Thursday, August 14, 2008

August 14th Update

I won't provide an update every day until after the convention, this will be more of a "when I get around to it" process. I'll try to post a new map every week -- I didn't bother in this case since Ohio is the only change.

Today's tally is Obama 298 to Mccain 240.

Same bad news for Mccain, good news for Obama. Obama has multiple paths to win -- basically if he wins the states he is heavily favored plus any of the battlegrounds he wins. He has a high-end potential of 354 EV's with all the battlegrounds. Mccain needs everything that he is leaving plus ALL of the battlegrounds, including Nevada, Ohio and even Colorado, which may well be the key to this election.

Changes today:
Ohio move from Tie to Obama. It also downgrades from "Key" to "Substantial" battleground -- Obama has reclaimed a narrow lead in a number of polls and adds 20 EV's to his column.
Nevada upgraded from substantial to key battleground -- it stays in Obama's camp barely but as posted yesterday, Obama's underperformance with Hispanics make this less of a lock than it should be for him.
Missouri moves from potential to fringe battleground. I still think this is probable Mccain country, but some projections show it even now.
North Carolina moves from potential to fringe battleground. If Obama wins here, he will win the election in a landslide.

Wednesday, August 13, 2008

First Run of Predictions

You have heard of polls, "polls of polls", well this site is aimed at projecting electoral votes in the presidential election based on a poll of the available poll-based projection sites. As each of the sites uses a slightly different formula for inclusion of national and state-based polls, each reaches slightly different conclusion. This projector takes a consensus approach and statistically looks at the current projections of prediction sites as well as state's voting histories to determine the most likely winner of each state's electoral votes. In total, the top 11 electoral vote prediction sites are included. Excluded are sites which are primarily opinion-based versus poll oriented. Also excluded are sites that typically list a large number of states as "undecided" -- anyway can draw an electoral map that gives DC to Obama and Utah to Mccain. We are interested in who is going to win the key states

Methodology:
The 11 sites are surveyed for their projection. No weight is given to whether a state is projected as "weak Obama" or "strong Obama", if it is for Obama it counts the same. This is done because different sites have differing definitions and categories for the strength of their ratings. A state is categorized on each site as either "Obama", "Mccain" or "Tie". Taken with the results of these 11 ratings are the results of the last 4 elections. Each site counts as 1 of 11 points -- a tie counts as .5 points for each candidate. If a state has gone Republican the last 4 elections, Mccain is award 4 extra points, Obama none. If a state has split the last 4 elections, they are each awarded 2. The candidate with the most total points is projected the winner

Definition of Battleground States:
Battleground states are defined as states where the 15 points are split between the candidates.
"Key" Battlegrounds are considered to be within 3 points
"Serious" Battlegrounds are considered to be within 6 points
"Substantial" Battlegrounds are considered to be within 9 points
"Somewhat" Battlegrounds are considered to be within 12 points
"Fringe" battlegrounds are any other state where the 15 points are not all awarded to 1 candidate

"Potential" Battleground States:
These are the states that, while they are not battlegrounds now, could potentially become battlegrounds based on relatively close polling. These states are not scientific, but simply my opinion based on poll-reading.

First Set of Predictions
Obama leads, which should be no surprise to those of you following the national news. He has been ahead in national polls by 4 or 5 points (with a few outliers.) The conventions could potentially shake things up one way or another, either giving Obama a more substantial lead or allowing Mccain back into contentions.

The first projection shows Obama with 278 Electoral Votes, Mccain with 240 and Ohio with its 20 electoral votes listed as an exact tie. This would not be so bad for Mccain, but looking at the battleground states, they are mostly states already showing in the Mccain column. If you give Mccain every single state listed under "Key", "Serious", "Substantial", "Somewhat" and "Fringe" battlegrounds, he gets to only 274. On the flip side, Obama would have a 328 electoral vote landslide.

The good news for Mccain is that there is still a path to win. He will likely need to shake loose some "potential" battlegrounds to make it happen though.

August 13th Prediction:
Predicted Obama States:
California -- 55
Colorado -- 9 (somewhat battleground)
Delaware -- 3
District of Columbia -- 3
Hawaii -- 4
Illinois -- 21
Iowa -- 7
Maine -- 4
Maryland -- 10
Massachussets -- 12
Michigan -- 17
Minnesota -- 10
Nevada -- 5 (substantial battleground)
New Hampshire -- 4
New Jersey -- 15
New Mexico -- 5
New York -- 31
Oregon -- 7
Pennsylvania -- 21
Rhode Island -- 4
Vermont -- 3
Washington -- 11
Wisconsin -- 10
Total EV's = 278

Predicted Mccain States
Alabama -- 9
Alasaka -- 3 (fringe battleground)
Arizona -- 10
Arkansas -- 6
Florida -- 27
Georgia -- 15
Idaho -- 4
Indiana -- 11 (somewhat battleground)
Kansas -- 6
Kentucky -- 8
Louisiana -- 9
Mississippi -- 6
Missouri -- 11
Montana -- 3 (substantial battleground)
Nebraska -- 5
North Carolina -- 15
North Dakota -- 3
Oklahoma -- 7
South Carolina -- 8
South Dakota -- 3
Tennessee -- 11
Texas -- 34
Utah -- 5
Virginia -- 13 (substantial battleground)
West Virginia -- 5
Wyoming -- 3
Total EV's = 240

Exact Predicted Ties
Ohio -- 20 (key battleground)

Battleground States & Commentary
Key
Ohio -- this always critical state is exactly split in projections. Obama's strength in urban centers offset by Mccains large lead among rural white voters. Very, very tough state to predict.

Serious
No states fit this category in this projection

Substantial
Virginia -- has voted Republican for 10 straight presidential elections, but the growth of the DC suburbs have significantly changed the profile of this state. Obama's chances would be bolstered by picking an Old Dominion running mate.
Montana -- surprise polls have shown it very close. Republican 3 out of the last 4 elections, but very high presence of independents puts it up for grabs.
Nevada -- Obama leads but his underperformance with Hispanic voters in the primary could put it at risk

Somewhat
Indiana -- primary results and voting history indicate this should be solidly Republican. Suprisingly, several polls have shown Obama even.
Colorado -- put it in Obama's column for now, but polls are still relatively close.

Fringe
Alaska -- favored Bush in 2004 by over 20 points, but the Ted Stevens scandal has severely damaged the GOP there. A recent poll showed Obama up by 5 points, but until it is validated with more data, this one remains in the GOP column for most.

"Potential" Battlegrounds
Florida -- universally projected Republican now, but polls are still relatively close. Mccain should fare well with the retired crowd, but Hispanics and Jews will be swing votes.
Iowa -- showing all Obama now, but has a demographic profile that may favor Mccain in the end.
Michigan -- Obama leads on the strength of his support in urban areas, but polls are still relatively close. Mccain will need to make his economic case forcefully to win here.
Minnesota -- Obama leads but not by a lot. Small African-American voting base here and a state with a big independent streak.
Missouri -- showed close early with Obama even leading in some polls. Mccain has resolidified his lead here, but Obama could make a comeback, particularly if he can recapture some of the magic he had in the primaries.
New Hampshire -- another state with an independent streak that currently favors Obama. Used to be a GOP stronghold but has drifted more Democratic with influence from neighboring New England states.
North Carolina -- a long-shot for Obama, but several polls recently have put him close.
Pennsylvania -- Obama solidly in the lead for now, but he was crushed in the primary and this one always seems to be close in the end.