After every election since 1980, Republicans wound up with more seats in the United States Senate than share of the popular vote would suggest. On average, Republicans took an extra six seats. The permanent skew of the Senate toward Republicans can be seen in a graph comparing the popular vote for all seats to the number of seats Republicans held. The dark red line at the top is the number of seats Republicans held. The bright red line below is the popular vote. The shaded pink area is the skew, large, chronic, and consistent.
The advantage to the Republican Party is remarkable. Republicans hold an average six seat advantage. Republicans hold more seats than the popular vote suggests in every single term. In just four of the past forty-two years, Republicans won the popular vote: 1997 through 2000. However, for 24 years Republicans held control of the Senate (except that in 2001, a party defector handed control to the Democrats for the remainder of the term). Republicans control the Senate more than half the time even though the party wins the popular vote under ten percent of the time.
The skew appears to be getting worse. Republicans have not won the popular vote this century, but controlled the Senate six of eleven terms. Last decade the skew increased to well over eight additional seats.
CALCULATING THE DATA
Looking at all election data from 1976 through the present, including appointments and special elections, take total vote counts for candidates of each party for all one hundred Senators serving at a given point, specifically February 1 of odd years, add them together and calculate an aggregate percentage. One exception was made. Al Franken was included in the calculation even though he was seated after February 1.
In some cases where candidates from other parties gained over one percent of the vote, third party data was added to totals for a particular party where appropriate: for example conservative parties may be added to Republican totals or multiple Democratic candidates are added together. As it turned out, even in the aggregate, inclusion of this extra data did not move the graph noticeably.
In the case of California since the jungle primary system, judgment had to be used. For example, in the 2018 Senate race Democrat Kamala Harris beat Democrat Loretta Sanchez. Noticing that Harris beat Sanchez by nearly the same percentage as Hillary Clinton beat Donald Trump in 2016, it seemed appropriate to count the Sanchez vote as for Republicans because percentages matched.
Many hours went into this research. The resulting numbers were so starkly skewed there wasn’t much text needed: The United States Senate does not represent the American population.