This is Part 4 of 4 of a series on election theft
The Democratic Primary
A year ago, before the first primary vote had been cast, the usual monied interests had lined up behind the usual establishment candidates. But two upstarts with anti-establishment messages were attracting genuine, grass-roots crowds with genuine, grass-roots enthusiasm. These were Bernie Sanders and Donald Trump. Sanders's appearances were more popular even than Trump's, and media polls showed he had a much larger following. And yet news coverage in the "liberal" press gave Trump about ten times as many column-inches as Sanders. Trump's flamboyance was a subject the press loved to harp on, albeit with coverage that was harshly critical in the press outlets of the Eastern establishment. The real injustice was to Sanders, whose popular successes were ignored where possible and marginalized in the few reported articles. This story was documented in detail by an article by Thomas Frank in Harper's last month, titled Swat Team: The media's extermination of Bernie Sanders, and real reform.
Sanders was dismissed with the meme that he was unelectable--his populist socialism was too far from the American mainstream. The best evidence at the time was exactly opposite: In polling projections, Sanders beat Trump handily, while Clinton was projected to be an even match.
Months later, Wikileaked emails revealed that the Democratic leaders knew that Sanders was the stronger candidate, knew that he threatened their policies and tradition. "Many of the most damaging emails suggest the [DNC] was actively trying to undermine Bernie Sanders's presidential campaign." [Washington Post] Un-elected "super-delegates" gave Clinton a 300-vote lead over Sanders before the first vote was cast. It was the Democratic National Committee, with the cooperation of America's press, that quashed Sanders's candidacy in favor of Clinton--but it took some help from the voting machines to complete the job.
Evidence from the exit polls suggest that the nomination was stolen from Sanders, and that were it not for electronic vote-shifting, Sanders would have been the Democratic nominee, and probably the President Elect today. There were exit polls in 27 states, and in 23 of these, Sanders did better in the exit polls than in the officially-reported count [data by TDMS Research]. 11 of the 12 statistically significant exit poll discrepancies favored Sanders, and the 12th was the interesting case of OK. In Oklahoma, Clinton did better in the polls than in the official results. Oklahoma votes on paper ballots, some counted by hand and some by machine. OK is unique among the 50 states in that the scanning machines are programmed for each election by state employees, rather than the private vendors who do the job on contract in all other states.
If we assume the Oklahoma election was honest, it suggests an interpretation that corroborates the hypothesis of creeping, built-in shift in the polling methodology, which I described above. If you have accepted the logic of this article thus far, then the polling companies have learned to weight conservative demographics more heavily in their quest to match the reported results. In this case, the disparity in Oklahoma's primary could mean that the polling weights have a built-in bias for Clinton over Sanders, which should be added to the reported disparities in all other states.
In June, Clinton won the nation's largest primary in California by 53-46%, but there were no media exit polls. My daughter and I helped to administer a small, citizens' exit poll in the Bay Area, which showed no consistent anomalies. California has one of the most verifiable election systems in the nation, with paper ballots counted on op-scan machines kept honest by spot checks and sample hand counts. The California primary, however was tainted in a way that did not involve computerized manipulation. The Democratic party in California has an open primary system, in which people who registered with "no party preference" can vote in the Democratic (but not the Republican) primary. NPP voters are a big constituency in CA, and they favored Sanders by 40 points. On the other side, registered Democrats favored Clinton by 30 points. So reported suppression of the NPP vote damaged Sanders's chances in a way that is difficult to quantify.
Overall, exit poll evidence indicates that Sanders actually beat Clinton in the primaries. This conclusion must be qualified because there was no tight control to validate the exit polls. The best we have was the parallel exit polls, held in the same places at the same times for the Republican primary. Republican exit polls agreed consistently with the reported vote.
The General Election
It is well known that Clinton won the popular vote. In the exit polls, Clinton handily won the electoral college as well.
In the context of all that has gone before, I am inclined to believe that the election was stolen for Trump by the same Republican operatives who have been stealing elections since the computerization after HAVA. But the evidence in this case is not as overwhelming as it has been in some past elections. The best case for electronic election theft is the fact that the exit poll discrepancies favoring Clinton are larger in just those swing states which were critical for Trump's victory. This is a clearly visible, but not striking effect.