Unless Jack Dorsey knows the future, I'm not sure you can fact check something that hasn't happened yet.
https://www.snopes.com/fact-check/mail-in-ballot-voter-fraud...
What's True
While no U.S. government agency officially compiles state-by-state data on voter fraud, and requirements for mail-in voting vary by state, analysis by elections experts shows that fraud is slightly more common with mail-in voting than in-person voting at polling places.
What's False All types of voter fraud in U.S. elections is minuscule in comparison to the number of ballots cast, according to elections experts. Taking that into consideration, it is problematic to make comparisons between types of ballot-casting systems and erroneous to claim mail-in voting "substantially" increases the risk of fraud.
So where is the line between slightly and substantial?
If fraud is more common with mail in voting and some states (or everyone?) converts entirely to mail in voting, how much will fraud increase overall?
Will it increase enough to change the overall results? With Michigan and Wisconsin being decided in 2016 by less than 1% of the vote, there's not much margin for error, fraud, or mistakes.
Trump is making an extraordinary claim. He must back up that claim, whether that's by revealing that there's a true plot against him; referencing historical data; or something else.
Even the above linked claim in snopes says fraud is more common with mail in ballots.
Claiming mail-in votes will be majority fraudulent, and by implication that the entire vote is invalid... is a much stronger claim, which IMO requires much stronger proof.
Given that mail-in ballots have been in used for a long time, there's a good history of data, so it's not predicting the future out of nothing, but based on past evidence.
The twitter fact-check link in fact goes into that precise thing.
