By Phil Harrison
A dedicated group of election integrity activists have been studying US election data for over a dozen years and have observed many statistical anomalies that shake their confidence in the integrity of our elections, which are counted in secret by computers using proprietary software that the public cannot review.
Some anomalies have occurred with the MA Democratic Presidential primary conducted on March 1st of 2016…
Richard Charnin, one of the activists who has written extensively on this issue, has presented data on his blog that he claims shows that the primary vote was manipulated to favor Hillary Clinton.
First, the raw exit polls differed significantly from the recorded vote, showing Sanders ahead 52.3% to 45.7%. The pollsters then “adjust” the exit polls to force agreement with the final tally before they report the results, which is a very unscientific practice, and is not needed in European countries with hand-counted ballots that show close agreement with exit polls.
A second anomaly shows up in a Cumulative Vote Share (CVS) graph which shows the percentage of each candidate’s vote as the precincts are added in, starting with the smallest precincts and then including the rest in order of precinct vote size. According to statistical theory, these graphs should not show a systematic slope in the curves as the votes are accumulated. The systematic shift toward Clinton shown in Charnin’s graph is not expected according to statistical theory and is evidence (but not proof) of vote manipulation:
All of these anomalies would be moot if elections were subject to audits along with a rigorous chain of custody for paper ballots. In MA, paper ballots are mostly counted by optical scanners and a hand recount can be requested by a candidate, but none has ever done so. Until the public demands and gets verifiable, audited elections, it will not be possible to know if we really have democracy.
You can read Charnin’s full blog entry at: