The Real Comeback Kid: Ralph Nader?

In what continues to be a bad polling week in national trial heat matchups for both Democrats, Zogby has released a poll showing John McCain 6 points ahead of Hillary Clinton, and 5 points ahead of Barack Obama. What's a little bit more interesting about these matchups, however, is where Ralph Nader polls. Nader pulls 6 points in the McCain-Clinton matchup and 5 points in the McCain-Obama matchup; in other words, exactly the amount by which Clinton and Obama respectively trail McCain.

Now, a couple of caveats here:

1. Third-party candidates almost always poll better than they ultimately do in the ballot booth, especially early in an election cycle.

2. Zogby mentioned Nader by name in its telephone interviews, whereas most other pollsters have not. There is not necessarily anything wrong with mentioning Nader specifically, but sometimes the mere power of suggestion is enough to get a candidate a couple of percentage points. If instead of using Nader, you asked instead: "If the election were held today between Barack Obama, John McCain, and former Minnesota Vikings quarterback Fran Tarkenton, who would you vote for?" -- Tarkenton would probably get a point or two.

3. It's Zogby

Still, Nader's numbers are presumably being driven upward by the current Democratic pie fight. I don't particuarly believe that Nader is likely to poll at 5 or 6 percent when it's all said and done, but it's plausible that some of those voters will not turn out -- or will require time and energy to bring back into the fold that might otherwise be spent on more traditional 'swing' voters.

There's More...

Contract Post

Beware of (Temporarily) Falling Polls

I am generally not blogging about national polls at this site, but today's Rasmussen tracking poll shows a marked decline in Barack Obama's support after a bad day in Friday interviews. It would seem likely that this is related in some measure to the controversy over Reverend Jeremiah Wright.

This downtrend could be a statistical fluke -- but more likely it is real. However, the important question is not if the downtrend is real (i.e. statistically significant), but how long it will last. At the extremes, the goalposts are that it will be over and done with within 24 hours -- now that Obama is starting to press his defense -- and that it will last throughout the balance of the election cycle.

At fivethirtyeight, we tend to take the long view. We don't look at polls just from the last week. We don't throw a previous poll from an agency out just because it has published a newer one. We recognize that there is a lot of uncertainty related to all polling data taken this far from an election.

When you see a poll that is '50' today, and was '40' yesterday a lot of people will be inclined to extrapolate the trendline, and assume that the poll will be '60' tomorrow. Over the long run, however, the poll is more likely to go back down to '40' than to continue to move upward to '60'. The very definition of the term 'bounce' (and there are negative bounces as well as positive ones) implies that it is something temporary -- you go up, and then you go back down.

We see these bounces almost every year around the conventions. We have seen them at various times in the primary campaign. Barack Obama and Hillary Clinton have been pretty much locked in deadstep since Super Tuesday in primary tracking polls. Whenever one has started to pull away, based on the latest events in the news cycle, they have shortly thereafter come back down.

When I studied 2004 polls to come up with the time-based weightings that I use at this site, I found that it was helpful to account for older polls, even when you have newer ones. There were several points in time, in fact, at which the newest polls were less accurate predictors of election outcomes than somewhat older ones.

I recognize my discussion here is a little obtuse -- at some point, we will get more quantitative about how to assess polls from different points in time. But the point is, don't assume that any "bounce" you see in the numbers is a "trend". More often than not, bounces are fleeting and temporary things. Trends are important too, but they operate at a much slower, glacial kind of level, particularly in general election campaigns.

UPDATE: Gallup shows a more modest 3-point in its primary election tracker, and no change in its general election tracker.

BTW, one thing to keep in mind is that you always have multiple events cycling into and out of the news cycle at any given time. One benefit of the Wright story to Clinton, for instance, is that it took the Ferraro story out of the news -- which itself may have been producing a temporary downward bounce in her numbers.

There's More...

Contract Post

Today's Polls, 3/15

Polls continue to come in at the lower end of the spectrum for both Democrats. In today's batch, that is somewhat more true for Clinton than for Obama.

