I have found 12,301 votes cast in the 1940 General Election that have not been tabulated in any secondary source that I have looked at (including Congressional Quarterly and Clerk of the House).

In my annual drive between Eastern Massachusetts and Upstate New York, I often stop into a very deep repository of past election data – the New York State Library in Albany to collect additional past election data. This trip was a bit different, as I spent some time four floors above the library in the New York State Archives. Here, they have many original hand-written and typed records of election returns. Among the new data that I collected, was a sheet of write-in returns for the State of New York for the office of President in 1940. The document, titled “Statement of scattering vote cast as the General Election for Electors of President and Vice-President”, tabulated 12,301 write-ins for President. These votes include 11,289 for Communist Party candidate Earl Browder, 121 votes for Socialist Labor candidate John Aiken, and 891 votes for scattered write-ins. The vast majority of these write-in votes were recorded in the New York City boroughs of Bronx, Kings, and New York, where Browder received about 0.5% of the vote in these Counties as a write-in. The document only includes write-ins for 12 of New York’s 62 counties. Some of the larger counties, such as Erie, are absent, so it is likely that there are even more votes for Browder that are uncounted. These votes will soon be added to the Atlas election results database.

I was contacted this week by a reporter from the Austin American-Statesman newspaper in Austin, TX with a question about a story that they are working on. He wants to present a data that can visually represent the relevance of the Texas Primary to the Presidential nominating process. Texas holds its primary on the old “Super Tuesday”, March 4, 2008 – after 35 other states have already held their contests (20 states – give or take – now vote on a single day, February 5, 2008, a date that has been dubbed “Super Duper Tuesday”). The Republicans allocate 65% of their delegates prior to March 4 and the Democrats allocate 55% (the Democratic number is a reduced due to the penalty applied to Michigan and Florida for violating party rules by placing their nomination contests prior to February 5. These two states have been stripped of all delegates. The Republican number is reduced due to the penalty applied to Florida, Michigan, New Hampshire, South Carolina, and Wyoming for violating party rules. The delegate count for these states has been reduced by half.)

In response to this request, I have created an additional feature – on each of the polling state summary pages, there is a new section that shows the number of delegates awarded prior to the date on which the state-under-view holds its contest. In addition, a new graph shows the projected delegate count for the top candidates as allocated by the states holding primaries and caucuses earlier than the given state. This graph also shows the projected delegate standing as a function of time – as the polling data evolves with the campaigns.

If the line representing the leading candidate is above the bold win line (half-delegates + 1), then the state contest is no longer relevant, otherwise, the delegates awarded for the state contest are helping to determine the party nominee. Currently, the Texas Democratic Primary and the Texas Republican Primary are both relevant, as no candidate has exceeded the number of delegates needed to secure the nomination prior to March 4. These charts will be updated as the campaign unfolds.

Projected Delegate Allocation Prior to this Contest: