Prediction scripts are now available for the 2008 Senate Election and the 2008 Gubernatorial contests (joining the already-available 2008 Presidential Predictions. The 2006 Senate aggregate prediction called 32 of 33 contests correctly - only missing Virginia. The 2006 Gubernatorial aggregate prediction called 35 of 36 contests correctly - only missing in Minnesota. Give a try at election prognostication!

Yesterday, the Atlas operating system software and core services were upgraded. This included the core OS, Apache, MySQL, and PHP. Please submit a bug if you happen to find that any of the site features were broken by this upgrade. Many thanks and enjoy!

With all the clamoring amongst the states to move primary contests to the front of the line, the Atlas has decided to leapfrog them all, unveiling today the 2008 Presidential Primary Mock Election contests: the 2008 Republican Mock Election and 2008 Democratic Mock Election. The Atlas Mock Elections in each state are set up as open primaries (meaning that any registered voter of any party affiliation may cast a ballot in either the Democratic or Republican primary). In the real primary and caucus system, each state sets the rules for participation:

• Open primaries allow voters of any party affiliation to vote
• Modified open primaries allow independent or unaffiliated voters as well as those affiliated with the party holding the election contest
• Closed primaries limit the participants to those registered with the party holding the election contest

The allocation of delegates for the Atlas Mock Primaries are modeled after the rules for each state - typically awarding delegates proportional to the popular vote in a given state or via winner-take-all (the same system as 48 states and the District of Columbia use in the general election to award Presidential Electors). In the real primary system, a certain set of delegates are allocated via the vote in congressional districts - this level of detail has been omitted on the Atlas mock election - where all the district delegates are awarded as part of the "at large" group. Overall, the primary process is quite complex.

Go now and cast your ballot!

The Atlas Server was replaced today with a new motherboard, faster processor, and more memory. Enjoy the improved performance!

Election data and maps from the 2006 General Election for U.S. Senate and Governor for most states have been uploaded to the site. A few states remain to release their official data, including my home state of Massachusetts. The expectation is for these data to be available within the next couple of weeks. The detailed county and town-level spreadsheets have also been updated with the most recently-available information.

Preliminary, unofficial election results (data and maps) for the 2006 Senate and Governor's races are on-line. These data will be updated as election agencies across the country finalize and post official tallies. A final official compilation of data for all states is expected in early January. Population data for 2006 will be added when these data are released by the Census Bureau. Preliminary detailed spreadsheet files for these contests are available at the store. As always, once purchased, updates to spreadsheets are free.

Today is election day. If you are reading this, you are most likely an election enthusiast and will likely vote - but in the off chance that you are considering not voting, here is my encouragement to do so. Exercise your right to choose the people who govern us. Don't like the candidates running? write someone in. Turned off by the negativity and lack of civil debate? No one will ever know if you stay home. Most states have several choices on the ballot - Politics1 highlights many of the candidates running for state and federal offices.

I cast my ballot this morning - a fill-in-the-oval paper ballot to be read by an optical scanner. This is a fast, relatively accurate system with full independent audit capability. Here in the Commonwealth of Massachusetts, there are unfortunately not a lot of contested races - my Representative in Congress in the 4th district, Barney Frank (D), is unopposed. The Senate race (Edward Kennedy (D)) has not been competitive, Secretary of State, Treasurer, and Auditor have no Republican challenger (they do however, have either Green-Rainbow or Working Families party challengers). The Governor's race has been, by far, the most interesting and productive public debate (although the campaign was too negative in my opinion). There are three interesting ballot initiatives, however, and I cast my support for Question 2, allowing candidates to run on multiple ballot lines, more commonly referred to as fusion.

The Atlas will be posting updates throughout the evening for Senatorial and Gubernatorial races across the nation.

Several members of the Atlas community have suggested providing a feature for members to share their endorsements of candidates for the 2006 Senatorial and Gubernatorial contests. This feature has now been added to the Atlas (Links: Gubernatorial Endorsements and Senatorial Endorsements). A compiled map on the endorsements home page shows the aggregate of community endorsements. The new scripts include pages to show state-by-state listing of endorsement totals for each candidate.

The unofficial results are in from the competitive 2006 Republican Primary for Seante in Rhode Island. In this race, incumbent Senator Lincoln Chafee was challenged by Cranston Mayor Steve Laffey. The unofficial result published by the Rhode Island Board of Elections is: Chafee (i) 34,407 (54.22%); Laffey 29,052 (45.78%), a margin of about 8 percentage points.

The township map reveals the strength of each candidate. Laffey performed best in the North (Providence County) outside the city. His best municipality is the town of Johnston where he won 62.8% of the vote. In his home-town city of Cranston, he received his largest vote margin of 447 votes, capturing 52.5% of the vote. Chafee had his strength in the city of Providence and the south of the state (Bristol, Newport and Washington Counties). His best municipaility is the town of New Shoreham (Block Island) where he received 78.3% of the vote. Chafee received his largest vote margin in the City of Warwick - 2,134 votes where he won 62.3% to 37.6%.