Rasmussen has new polls out in California, Ohio, and Connecticut. In Ohio, McCain leads both Democrats by 6 points, a result that either balances out Survey USA's recent poll that had both Democrats ahead by 10 in that state, or shows in which direction this race is presently moving. In California and Connecticut, meanwhile, Obama has relatively safe leads over McCain, but McCain is within single-digits of Clinton:

State        Obama     Clinton
California   +15       +7
Connecticut  +12       +3
Ohio         -6        -6

We also tracked down an Ayers poll in Tennessee, which shows McCain leading Clinton by 8, and Obama by 17. Tennessee is a state in which we see Clinton potentially being competitive, but not Obama.

Overall, Obama's Win % is now at 60.0%, and Clinton's is at 40.3%. A week ago, they were at 62.4% and 42.8%, respectively.

There's More...

Contract Post

New Maps

I've replaced the old, highly-stylized but "electorally correct" maps...



With new, more traditional, "geographically correct" maps...



I'm hoping that most of you will see this as an improvement, but I'm happy to hear any objections. We've all become so used to looking at these red-blue maps that I just think, on balance, these convey the information a little better, even if they lead to the false impression that Montana is more important than Massachusetts. If I saw an "electorally correct" map I reaaaaaly liked, I might use that as a compromise, but the old ones were just distorting things a little bit too much, such that it was hard to tell which state was which.

Postscript/Full Disclosure That Like Four of You Will Care About: This is actually at heart a technological issue. I need some way to generate maps automatically when I run a site update, and MS-EXCEL has no function to do this. (Actually, it used to have this functionality, but Bill Gates decided to remove it and sell it as an add-on so that he could buy a new yacht). So the maps you see are generated using conditional formatting, basically by creating a huge grid and assigning each cell in the grid to its corresponding state:





At first I'd borrowed such a grid from Wikipedia, which had the virtue of being "electorally correct" but produced some very blocky-looking maps, but I realized (duh!) that I could improve the resolution by increasing the number of squares in the grid, and then laying a generic state outline map over it to sort of cover my mistakes. I'm like the MacGyver of Microsoft EXCEL.

There's More...

Contract Post

Swing State Analysis

You should now an additional graph along the right-hand side of the page, which I have dubbed the 'Swing State Analysis'. What this tells you is which states are likely to 'swing' the election. As I describe it here:

I'm now reporting another new parameter in my output, which is the state that "swings" the election in each of the simulation runs. The way that this works is as follows: I arrange the states from best to worst in order of Obama's (or Clinton's) vote share in each of the 5,000 simulations. I then count electoral votes upward until he equals or exceeds 269 EV. The state that puts him over the top is literally the swing state for that simulation run.

This is one of those things sounds fancier than it really is, and produces what should be fairly intuitive results. Obviously, the key variables in this calculation are the number of electors in a state, and just how tight the state is expected to be. These figures might be thought of as a good proxy for how the campaigns should allocate their resources, among other things.

What's interesting is that we come up with relatively different lists in the Obama-McCain and Clinton-McCain scenarios. Pennsylvania, Ohio, and Michigan are important to both candidates to a lesser extent so is Missouri. But New Jersey and Virginia are featured prominently on Obama's list, while they don't make Clinton's Top 15 (New Jersey because she won't lose it; Virginia because she won't win it). Florida, meanwhile, ranks much lower on Obama's list than on Clinton's, because the model thinks that he has a lot of alternative paths to 269 electoral votes that should provide a better return on his investment.

There's More...

Contract Post

Does Obama benefit from an active primary campaign?

I decided to look at something relatively basic tonight that I should have looked at a long time ago. Do the Democratic candidates fare better in the polls in the states in which there has been an active primary campaign?

It turns out that the answer appears to be yes for one of the candidates.

Consider the following:

In Ohio, where there has been a primary campaign, Barack Obama leads McCain by 2.6 points in our weighted average of polls. In Pennsylvania, where there hasn't yet been a campaign, Obama trails McCain by 1.3 points.

