About this Site
Welcome to Electoral Vote Predictor, which tracks political polls for U.S. federal elections. The site was immensely popular in 2004, ranking in the top 1000 Websites in the world and the top 10 blogs in the world, with about 700,000 visitors a day. In some surveys, it was the most popular election site in the country. In 2006, it tracked the Senate and House elections. In 2008, it tracked the presidential and congressional races and was getting over 1 million visitors a day at the end. In 2010 it did the Senate and House. In 2012 it tracked the presidential race and the Senate but not the House because House races are polled too infrequently and there is no historical data for the new congressional districts to fill in the many gaps.
Unlike other sites, which track generic national polls, this site tracks the state-by-state polls. After all, the presidency is decided by 51 separate state elections and the Senate by elections in almost three dozen states. As new state polls are released, the maps, spreadsheets, tables, graphs, etc. will be updated. In the maps, the states with white centers are statistical ties and are subject to rapid fluctuations.
The main currently contains a map showing the state of the polls for the Senate. Putting the mouse on a state pops up information about the polls there. Clicking on a state goes to a graph of the polls for that state. There are numerous other pages with election data on the site. From any page, clicking on the "electoral-vote.com" name on the red banner at the top of the page takes you back to the main page.
The news items on the main page are written by two front-page authors, "The Votemaster" (Andrew Tanenbaum) and "Zenger" (Christopher Bates). The (V) or (Z) after each item indicates who wrote it.
To the right of the map are two clickable buttons:
|Previous report||Link to yesterday's page|
|Next report||Link to tomorrow's page (useful when you have gone back several days and want to go forward)|
Polling methodology and using polls is not as simple as you might think. The Polling methodology file explains polling and the methodology used here in detail. The site is updated once a day, generally by 7 a.m. Eastern Time. For this reason, polls released during the course of a day will not show up until the next morning. This is the same news cycle as a (printed) morning newspaper.
Before sending e-mail, PLEASE, PLEASE read the FAQ first. Your question may be answered there. Please DO NOT send e-mail announcing a new poll until you have checked the site the next morning. Getting 1000 e-mails about a poll we aready know about is not so much fun.
We have bent over backwards to be scrupulously honest about all the numbers, carefully write the main page to be strictly nonpartisan. If you want an election site that has a pro-Republican bias from beginning to end, including all over the main page, try www.electionprojection.com.
Menu of Icons to the left of the Map
|Welcome + FAQs||This page|
|All Senate races||Descriptions of all 2018 Senate races and candidates|
|Senate polling data||Senate polling data in various formats|
|Control of the Senate||A time series showing how many seats each party has over time from Jan. 2018|
|Tipping-point state||Easiest way for each party to win the election|
|Icons for bloggers||Icons with up-to-date projections for your blog or Website|
|2020 Dem candidates||Every Sunday another potential 2020 Democratic presidential candidate will be profiled here|
|Feature stories||Longer pieces about various aspects of elections in general|
|Data galore||Lots and lots of other election data|
There are FAQs (Frequently Asked Questions) files about:
Data Reuse Policy
You are free to reuse the data, maps, graphs, tables, charts, and other information in any way you like if you credit the source and provide a link back to this site if feasible.
The primary host for this site is at HostRocket.com. They have fast servers, redundant fiber-optic connections to the Internet, and a knowledgeable technical staff. I recommend them highly, especially for business applications where reliability and customer service are important.
The nice colored maps with the pop-up boxes are produced completely automatically from each new day's spreadsheet by code written by Eric Paulson of Viking Web Development. The code uses a library written by Walter Zorn.
Thanks to Herb Weisberg of Correlation Research for statistical consulting.
I would also like to thank 'Grumpy' for producing the software that makes the time series graphs as well as much other help.
John Sinteur has provided technical assistance in many ways as well as writing an always-amusing blog The Daily Irrelevant.
Sally Montoya has been a great help with data collection. Without her I couldn't have done this.
Thank you very much.