Swing State Overview (270 to win, 269 to tie)
(Swing States are defined as states closer than 5.0%. Before you ask, if a state isn't listed, it isn't that close right now.)
I noticed, with interest, Nate Silver's takedown of Real Clear Politics earlier today. I have to say, Nate makes a an extremely compelling case that RCP is internally inconsistent in which polls they include in their averages, making it more difficult to take them seriously as a polling resource.
Then again, I stopped using RCP's polling averages a long time ago. They always seemed to be using arbitrary dates for their polling averages, and the information provided at Pollster.com left them behind a while ago. RCP is useful in the way they display the most recent state polls, and I will continue to use them for that purpose.
This also seems like as good a time as any to bring up my disagreements with other forecasting methodologies, including Nate's. While I am fully aware that most, if not all, other forecasters have a far better grasp of statistics than I do, in short I still believe that virtually all other methodologies are simply too complicated for their own good. My specific problems can be found in the extended entry.
(Dark Blue (239): Obama +7.0% or more
Lean Blue (61): Obama +2.6%-+6.9%
White / Toss-up (75): Obama +2.5% to McCain +2.5%
Lean Red (5): McCain +2.6%-+6.9%
Dark Red (158): McCain +7.0% or more)
Swing State Overview (270 to win, 269 to tie)
(Swing States are defined as states closer than 5.0%. Before you ask, if a state isn't listed, it isn't that close right now. Also before you ask, yes I have included whatever new poll you can show me.)
With only 33 days until the election, and with early voting under way in a few states, Obama has taken a commanding lead in the 2008 presidential campaign. Obama is up to 353 electoral votes, and has a comfortable lead outside the margin of error in states worth 273 300 electoral votes. If Election Day were tomorrow, I would estimate McCain's chances of victory at around 5-10%. That ios pretty similar to fivethirtyeight.com, which currently gives Obama an 85.4% chance of victory.
The bad news is that Election Day isn't tomorrow, and we still have quite a bit of campaigning to do. There are three more debates, including one tomorrow where, in order to surpass expectations, all Palin has to do is show up and demonstrate that she can speak English. A lot can happen in 33 days, and it is best to stay on the attack as much as possible. Winning isn't the only goal right now--building a working majority to in order to pass legislation also matters quite a bit. Kos has a good list of ways of looking at the current situation as well.
Here is the latest swing state chart, using only polls with a majority of interviews conducted since September 22nd (except for states that didn't have at least two polls conducted in that time frame, in which case I used the two polls with the most recent mid-point dates for their interviews. Each polling firm is only allowed one survey per average):
Swing State Overview (270 to win, 269 to tie)
(Swing States are defined as states closer than 5.0%. Before you ask, if a state isn't listed, it isn't that close right now.)
Quick summary: Pennsylvania moves off the swing state chart, and Obama gains in virtually all remaining states.
In 1992, Bill Clinton scored 370 electoral votes. In 1996, he pulled in 379. Right now, Obama is only 1.5% away from scoring 364, meaning that he is now very close to Bill Clinton's margins of victory. Obama's national lead of 4.7% is also much closer to Bill Clinton's victories (which averaged +7.0%) than to Gore and Kerry's performances (which averaged -1.0%).
Obama has frequently been compared to Dukakis, Gore and Kerry during this election, but I haven't heard many comparisons to Bill Clinton. Current polling indicates, however, that Bill Clinton is exactly the Democratic nominee with whom Barack Obama should most regularly be compared.
(Dark Blue (194): Obama +7.6% or more
Lean Blue (75): Obama +2.6%-+7.5%
White / Toss-up (85): Obama +2.5% to McCain +2.5%
Lean Red (27): McCain +2.6%-+7.5%
Dark Red (158): McCain +7.6% or more)
Swing State Overview (270 to win, 269 to tie)
(Swing States are defined as states closer than 5.0%. Before you ask, if a state isn't listed, it isn't that close right now. Also before you ask, yes I have included whatever new poll you can show me.)
Polling now shows that Obama has acquired a substantial lead in states with 269 votes. The Colorado, Iowa and New Mexico firewall is holding nicely, as our the Great Lakes states of Michigan, Minnesota, Pennsylvania and Wisconsin. the problem shows up after 269, in search of the state to put Obama over the top. As the swing state chart shows, there are a wide variety of possibilities, but McCain maintains narrow leads in virtually all of them. Even in New Hampshire and Virginia, McCain leads according to the polling median, as 2 of 3 polls show him ahead in NH, and 4 of 6 polls show him ahead in VA. The two NV polls are split.