In Washington, where there has been a primary campaign, Obama leads McCain by 9.7 points. In Oregon, where there hasn't been a campaign, Obama leads McCain by 6.3 points.

In North Dakota, where there has been a campaign, Obama led McCain by 4 points in the only poll. In South Dakota, where there hasn't yet been a campaign, he trails McCain by an average of 6.6 points.

In Tennessee, where there has been a campaign, Obama trails McCain by 15.4 points. In Kentucky, where there hasn't been a campaign, he trails McCain by 23.3 points.

These numbers -- apples-to-apples comparisons of apparently similar states -- seem to lead to the conclusion that Obama might benefit from having had the opportunity to campaign in a state, perhaps on the order of about 5 points in his general election polls. And if we plop a 'campaign' dummy variable into our regression analysis -- which is set to '1' in every state that has voted in a primary or caucus so far, and '0' in the ten states (including Florida and Michigan) that haven't, it turns out to be highly statistically significant:

Variable    Coef.   St. Err. t-score P>|t|
Campaign    5.70    2.11     2.70    0.010
Kerry       0.43    0.65     6.65    0.000
Baptist    -0.53    0.12    -4.36    0.000
AfAm        0.21    0.14     1.49    0.144
$_Obama     8.29    2.19     3.78    0.000
$_Clinton  -6.38    1.96    -3.25    0.002
$_McCain   -6.77    3.50    -1.94    0.059

Constant    2.68    2.31     1.16    0.252

OK -- so I know that you didn't come here to read regression output. But what that is saying is that Obama is polling about 5.7 points better in states that have participated in the primary process so far, all else being equal. And it's saying that this finding is highly unlikely (around 1000-to-1 against) to be the result of chance alone. This is a robust finding too. If you remove other variables that might be related to the presence of a campaign -- the fundraising numbers, for instance -- the campaign variable continues to show up at about the same level of significance.

This pattern does not show up for Clinton. Her campaign variable is not statistically significant.

But this has been one of the big themes with Barack Obama this year: the more voters get a chance to know him, the more they seem to like him. That's how he improved his standing in essentially every state that has voted so far as time wore on; check out the pollster.com graphs for more detail.

Now, there are two ways to interpret this data. Number one is that the states where Obama campaigned are still basking in his afterglow, and will eventually come back down to earth. That's the pessimistic interpretation. The optimistic interpretation is that these improvements in his poll standing are permanent, and that he can therefore expect to gain in general election polls in states like Pennsylvania, Indiana and North Carolina as the campaign wears on. If the optimistic interpretation is true, that means that both Florida and North Carolina should be highly competitive in the fall, and maybe Indiana too, and that that Obama can expect to move back ahead of McCain in the Pennsylvania polling averages.

There's More...

Contract Post

Carolinas Down, 48 to Go

Although the Clinton campaign has been dismissing the importance of primaries in any states that have fewer than 1,000, or more than 100,000 black people
contain more Starbucks than Dunkin' Donuts
had the audacity to vote for Obama, this may be the first time they've written off a state in the general election. Actually, four states, according to Harold Ickes in today's New York Times:

They're great states, but Idaho, Nebraska and the Carolinas are not going to be in the Democratic column in November."

There's a fair amount of polling in North Carolina, and all of it points to a potentially competitive race in November. Clinton is presently credited with a 16.6% chance to win the state, which is not fantastic but marginally better than her chances in Colorado (15.1%) and Virginia (12.5%). Her percentage is about the same in South Carolina (15.8%), although with less polling data to support it. Obama's numbers are comparable -- actually a smidgen better -- in both states.

The South Coast is one of two frontiers -- the other being the Interior West -- in which the Democrats have a chance to expand their base. Virginia has already turned at least purple, and there are other, more subtle factors at play in the other states in the region: North Carolina's high-tech sector, the booming and sophisticated black metropolis in Atlanta, Georgia. These states are beginning to differentiate themselves from what might be called the Deep South or Interior South, away from the Atlantic Coast.

