be small. For example, in sets that obey the law, the number 1 appears as the leading significant digit about 30% of the time, while 9 appears as the leading significant digit less than 5% of the time. If the digits were distributed uniformly, they would each occur about 11.1% of the time. and this principle is used by the US state department and international authorities use to verify election integrity,
here's a mathematician that explains the concept
there's a program the source code of which is publicly available on github here
this aggregates and graphs data sets based on numbers the state governments have made publicly available and for milwakee we get this for biden's votes for comparison here's an election in Iran that was found to be fraudulent it's notably not nearly as bad as biden's and in multiple states only biden's data has this much or more variance from bedford's law. Trump 3rd party and write-ins conform more closely to the expected curve here's the rest of the Milwaukee data for example and the Github has the datasets for multiple counties in battleground states
the 3rd thing is Wisconsin voter turnout which is well documented and publically available for the last serval decades
the 4th is Bellweather counties. Election bellwether counties in the United States are counties which vote in alignment with the country as a whole in United States presidential elections, so that the county votes for the candidate who ultimately wins the election. The strongest bellwether counties are those which back the winning candidate in all, or almost all, elections. Trump won 9 of the top 10 most reliable. several of these failing to predict the results is extremely unlikely but the odds against this many being wrong are again astronomical.
Lastly there is a model based on turnout in the primaries that is used to predict the general results called the primary model. For the record, the Primary Model gets the electoral-vote winner right in 25 of the 27 elections from 1912 to 2016, or 92% of the time it's right and it showed trump would win: http://primarymodel.com/2020-1
None of this is conspiratorial or theoretical it's all based on publicly available data and widely accepted and professionally used math regarding statistics and probability anyone can do. Remember this when the media tries to say there's no evidence of fraud or widespread errors.
also long list of archives to examples of issues with mail-in ballots