An interesting statistic with regard to the lack of Republican Strength in Rhode Island - in the City of Providence (2000 Voting-Age Population of 128,341 - almost twice as many people as in Warwick, the second largest city): the total number of votes cast in the Republican primary was 2,941 votes, 1/5th the 14,861 votes cast in a far-less-competitive Democratic Primary.

The section of the Atlas that includes primary returns for US President have been updated to include official results from Presidential Primaries in the 2004 Democratic contest (county maps, congressional district maps, and data). All caucus events and several primary events were run by the Democratic party and therefore these results are obtained from the respective state Democratic Party organizations (and not from official state election agencies). In some cases, these data are incomplete due to the inability to obtain final and complete data from the state parties. A complete spreadsheet of the data is available for purchase from the store page.

The preliminary results of the 2006 Connecticut Senatorial Primary between incumbent Democrat Joseph Lieberman and challenger Ned Lamont are in. With 99.6% of the precincts having reported, Lamont defeats Lieberman 146,065 to 136,150 (51.8% to 48.2%). This is a more narrow margin than the more recent polls by Rasmussen and Quinnipiac University had predicted.
The preliminary town results map (shown at right) shows the relative strenghts of each of the two Democratic candidates. Much of the state was relatively evenly split. Lieberman was strongest in the route 8 corridor - West Haven, Derby, Ansonia up through Naugatuck, Waterbury and Thomaston. Lamont's strength was centered in the northwestern portion of the state (Litchfield County - with his strongest town Cornwall), the south near Lyme, the northeast near Mansfield, and Lamon't hometown, Greenwich.

The site has recently added state-level (and some county-level) data for the Presidential Elections of 1876, 1880, 1884, and 1888 to the database. The total figures for some states have changed slightly from the original html files due to differing figures between the primary sources referenced and the source used to create the original html tables (see What was the Real Popular Vote? entry for some additional information with regard to why popular vote figures for older elections are so variable.). Work continues to add data from older elections (state-level data, county-level data, county-boundary accuracy and image-mapping, etc.).

Two new sections are now available on the Atlas - Gubernatorial Election Polls for 2006 and Senatorial Election Polls for 2006. The new polling sections include the ability to view individual polls in detail as well as summary pages comparing recent polls by state and by polling agency. The main page includes a compilation of the most recent three polls for each state and presents a map of the result showing the current status of who is ahead on a per-state basis as well as a table showing the current view of the pick-ups for each part. Also included is a trend graph illustrating the number of states ahead for each party starting in February of 2006.

The 2004 Presidential Predictions turned out to be a very popular Atlas site feature - with almost 2,000 unique entries (and over 12,000 total version entries). The compiled map showed the correct winner of every state - with the exception of Ohio - which missed being called by merely two Republican predictions (1,000 Democratic calls vs. 998 Republican).

For 2006, new prediction scripts and databases for the Gubernatorial and United States Senate races have been created. The predictions have been expanded to include independent candidates, show party "pick-ups", and has more integration with the Atlas Forum.

Participate and Enjoy!

Last week, the Atlas underwent an expansion, adding Gubernatorial and Senatorial election results and integrating these data with the Presidential returns. There was already a special section for Gubernatorial results for a number of years - however the feature set for these data was limited. The site expansion makes available the complete Gubernatorial results from 1998 through 2005, including all the features already available for the Presidential returns (map comparisons, party maps and data, image-mapped county maps with county and township summary pages, customizable data tables, etc.). In addition, the first complete year of Senatorial results - 2004 - is also included (many thanks to True Democrat for contributing the data). Using the frames version for navagation, a new menu appears in the top frame at the right - allowing the user to select which election to view. In the no-frames version, new navigation appears at the bottom to select a different office.

Further integration of the data is planned - such as side-by-side map comparisons between offices for a given year, possible mouse-over maps, data comparisons, etc. Please send suggestions. If you would like to contribute state-wide election data (perhaps even other offices), please send an email. The newly-included election results may be access in the frames version or the no-frames version.

Total Ballots Cast are figures published by most election agencies that show the total number of people having cast ballots. Typically, a number of these ballots for any given contest (e.g. President) are invalid - either the voter decided not to choose a candidate for President (called a "blank" or "under-vote") or the voter selected multiple candidates (called a "void" or "over-vote").