There's a long time between now and Pennsylvania, and one thing I'd like to do between now and then is explore the demographics of the South Coast region; my hunch is that you had a medium-term sort of revival of Southern exceptionalism, which took a decidedly culturally conservative bent, which is beginning to be outweighed by a longer-term trend toward greater liberalization up and down the entire Atlantic Coast, as Boomers seek to relocate to warmer climes and companies seek new pools of resources.

Certainly, both Clinton and Obama have the capacity to appeal to certain groups in the South Coast in ways that John Kerry could not or would not. But Clinton seems to insist on starting with a map that already has at least four states colored in red.

(hat tip to the Public Policy Polling blog)

There's More...

Contract Post

New Pennsylvania, Michigan polls are good news for McCain

A couple of new polls from Pennsylvania: Rasmussen shows McCain leading Obama by 1 and Clinton by 2, whereas Susquehana shows a bigger split between the two Dems: McCain leads Obama by 4, but trails Clinton by 3.

Rasmussen also has a new Michigan poll out, with both Democrats trailing McCain by 3 points.

These polls, when combined with the new pollster weightings that we rolled out yesterday, reduce Obama's overall win percentage from 61.7% to 59.5%. Clinton's number is essentially unchanged, at 41.4% compared to 41.8%.

Should the Democrats be worried about Pennsylvania and Michigan? I'll give my generic answer to this question: you'd always rather the polls tipped in your direction than the other way around. Both Democrats should have the opportunity to move those numbers upward -- or downward -- based on the tenor of the campaign over the next six weeks.

NOTE: The Susquehanna poll also contains some interesting splits in the Democratic primary Obama leads Clinton by 19 points in Philadelphia, and has a slight lead on the Philly suburbs, but trails her 59-10 in the Southwestern/Allegheny region. What was that James Carville quote about Pennsylvania again?

EDIT: Also a new Arkansas poll added after the buzzer (McCain +16 over Obama, but -15 versus Clinton) that affects the numbers very slightly.

And yet another new Pennsylvania poll, this one from Strategic Vision. The good news for Obama is that it shows him doing better than Clinton in that state. The bad news for him is that both Democrats are trailing McCain in November trial heats -- Obama by 3 points to Clinton's 6.

Although earlier polling data had been better in Pennsylvania, both Democrats now rate as (very slight) underdogs to McCain in that state.

There's More...

Contract Post

Pollster Ratings v2.0

I recently subscribed to PollingReport.com, and had some time today to backfit a whole bunch of polling data from previous election cycles into my pollster report card. I now have a database of over 150 competitive contests since 2000. This includes essentially every competitive presidential race, and most competitive primary races, Senate races, and Governor races. I also expanded the playing field literally to include a number of new pollsters, including both some national pollsters and some regional ones, up to a total of 32 pollsters in all. A pollster was included if it had at least 5 surveys in my database.

My methodology remains essentially the same as in this Daily Kos diary. My process is to look at the average miss for each pollster across each contest they polled, and compare it to the average miss of other pollsters in those same contest, after going through a more-complicated-than-it-needs-to-be iterative process.

The results are below, split into groups for 'regional' and 'national' pollsters. (This distinction is arbitrary -- some pollsters like Insider Advantage and Quinnipiac straddle the line between being regional and national -- but helpful for presentation). 'Error' represents the average error for the particular pollster, as compared to the 'IAE', which is the iterated average error for other pollsters in those same contests.



The way to read these numbers is that Survey USA, for example, has been an average of 1.34 more accurate than an average pollster; this is an outstanding number, and remains something of an outlier. We've also included a couple of new major media pollsters: ABC/Washington Post, which is fairly decent, and CBS/NYT, which is pretty awful. Oh, and the Zogby Interactive polls, which are off-the-charts bad -- considerably worse than Zogby's telephone polls, which are themselves quite poor. Insider Advantage has moved up a little bit; I don't like certain elements of their methodology, but their results have been fairly decent, and they're smart enough to stick to polling in the South, which is a region that they know well. Gallup continues to subsist on name recognition only. Their national polls may be fine, but they aren't doing anyone any favors when they deign to poll state-by-state matchups.