McCain has stayed competitive with Obama by concentrating all of his resources into an extremely narrow group of states. As such, it is growing difficult for me to see how McCain wins the national popular vote this year, since the DNC and the Obama campaign have sent resources to about 35 states where they will operate uncontested. It is also growing difficult for me to see where McCain punctures Obama's current 269 total. None of the "Lean Obama" states seem like realistic McCain states to me, with the possible exception of Pennsylvania. Even there, McCain hasn't led in 25 consecutive non-Zogby internet polls. So, I don't really buy it.
I'm feeling good about 269 and the popular vote, but I still don't see an obvious 270+ for Obama. Complete state by state details, plus a methodological update, can be found in the extended entry.
Update: New Colorado polls, and CNN battleground polls added. Still looks great for Obama.
(Dark Blue (133): Obama +7.6% or more
Lean Blue (96): Obama +2.6%-+7.5%
White / Toss-up (68): Obama +2.5% to McCain +2.5%
Lean Red (78): McCain +2.6%-+7.5%
Dark Red (163): McCain +7.6% or more)
Here is the swing state chart:
Swing State Overview (270 to win, 269 to tie)
(Swing States are defined as states closer than 5.0%. If a state isn't listed, it isn't that close right now.)
* = Reflects polling from before the Republican convention
Just as he falls behind in national polls, McCain has just moved into the lead in the electoral college. This odd development is due to the information lag between new national polls and new state polls. Much the same happened last week, as McCain moved clearly into the lead in national polls, but never lead by more than 270-268 in the electoral college (and sometimes didn't lead at all). McCain's lead in today's chart probably reflects where the campaign was five or six days ago, just as Obama's swing state lead last week represented the state of the campaign a few days before that.
Hopefully, at some point, there will be tracking polls in the closest states, thereby giving us up to date information on the actual state of the campaign. For now, what state by state information we have can be found in the extended entry.
(Dark Blue (133): Obama +7.6% or more
Lean Blue (84): Obama +2.6%-+7.5%
White / Toss-up (100): Obama +2.5% to McCain +2.5%
Lean Red (48): McCain +2.6%-+7.5%
Dark Red (173): McCain +7.6% or more)
For the first that I can remember, the dark red states outnumber the dark blue states--and by a substantial amount of electoral votes, too. Here is the swing state chart:
Swing State Overview (270 to win, 269 to tie)
(Swing States are defined as states closer than 5.0%)
* = Reflects polling from before the Republican convention
The state of the campaign has reversed itself entirely from over the summer. Now, McCain holds a 2-3% national lead, while Obama needs to sweep a long run of swing states in order to pull off the victory. Hopefully things will turn around, but I see no indication of Obama improvement lately. At best, the national polls are stagnating, and the state by state polls continue to swing toward McCain.
Sorry I don't have more good news, but I am no longer one of those bloggers who argues with polls instead of accepting them. Hopefully, Palin's recent slide will impact McCain, given that there is some indication she was partially responsible for his recent rise. There also has been some incremental imporvement in the tracking polls recently, so hopefully that is a sign of a positive rend.
Also, I am travelling today, so light blogging until this evening. State by state details in the extended entry.
(Dark Blue (181): Obama +7.6% or more
Lean Blue (40): Obama +2.6%-+7.5%
White / Toss-up (88): Obama +2.5% to McCain +2.5%
Lean Red (59): McCain +2.6%-+7.5%
Dark Red (170): McCain +7.6% or more)
New polls from Alabama, Colorado (2), Florida (2), Georgia (2), Maine, Michigan (2), Mississippi, North Carolina (3), Ohio (3), Pennsylvania and Wyoming are added to the averages, making for a nearly complete set of post-convention swing state polls. Here is the swing state chart:
Swing State Overview (270 to win, 269 to tie)
(Swing States are defined as states closer than 5.0%)
* = Reflects polling from before the Republican convention
Note that when the states are pushed to make a choice, Obama still leads. And the polling in this chart is almost entirely post-Republican convention. Further analysis and state by state details in the extended entry.
(Dark Blue (215): Obama +7.6% or more
Lean Blue (49): Obama +2.6%-+7.5%
White / Toss-up (91): Obama +2.5% to McCain +2.5%
Lean Red (87): McCain +2.6%-+7.5%
Dark Red (96): McCain +7.6% or more)
New polls from Florida, Kansas, Kentucky, New Jersey, Nevada, North Carolina, Pennsylvania, Virginia (2), Washington, and Wisconsin. the only category change is in Pennsylvania, which downgrades from "Solid Obama" to "Lean Obama." It should be noted that Adam B has a good piece on why recent Pennsylvania polling, which still shows Obama ahead with greater than 95% certainty, is far more pro-McCain than reality.