Figures for Total Ballots Cast in the 2004 and 2000 general elections for President have been added to the database at the state, county, township, and precinct levels for all states in which this data is available. These data may be accessed by using the customize form at the bottom of any detailed data table (Atlas Membership required for county, township, and precinct-level). In addition, there are options to display the total number of invalid ballots and percent invalid for each in the table. For an example, take a look at the 2004 state-level data table with the new data included. Data for previous years are also in process of being added to the database.

The Politcal Matrix is a mechanism to map and compare one's political views in a two-dimensional space. Unlike the simple left/right (liberal/conservative) schema, the Political Matrix uses two axes, the right/left economic axis and the authoritarian/libertarian social axis. The two-dimensional map then places your result in one of four quadrants: authoritarian-right, authoritarian-left, libertarian-right, and libertarian-left (moderates would be points that fall close to the origin). Modeled after the Political Compass, and similar to the Libertarian diamond, the Atlas community has created and tested the new matrix to be a more detailed and accurate representation of one's political philosophy. The Atlas community political matrix chart may be found here. Go ahead and take the test, add the results to your myatlas page and compare your philosophy with the community.

Happy President's Day! In honor of the past-serving presidents, there was a proposal put forth in 2004 by Rep. Mike Castle (R-DE) and Rep. Carolyn Maloney (D-NY) that would place images of the Presidents on the obverse side of the $1 coins with the image of the Statue of Liberty on the reverse side. There would be four per year in order of service (not including those Presidents who are still alive).

The bill (H.R. 902) passed the House on April 27, 2005, passed the Senate in November of 2005, and was signed into law by President Bush on December 15, 2005. The US Mint will present the new 2007 $1 coin design candidates at a public meeting on February 28, 2006. Expect the new George Washington coin early next year. A new coin will enter circulation every three months depicting each successive President (and yes, Grover Cleveland will have two!) - (see the chronological list of Presidents).

The New York project (first discussed in this weblog entry) has made a lot of progress. The township maps and data are now complete for all counties for the general elections of 1992, 1996, 2000, and 2004. The precinct data is now complete for all counties outside of the five New York City Boroughs for those four elections (New York City Borough precinct data is complete for 2000 and 2004). This was no small task as most counties required hand-entry of precinct data. In addition, the further complication of New York's fusion system added an average of four more ballot lines per election. Overall this required a lot of number typing. For example, the counties of Erie, Monroe, and Nassau had a total of 2,897 precincts in 1992 (only to grow larger by 2004). With the exception of Monroe in 2004, all three of these counties required manual entry of the data between 1992 and 2004 from the canvass books (they are legal size documents about 3 cm thick). Total precincts for New York in 1992 are 9,764 (outside of NYC).

The New York township map for 1992 is partiularly interesting. Perot was very strong in Western New York, winning numerous towns in Chautauqua, Cattaraugus, Erie, Wyoming, and Niagara counties (Perot finished second in Wyoming and Cattaraugus counties and a very close third in Niagara County). Perot also had a strong showing in central New York in an area up the I-81 corridor from Binghamton through Northern Jefferson County, having won a number of towns in the region. Perot placed well-enough in this region such that in most towns, the winning candidate received less than 40% of the vote. Bush was strong through the the Adirondacks, perifieral Catskills, and the Genesee River Vally counties (Steuben , Allegany, Livingston, Ontario, Yates, Genesee, Orleans, Wayne). Clinton had strong support in New York City and the large upstate cities (Buffalo, Rochester, Syracuse, and Albany), in the northern tier (St. Lawrence and Franklin Counties), Ithaca, and the central Catskills (Ulster and Sullivan Counties).

County-level township maps for all counties in these four years are now available on-line to members. Township data and precinct data is also available on-line to members. There are also detailed spreadsheets with boundary codes that include both unfused and fused data for the Presidential vote at the precinct, township, and county level. See the store page if for purchasing information on the spreadsheets.

The latest book to incorporate material from the Atlas is Anthony Fairfax's The Democratic Trend Phenomena. The paperback book is an analytical work with a decent amount of statistical mathematics with the goal of proving a predictable trend in the popular vote for the Democratic Presidential candidate.

From the Author: "The goal of forecasting the outcome of U.S. presidential elections has been around for decades. However, the belief is that predicting, especially the popular vote, with a high degree of accuracy is difficult due to different national or even local conditions that change with each election (e.g. economic, social, political, and global changes)."

"Nonetheless, over the last three decades there exists a little known exception to the rule pertaining to forecasting presidential elections. The exception is that the popular vote for the Democratic candidate for president has trended in a predictable pattern since 1980. If the election of 1976 is disregarded the trend is revealed to actually begin in 1972."

"This unique predictability has been deemed by the author as, The Democratic Trend Phenomena. This book describes the cause of the phenomena, measures the accuracy, and outlines the future effects."

On the acknowledgements on page ix: "Additional thanks to David Leip who provided critical election data during the development of this book"

The book sells on Amazon for $10.17.