The Special Senate and 18th Congressional District Elections Give 3 Important Political Answers and 1 Observation


            1. Fraudulent votes.  Donald Trump claimed that there were millions of illegal votes cast in the 2016 presidential race. There is certainly no evidence of illegal voting in the Special Elections in Alabama and Pennsylvania.


            The turnout was down, down, down.  In Alabama, 2,123,372 people voted in 2016, 66.4% of the registered voters.  One out of every three people eligible to vote did not cast ballots in the hotly contested presidential race.  Do you think there are hordes of illegal voters out there willing to risk jail or deportation to vote in an election that 1/3 of the eligible voters shun?


            Along comes the Special Senate Election to fill the vacancy created by Jeff Sessions' appointment as Attorney General. In 2017, only 1,348,670 people voted, 45.2% of the electorate. That means that more voters stayed home than went to the polls. What happened was that 666,333 of Trump's voters did not turn out to vote for Republican Moore. But seeing as Jones got 673,896 and Moore got 651,922 - it's a sign that there is plenty of room out there for a third or independent party candidate. And this is just among the people who voted in 2016, ignoring the 1,073,075 who did not.


            2. Inaccurate vote counts.


            The 18th District Special Election proves that the electronic voting machines are accurate. Absentee ballots are paper ballots. Pennsylvania has stringent rules for casing absentees, so less than 5% of the voters cast paper ballots: 4.3% in Greene County, 4.6% in Washington County and 2.4% in Westmoreland County.


            The results in the paper ballots are mirrored the results in the machines.


Machine vote

Absentee vote


Lamb - 41.3%

Lamb - 46.2%


Saccone - 57.7%

Saccone - 52.1%


Miller - 0.9%

Miller - 0.5%


Lamb - 46%

Lamb - 47.1%


Saccone - 53.3%

Saccone - 52.1%


Miller - 0.7%

Miller - 0.76%


Lamb - 42.2%

Lamb - 47.6%


Saccone - 57.1%

Saccone - 51.3%


Miller - 0.64%

Miller - 1%

            Even though Lamb did better in the absentee ballots than he did in the machines, it was never more than 5% and never changed the outcome, even in the absentees alone. And seeing as the absentees were 4.6%, 4.3% and 2.4% of the vote respectively (Allegheny's absentee breakout won't be available for some time), 5% of 5% is 0.25%, a really small number and unlikely to change the outcome of an election, even a close one.


            Lamb is ahead by 627 votes out of 228,378 cast. That comes to 0.274% more than the outlier 0.25%. The margin in the Alabama Senate race was 1.62%.

Trump's last-minute campaign appearance boost probably explains Saccone's relatively better performance in the machines cast by late-deciding voters, but it was so small and made almost no difference. The voters are not buying what Trump is selling. The message of the Moore and Saccone losses in the Special Elections is that voters are not supporting candidates who seek to replace the Constitution with the Bible.


            To an extent, the similarity between the machine and absentee vote is a counterintuitive result. It is one I discovered by examining thousands of election districts. Absentee ballots are perfect predictors of turnout, as well as almost always mirroring the vote in the machines. Many people find it difficult to imagine that the people voting absentee are a cross-section of the voters who vote in the machines, but that seems to be the case.  In any event, it is hard to make the case that the vote count in electronic machines is compromised when the absentee ballots constitute a paper checksum and confirm the accuracy of the count.


            3. Gerrymandering. There is a lot of fighting over the drawing of election districts designed to favor one party over another. The Alabama and Pennsylvania Special Elections prove that anyone can win in any district, but they have to run a reasonable campaign. Alabama was considered so red that the Democratic Senatorial Campaign Committee hesitated to put resources into Doug Jones' race. In the 18th Pennsylvania Congressional District, Democrats had not fielded a candidate in the past two elections.


            The message to both parties is to run candidates everywhere.


            4. Another message of this election, which the media will assiduously ignore is that every vote counts. Both the Alabama Special Senate and Pennsylvania 18th Congressional District races were very close, with the winner failing to win a majority. Senator Jones won with 49.97% of the vote and Conor Lamb won with 49.84%. Both races had only two candidates on the ballot, so the close result and the failure to win a majority by the winner shows that the voters are evenly divided on the parties and that the independent voters are determining the outcome.


            The similarities of the result, whether in Alabama or Pennsylvania, shows that people are the same all over. The low turnout and close result show that there is room for successful independent candidacies. These elections show that voters are disgusted with the performance or in this case non-performance, of the two-party system.


Return to Institute of Election Analysis Home Page

Contact: Joshua Leinsdorf