It seems to me your criticisms very much fall into "letting the perfect be the enemy of the good" territory. States have conducted some version of vote-by-mail or absentee balloting for decades, and there's no evidence I'm aware of that either of these have, in practice, materially increased voter fraud. Furthermore, studies on existing voter fraud conducted by groups like the Brennan Center for Justice and the Heritage Foundation have concluded the incident rates are around 0.0025% -- and that's the high end of the estimates. Even if your concern that a push to move most states to vote-by-mail in the 2020 election causes that number to go up substantially proves valid, how likely is it, truly, that it increases by the two orders of magnitude it would take to bring it up to a quarter of a percent -- and that such an incredible increase goes essentially unnoticed and unchallenged?