Monday, October 13, 2008
Obama Holds His Lead, The Dow Surges and I Introduce New Election Tools
Electoral Projection: Obama/Biden 349, Mccain/Palin 189
(Obama retakes Missouri)
Popular Vote Projection: Obama/Biden 52.6%, Mccain/Palin 45.4%
Days Until Election: 22
There really isn't much happening on the electoral map these days. Missouri is just so close in my projection model that it just keeps flipping back and forth. Obama still holds a commanding lead going into the last debate. As a side note, now that the polling history in all the key states is robust and the election is close, I have fully phased out the historical component of my calculation -- the projections are now entirely poll-based. As we get really close to the election, I will be adding a time-phase adjustment for the really close states to account for the age of the polls and the movement of the race since they were taken. I have also eliminated the "potential" battleground category as we are too close to the end to be talking theoreticals. I've combined somewhat & fringe battlegrounds in the new "fringe battleground" category, which is slightly expanded to incorporate all the close states.
Missouri -- flips to Obama -- this one hasn't been moving much, it keeps flipping because it is so close
Missouri -- remains a key battleground after the flip back to Obama
North Carolina -- no change
West Virginia -- promoted from serious to key -- again, this is still all on the strength of a single poll. But it WAS a poll that showed Obama up 8 points. We'd like to see another.
Ohio -- promoted from substantial to serious battleground -- Mccain is working hard here to close the gap.
Nevada -- promoted from fringe to serious -- the polls are mixed here -- the margin and turnout of Hispanics will be key.
Indiana -- promoted from potential to serious -- it is close here, and Obama is spending to try to score this improbable upset.
Florida -- no change
Virginia -- demoted from serious to substantial. Obama would appear to be pulling away slightly here.
Colorado -- promoted from somewhat to substantial. This one appears it will stay close the whole way.
Fringe Battlegrounds (formerly Somewhat & Fringe)
Georgia -- added as fringe battleground -- with huge early voting turnout by African-American voters, could Obama win here after pulling resources?
New Mexico -- promoted from potential to fringe -- still close, could Mccain steal this Arizona neighbor?
Minnesota -- promoted from potential to fringe -- Mccain's hard campaigning showing some progress
Note: New Hampshire drops off the list. This one is probably too far gone for Mccain to recover. Amazing for a state that always loved him.
As a service to my readers, I've introduced a few new tools for analyzing the election:
(1) The first is my projected spread chart. It shows, based on my modeling, the projected differential in each state. It also shows the "transfer" state, the closest state each candidate needs to win the election. As a refresher, for even remotely close states, I have utilized my non-partisan sample-size weighted methodology. For states where the margin is >10% in all available polls, I've used less rigor and used the RealClearPolitics average. The numbers are the MARGIN of Obama minus Mccain.
(2) My projected popular vote chart. I have based this on the 4 major tracking polls that have at least 3 weeks of history. These are Gallup, Rasmussen, Hotline & Battleground. Full disclosure: I have excluded Zogby due to its insufficient history. Including Zogby would have less than 0.5% impact on the days the Zogby poll has been in place, so this is more about convenience than doubting Zogby's accuracy. I have weighted the polls, similar to all my weightings, based on sample size. This gives Gallup 33% weight, Rasmussen 43%, and Hotline and Battleground 12% each (Gallup and Rasmussen use much larger samples than the others.) The charts show the MARGIN of Obama minus Mccain.
No Dent By Mccain Yet, Can the Market Help?
The tracking poll results show no sign Mccain has cut into Obama's lead yet. But given that Mccain's crash correlated with the stock market, could the big recovery today mark a turning point for him? Will people refocus from the economic issues, which clearly favor Obama and focus on questions about Obama's past? I don't know, but Mccain certainly hopes so.
Virginia in the Transfer Seat
The latest projection shows that to win, John Mccain would need the states that he currently leads (holding on to the battlegrounds of Georgia, Indiana, West Virginia and North Carolina) while picking up Missouri, Ohio, Nevada, Colorado, Florida AND Virginia. Obama needs to only hold Virginia (or any of the other states I just mentioned) and he wins. So, Virginia is the decider at this point.
Mccain and Palin are in Virginia today (they must've seen an advance copy of this blog), Mccain also visits North Carolina.
Obama is in Minnesota, Joe Biden in New Hampshire (clearly they didn't get the memo that New Hampshire is no longer in play)
Mccain is finally doing what I thought he should -- defending the Bush map. He needs to shore up North Carolina and aggressively cut into Obama's lead in Ohio, Florida and Virginia. Obama is outspending him like crazy, so his only asset is campaign stops that generate free media.
Poll Closure Times
Interestingly enough, even if it is close, it could be over relatively early on election night. That's because most of the battleground states are in the Eastern or Central time zone. I've included above the closure times with the battlegrounds highlighted. Note that I have included the time the LAST poll in state closes for states that have multiple closure times in multiple time zones.
Some Business Travel
I will be traveling on business for good parts of this week and next. I will post as often as is practical, but you may notice a drop-off in the pace. Nothing I can do about it -- I have to spend the next 10 years making back what I lost in the last two weeks in the stock market. Rest assured, you will have up-to-date information leading up to the election.