4.5 months out

Some notes. I'm refusing to call any race a tossup. I insist on leaning one way or the other. However, I will now note the difference between "Obvious" and "Probable" on the map, with probable Republican being pink rather than red, and probable Democrat being cyan rather than blue. I will also supplement my own predictions with the Consensus, an average ranking of up to ten predictions including my own. Currently, in order to be included in the average, the site must have been updated in the past week. More on that later. First, it's time for my updates, which will be covered in slightly modified regions.

West: No changes. Nevada is still slightly leaning Republican from what I can tell, unlike many other states.
Mountain: No changes. It's been discussed that Arizona may be in play. It's not.
Central (formerly Midwest, now includes Iowa): Arkansas never should've been considered close by me. Missouri has flipped, just one of several. Arkansas (6) probable Rep from leaning Rep, Missouri (11) leaning Dem from leaning Rep
Great Lakes (formerly North, now without Iowa): Michigan has forgiven Obama. Ohio, I'm still somewhat dubious, especially given the outlandish ten-point lead some polls are giving Obama. Michigan (17) leaning Dem from leaning Rep, Ohio (20) leaning Dem from leaning Rep
Gulf Coast: No changes. Florida is too chaotic for me to turn it over at this point, but it's only barely holding on for McCain.
Mideast: The Atlantic is tending towards Obama. Virginia (13) leaning Dem from leaning Rep, North Carolina (15) leaning Rep from probable Rep
Mid-Atlantic: No changes. Pennsylvania, I'll keep as probable rather than obvious because it's too far out. I still am sure New Jersey will be an easy Obama win.
New England: New Hampshire goes blue in this update. Connecticut remains cyan, because I'm just not sure. New Hampshire (4) probable Dem from leaning Dem

I'm going to ignore congressional districts until we get closer and have better data. This has the effect of moving two of Nebraska's EVs from probable to obvious.

Current numbers by group:
Obvious Dem 165, Probable Dem 86 (cumulative 251), Leaning Dem 66. Democratic total: 317
Obvious Rep 137, Probable Rep 26 (cumulative 163), Leaning Rep 58. Republican total: 221

So now here's the Consensus. First, let me introduce the 8 predictions being used. Election Projection (EP) by Scott Elliott is currently down, so I used the most recent cache on Google (June 16). Elliott uses a formula incorporating state and national polls and some other adjustments. He got 48 states right in 2004, and every 2006 Senate race.
OpenLeft is the first of a few liberal sites' prediction being included, mostly because I started my list from that at Democratic Convention Watch. This is a weighted average of polls to come up with a potential margin of victory. Another, more established, poll-based prediction comes from Electoral-Vote.com (E-V). Worth noting is that all three of these sites define their own ranges, however I have redefined them to be consistent.
A 15% or larger margin of victory is a solid win, worth 5 points if it's a Democratic win, and 1 point if it's a Republican win. A 10.0%-14.9% win is worth 4.5 or 1.5 for a strong victory. 6.0%-9.9% is a moderate win, worth 4 or 2 points. 2.6%-5.9% is a weak win worth 3.5 or 2.5 points. Any state within 2.5 points in either direction is a tossup, credited with 3 points.
This system is also used on the ridiculously mathematical site FiveThirtyEight (538). This site combines polls, trends, a historical regression, and 10000 simulations a day to estimate what will happen on Election Day (for example, there's a 4% chance the winner of the popular vote will lose the electoral college, and a 3.8% chance a decisive state will be decided by less than .5%, meaning a recount could be needed). Each state has a probability of victory, but I'm using the predicted margin.
Front Loading (FL) is another blog with a prediction map based on polls. They do not allow ties. Their "Strong" (10%+) is my solid, their "Lean" (5-10%) is my moderate, and their "Tossup" (0-5%, still assigned to the leader) is my weak.
Two more conventional sites are CNN and Rasmussen Reports. The same system is used here as on FL, though I note CNN is very willing to call states tossups to keep up interest.MSNBC's First Read and NPR are ineligible, because their polls haven't been updated in a week.
And then there's my seat-of-the-pants estimations. The eight 1-5 scores are averaged to produce the Consensus.