Now, for the regional pollsters:



The sample sizes here are obviously much smaller. But Sezler and the California Field Poll live up to their stellar reputations. The Ohio Poll (University of Cincinnati) and the University of New Hampshire polls have also done quite well, as have the Chicago Tribune polls, through the Market Shares polling firm. On the other end of the spectrum is the Columbus Dispatch, which conducts its surveys by mail, and pays the price in terms of its results. There are arguments for excluding Columbus Dispatch and Zogby Interactive polls entirely from the analysis.

The last step is to translate these numbers into weights, which involves some algebraic transformations that translate the pollster's average margin of error into an effective sample size, while also regressing to the mean for pollsters with a small number of surveys. These weights are calibrated such that the average poll (not necessarily the average pollster, since the better pollsters tend to be more prolific and poll more contests) will have a weight of 1.00.



Survey USA rates way ahead here, because it has not only gotten good results, but done so over a large number of contests. A Survey USA poll will be weighted around 3x as much as a Zogby or an ARG poll in our weighted polling averages.

NOTE: The recalibrated pollster ratings are not yet incorporated in our simulations and weighted averages, but they will be once we get a couple of new polls in and do a fresh run of the data.

There's More...

Contract Post

The Clinton voters who won't vote for Obama

From the recent Rasmussen Reports primary poll in Mississippi:

One measure of a deepening divide in the party is that just 56% of Obama voters have a favorable opinion of Clinton. Just 34% of Clinton voters have a favorable opinion of Obama.

If Obama is nominated, just 47% of Clinton voters say they are even somewhat likely to vote for Obama in the general election against John McCain.

If Clinton is nominated, 65% of Obama voters say they are at least somewhat likely to vote for her against McCain.

So among Clinton supporters in Mississippi, only about half are "even somewhat likely" to vote for Obama in November.

Sure, the Democratic primary has been acrimonious of late. But there's something more going on here. Who are these folks that would vote for Clinton in the primary -- but McCain in the general election? They are likely white Southern Baptists (and probably leftover Solid South Democrats), a group which, as I've been emphasizing all year, is not kindly disposed toward Barack Obama. I suspect the reason ultimately might have to do with Obama's race, but that's a subject for another day.

Now, how much of a problem is this for Obama? Probably not all that much. If we look at this subgroup in Mississippi, for instance, we're talking about half of the 40% of Clinton supporters in a state in which Democrats won about 40% of the vote in 2004. So 50% x 40% x 40% = about 8% of the voting pool.

But these voters are concentrated in states where Obama is not likely to be competitive anyway -- the states that we're referring to as the Deep South. In states that are even moderately more liberal -- states like North Carolina and Virginia in the South Coast region -- Obama's advantages with black voters and independents are enough for him to run at least evenly with Clinton. But in places like Mississippi, Tennessee, and Oklahoma, Obama has very little chance.

There's More...

Contract Post

Some "New" North Carolina polls

A little birdie told me about a couple of additional polls for North Carolina, from Public Policy Polling and Civitas.

The PPP poll shows both Obama and Clinton trailing McCain by 5. The Civitas poll shows Obama trailing McCain by 10 points, and Clinton trailing by 12.

The net result is to move Obama's Win Percentage in North Carolina slightly downward, but the state could be reasonably competitive.

Also, note that going forward, polls conducted within the past 10 days will now be highlighted in yellow in our detailed data table. This is not necessarily the same date that the poll is added to the website; the "date" column should reflect the median date of interviewing for the actual survey.

There's More...

