Thus, it is not unreasonable to expect Trump’s Democratic opponent in 2020 to gain on Trump by over 20,000 votes in Pennsylvania during the period between Election Night and the final, official certification of the canvass. The key question is whether this kind of gain simply extends a lead that the Democratic candidate already has, comparable to what occurred in two statewide races in 2018. Or whether, instead, it cuts into a lead that Trump starts with on Election Night—and, if so, whether it is enough of a gain for Trump’s Democratic opponent to overcome Trump’s Election Night lead. In 2016, Hillary Clinton’s gain of 23,659 votes during the canvassing process was not enough to flip Pennsylvania to her column. Instead, it reduced a Trump lead of 67,951 in the state to “only” 44,292.6 But in 2020 a comparable gain for the Democrat could erase entirely a 21,000-vote Election Night lead for Trump, converting the result into a 2,500-vote margin of victory for the Democrat. Pennsylvania is hardly aberrational in producing this kind of gain for Democratic candidates during the canvassing process. Although this phenomenon is still not widely understood by the electorate generally, scholars and even the media have begun to take notice. In 2014, I published an article entitled The Big Blue Shift to draw attention to this development, hypothesizing that it is best explained as an unintended byproduct of electoral reforms adopted in the wake of the 2000 fiasco, most specifically the advent of provisional ballots and the increased use of absentee voting.7 (One possible factor is that provisional ballots, which became nationally mandated by the Help America Vote Act of 2002 and which are necessarily counted during the canvassing process after Election Night once their validity has been verified, tend to be cast by voters of demographic groups who support Democratic candidates. But while this factor undoubtedly contributes to the phenomenon, the number of provisional ballots generally is not large enough to account for the entirely of the “blue shift” phenomenon, and the remainder of the explanation is still uncertain.) Whatever the exact causal mechanism— we are still in the early stages of studying the phenomenon—this kind of “overtime” gain by Democrats, after Election Night and before final certification of the canvass, achieved national salience in the 2018 midterms.8

Tl;dr - this dude noticed votes mysteriously swinging to dems overnight in previous races, and states that provisional and absentee totals cannot fully account for it. The steal has been going on a long time.