All that said, here's some stats about the Consensus. CNN is the only predictor which does not give Obama enough votes to win before tossups. I give Obama the most EVs, at 317, though I have no tossups. Overall, 538 seems to be the most aggressive, implying that the Republicans should purchase ass-whupping insurance, as they have Obama projected to win by 100+ EVs.
Republicans are nearly guaranteed to take 126 votes, those with Consensus scores less than 1.75. They are favored to win 75 more votes, with scores from 1.75 to 2.49. On the other hand, Democrats have 168 solid and 83 favored. Among the 86 tossups, 42 are currently on Obama's side, and 44 on McCain's, for a grand total of 293-245 Obama. The over-the-top states are Ohio and New Mexico, both of which have scores of 3.375.

Four states worth discussing for outliers.
North Dakota has had almost no polling, and one poll back in February had Obama in the lead. Thus, those projectors based on polls give Obama an even shot, those based on opinions do not. Score 1.88, standard deviation .79
West Virginia is a state I am pessimistic about, as are OpenLeft and FL. 538, however, gives McCain merely a 1.9% win here. Score 1.81, SD .65
South Dakota is another state 538 is unreasonably bullish on, predicting a 3.6% win for McCain where EP and E-V predict 15%+ blowouts. Score 1.38, SD .58
New Mexico is the only state with a blue and a red shading (moderate or more) on my spreadsheet. CNN has McCain favored here, but polls disagree and EP gives Obama a 7-point win here. Score 3.38, SD .58

Five months out

It's over. Hillary is done. Barack is the nominee. Here are my electoral predictions from five months out.

Obvious Democratic, Probable Democratic, Leaning Democratic, Leaning Republican, Probable Republican, Obvious Republican

West: Hawaii 4, Alaska 3, California 55, Oregon 7, Washington 11, Nevada 5.
Mountain: Idaho 4, Utah 5, Arizona 10, Montana 3, Wyoming 3, Colorado 9, New Mexico 5.
Midwest: North Dakota 3, South Dakota 3, Nebraska 2 (see below), Kansas 6, Oklahoma 7, Missouri 11, Arkansas 6.
North: Minnesota 10, Iowa 7, Wisconsin 10, Illinois 21, Michigan 17, Indiana 11, Ohio 20.
Gulf: Texas 34, Louisiana 9, Mississippi 6, Alabama 9, Georgia 15, Florida 27.
Mideast: Kentucky 8, Tennessee 11, West Virginia 5, Virginia 13, North Carolina 15, South Carolina 8.
Mid-Atlantic: New York 31, Pennsylvania 21, New Jersey 15, DC 3, Maryland 10, Delaware 3.
New England: Vermont 3, New Hampshire 4, Maine 2 (see below), Massachusetts 12, Connecticut 7, Rhode Island 4.
Maine and Nebraska CDs: NE-01, NE-02, NE-03; ME-01, ME-02.


Summary by level:

Obvious Dem 165, Probable Dem 82 (cumulative 247), Leaning Dem 9. Democratic total: 256

Obvious Rep 135, Probable Rep 37 (cumulative 172), Leaning Rep 110. Republican total: 282

By my current prediction, John McCain is in position to win the election 282-256. Here's why:
Florida and Ohio. They suck, and the Democrats should not count on these states.
Missouri. For now at least, I'm less optimistic about the Democrats taking it than many people. This could easily change down the road, especially with the Dems expected to win the governor's race.
Michigan and Virginia. The Democrats can win both of these states. I just think right now they are underdogs.

Edit: Texas is Obvious Republican.

Day 5 of Tech class

No, professor, I'm not going to share my livejournal with the class.

Stop asking.

Ranting

I'm doing this for class. In the event it exists beyond the end of this month, I'll use this space to rant about things that piss me off.

Preliminary example: One of my roommates parked her car behind me last week and then left for a walk. Her keys weren't in the communal "put your keys here so people can move your damn car" place. Thus I was 15 minutes late to tutoring a student.