Contract Post

Plenty of Democrats in Red States

Via the invaluable Pollster.com, I came across some information on the demographics of Democratic Primary voters from the current national Newsweek poll. Newsweek's sample consisted of 573 self-identified Democrats. Of those, 268 (46.8%) were in states that George Bush won in 2004, and 305 (53.2%) were in states that John Kerry won in 2004. That is hardly a seismic divide in red-blue allegiances, and speaks to why states that were not in play in the 2004 cycle may well be in play this time around.

For comparison's sake, there were 32,480,059 Kerry voters in Kerry states in 2004 (55.0%) and 26,547,419 in Bush states (45.0%). So, it looks like the map has turned at least incrementally more purple since then.

UPDATE: In fact, it appears that there has been slightly more turnout so far in Bush states than in Kerry states in Democratic primaries and caucuses. Using turnout figures primarily from this source, I get the following:

Red States

TX	2,865,945
OH	2,231,156
FL	1,734,456
GA	1,060,767
VA	973,931
MO	823,376
TN	622,586
AL	536,293
SC	532,468
AZ	443,324
OK	417,095
LA	384,243
AR	310,915
IA	236,000
NM	153,299
UT	124,307
CO	120,001
NV	117,559
KS	37,089
ID	21,224
NE	12,445
AK	8,621
TOTAL	13,767,100

Blue States

CA	4,416,779
IL	2,038,614
NY	1,891,143
MA	1,244,133
NJ	1,108,044
MD	760,314
MI	592,261
CT	337,269
NH	288,058
MN	206,559
WA	200,000
RI	186,036
VT	151,380
DC	114,012
DE	94,675
ME	44,670
HI	37,562
TOTAL	13,711,509

So thus far, about 50,000 more people have participated in Democratic nomination contests in Red States than in Blue States. And I don't know if the Blue States are going to have the chance to retake the lead. While there will be plenty of turnout in Pennsylvania, the same will also be true in North Carolina and Indiana; the Blue States may be banking on a re-vote in Michigan.

There's More...

Contract Post

Moving beyond the Big Three

For at least eight years now, Democrats have told themselves that they couldn't win the White House without carrying at least two of of the three "Big Three" states: Florida, Ohio, and Pennsylvania. And how has that strategy fared? Needless to say, it hasn't fared all that well: Democrats carried Pennsylvania in both 2000 and 2004, but lost each of Ohio and Florida both times, and thereby lost the election.

One of the advantages of my simulation model is that we can do some scenario testing. And today, the scenario I want to test is pretty simple. How do the two Democrats fare -- Clinton and Obama -- in a world where they never carry any electors from Florida, Ohio and Pennsylvania?

The procedure here is pretty simple. I ran a fresh set of 5,000 trial heats for each of Clinton and Obama. With all states included, Obama won 3,112 of his trials (62.4%) and Clinton won 2,062 (41.6%); these numbers differ very slightly from my last "official" simulation run because of random fluctuation. I then pulled states off the table one at a time: first Florida, then Florida and Ohio, and then Florida, Ohio, and Pennsylvania. That is, any electoral votes the candidate won from these states in the trial heats were eliminated.

First, let's see how this process worked for Obama:

With all states included, Obama wins 3,112 out of 5,000 trial heats (62.2%)

Without Florida, Obama wins 3,096 out of 5,000 trial heats (61.9%)

Without Florida or Ohio, Obama wins 2,867 out of 5,000 trial heats (57.3%)

Without Florida, Ohio, or Pennsylvania, Obama wins 2,389 out of 5,000 trial heats (47.8%)

There are several important things going on here. First, we see how utterly irrelevant Florida is to an Obama candidacy. Florida swung the election between a win and a loss for Obama in just 16 out of 5,000 simulations: or 0.32% of the time. According to our current polling averages, Florida is just Obama's 32nd best state, not counting the District of Columbia; states that rank ahead of it include North Carolina and Montana. By contrast, Florida is Clinton's 21st best state.

