Spent the weekend looking at AP "call" times from 2016. Here are three graphs. (I hope the thread stays together.) The first simply shows when the polls closed and when the AP declared a state for a candidate. The second graph is a scatterplot that shows the time (in hours) it took to declare a state (y-axis) against the ultimate margin of victory by the winning candidate. Intuitive result overall, but still significant variability.

In the process of writing a paper about the reliability of voter registration statistics in EAVS, I've done a survey of VR/list maint. reports issued by states. Especially good ones are available from Florida, Virginia, and Washington. Florida: dos.myflorida.com/elections/data…

Two graphs to illustrate the key take-aways from the #OH12 special election. First, Balderson's % plotted against Trump's % by county in the district. (The Trump numbers are only from precincts in the 12th dist.) Balderson just a little less than Trump in each county. (Circles ~ to turnout.)

1/7 The cult of the voting machine strikes again.
politico.com/story/2018/07/… 2/7 The security problems that were unearthed during the 2016 election have very little to do with voting machines, but rather, the computer systems that support the administration of elections.

Thought-provoking article by Nicole Goodman and @leahstokes on the effect of Internet voting on turnout in local Ontario elections: doi.org/10.1017/S00071… While framed as an article about Internet voting, it might be better to think about it as an article about convenience voting in low-salience elections. It has a lot for friend, foe, and indifferents about Internet and convenience voting to think about.

Q&A session on the use of the election security funds administered by the @EACgov just finished up. Here are some points addressed. Most are probably obvious, but you never know... There's an FAQ on the use of the funds here: eac.gov/faq/default.as…