The campaign continues to change very little, although there has been an incremental shift toward McCain over the last six weeks. Overall, since Obama's late June peek, it seems that McCain has gained about 1-2% both nationally and across the state polls. Even though the McCain campaign is regularly lying, contradicting itself, and running a truly nasty campaign, he has largely been given a pass. The amount of cognitive dissonance in the media coverage of McCain has reached Bush first term levels. The good thing is that Obama still leads despite it. Still, The coming VP picks and conventions will be absolutely key--much more important than the debates.
(Dark Blue (236): Obama +7.6% or more
Lean Blue (28): Obama +2.6%-+7.5%
White / Toss-up (91): Obama +2.5% to McCain +2.5%
Lean Red (87): McCain +2.6%-+7.5%
Dark Red (96): McCain +7.6% or more)
New polls from Colorado, Iowa, Michigan, Missouri, Oregon, Virginia and Washington since the last front-page update. Because both of recent polling and because I have tightened the categories a bit, Missouri moves from "Toss-up" to "Lean McCain." The categories are now in the final form they will take between now and Election Day. In the coming months, future changes will include increasing the number of polls in the averages from 4 to 5 and, come mid-October, removing all pre-October data from swing state polls.
The longer the campaign goes on, the more I am drawn to the theory that says the campaign is basically static, excepting major changes in mid-March around Rev. Wright and early June when Obama secured the nod. For one thing, the state polling picture has barely changed at all, even from the days in mid-June when Obama led nationally by about 5%. Second, while the national average appears closer than mid-June, it actually isn't. Right now, pretty much the only polls impacting the national average are Gallup and Rasmussen which, for some unexplained reason, have consistently shown the campaign to be about 3.5% closer than other polls.
Overall, at either the state or national levels, I don't see any evidence that the campaign has changed in about two months. That isn't to say that it won't change in the future. Rather, I would argue that the appearance of daily change brought on by the Gallup and Rasmussen tracking polls is a mirage. A more comprehensive look at polling indicated that this is where the campaign has been ever since Obama clinched the nod.
(Dark Blue (228): Obama +8.0% or more
Lean Blue (36): Obama +3.0%-+7.9%
White / Toss-up (102): Obama +2.9% to McCain +2.9%
Lean Red (76): McCain +3.0%-+7.9%
Dark Red (96): McCain +8.0% or more)
New polls from Alabama, Massachusetts, New Jersey, New York, Oregon and Wisconsin (2) today. Oregon moves from barely "Solid Obama" to the strongest "Lean Obama" possible. The lack of swing state polls (those closer than 5.0%) means that there is no change in the swing state forecast:
Apart from swing states, I find it interesting that the number of undecideds remains over 10%, even with many polls pushing leaners. The peculiar political environment, with Republicans extremely unpopular, but Obama under-performing under white Democrats, probably leaves an opening for third-parties to collectively reach 4-5% in this campaign, ala the 2000 election. The three "major" third party candidates even all have decent names, as we have two former members of Congress (Barr and McKinney) and also Ralph Nader.
The problem I think they are facing is one of organization, and a lack of grassroots support. In 1992, Ross Perot actually had a huge number of self-starting volunteers that put him on all fifty state ballots. In 2000, Nader raised about $8,000,000, mostly online. That same year, Buchanan also was able to use the millions of dollars in public funding the Reform Party secured by reaching 5% of the vote in 1992 and 1996. There just isn't the same level of organizational strength available to this year's third-party candidates.
The 4-5% opening probably could have been filled by Ron Paul, if he had run as a Libertarian or an Independent. However, I suppose that since that would have meant giving up the Republican nomination for his congressional seat, he didn't do it. Although there is no way to tell how that would have changed the campaign, the most likely outcome is that Obama would have a pretty huge lead right now.
(Dark Blue (236): Obama +8.0% or more
Lean Blue (28): Obama +3.0%-+7.9%
White / Toss-up (102): Obama +2.9% to McCain +2.9%
Lean Red (76): McCain +3.0%-+7.9%
Dark Red (96): McCain +8.0% or more)
New polls from Alabama, Arizona, Connecticut, Florida (2), Massachusetts and Washington show little change. Even though the map doesn't show it, Arizona moves from "lean McCain" to solid McCain." (I'll fix the map later--not in a position to do so right now.) Also, even though the state is still technically a statistical toss-up, McCain has regained a narrow lead in Florida. Over the past week, Obama had a narrow, statistically insignificant lead in the state. In national polling, the campaign continues to hover between a statistical toss-up and a statistically significant lead for Obama.
If this is as bad as the McCain barrage can get, then Obama should be in a strong position. McCain's best shot still has yet to put him in a statistically insignificant lead nationally, or at a statistically significant lead in states worth 200 electoral votes. I except Obama to gain in polls over the next week, relative to the current standings. From that point, we will have VP picks and the conventions to change and / or define the campaign. This barrage by McCain appears to simply have not been enough.
STATE BLOGS