Obama also isn't harmed all that much by pulling Ohio off the table in addition to Florida; he still wins the election a solid majority of the time. Losing both Pennsylvania and Ohio in addition to Florida is more problematic for him. However, he still wins the election almost half the time (47.8%). The most likely such scenario is as follows:

Kerry States               252
- Pennsylvania             -21
                       =231
+ Iowa                      +7
+ New Mexico                +5
+ Nevada                    +5
+ Colorado                  +9
+ Virginia                 +13
                       =270

That is, we take John Kerry's 252 Electoral Votes and subtract Pennsylvania, leaving Obama at 233. We then add five states to Obama's column that Kerry didn't win in 2004, but which Obama is presently favored in based on our averages. As it turns out, this is just enough to get him past the 269 electoral vote hump and win the nomination; he finishes with 270 electoral votes.

There are some other parlays that might work here too. If you eliminate Virginia -- the diciest of the five Red states that Obama is favored to win -- you could replace it with North Carolina. Or, you could replace it with Missouri plus any of the prairie states (North Dakota looks like the most promising). You could even replace it with Missouri plus one of the congressional distrcits from Nebraska, which would get Obama to exactly 269 EV. These are not the best scenarios for Obama by any means, but they are highly plausible wins, particularly since resources diverted away from the Big Three -- big, expensive, oversaturated states in which voters may be hard to move at the margins -- could be reallocated to shoring up victory in a place like Virginia, or making some plays in North Carolina, or in the Great Plains.

Now, let's try the same exercise for Clinton:

With all states included, Clinton wins 2,062 out of 5,000 trial heats (41.6%)

Without Florida, Clinton wins 1,811 out of 5,000 trial heats (36.2%)

Without Florida or Ohio, Clinton wins 1,510 out of 5,000 trial heats (30.2%)

Without Florida, Ohio or Pennsylvania, Clinton wins 1,151 out of 5,000 trial heats (23.0%)

Without the Big Three included, Clinton's win percentage falls to just 23.0%; these are essentially the cases where she's won a landslide victory elsewhere around the country. If you absolutely had to create a scenario for Clinton that doesn't involve Florida, Ohio, and Pennsylvania, you would probably do it like this:

Kerry States               252
- Pennsylvania             -21
                       =231
+ Arkansas                  +6
+ West Virginia             +5
+ Tennessee                +11
+ Missouri                 +11
+ New Mexico                +5
                       =269

The math works out there -- Clinton gets to exactly 269 EV and would presumably win in the House of Representatives -- but several of these states are tough for Clinton. While she's a heavy favorite in Arkansas and a slight favorite in West Virginia, she is an underdog in Missouri (46.9% to win), New Mexico (37.2% to win) and Tennessee (35.0%). Moreover, the non-Pennsylvania Kerry states are not gimmes to Clinton in the way that they are for Obama. She is actually an underdog in four of these states: Oregon (25.9%), Wisconsin (29.4%), New Hampshire (32.8%) and Washington (38.9%). So without the Big Three, Clinton would have to win seven states in which she is not presently favored. And that's just to pull into a tie; if the Democrats lost the House and needed an outright majority, Clinton would also have to bring in a state like Iowa, Nevada, North Carolina or Colorado, in each of which she has polled significantly behind McCain. Obama, meanwhile, would just have to hold the fort across a number of states where he was already ahead.

But perhaps the most powerful version of the formulation is this: even if he never won an electoral vote from Florida, Ohio, and Pennsylvania, Obama would still be a better bet to win the nomination (47.8%) than Clinton would with every state on the table (41.6%). That's the advantage of a 50-state strategy.

p.s. Notwithstanding all that, Obama still rates slightly ahead of Clinton in both Ohio and Pennsylvania in our weighted polling averages, although she is a heavy favorite over him in Florida. The group of swing voters is very different between the general election and the primaries, and so are the mechanics of inducing turnout. Although clearly there are some relationships between performance in the primaries and the general election, they are by no means dispositive.

There's More...

Contract Post

538 versus Intrade

Intrade already has futures ready to go on the state-by-state results for the 2008 general election (disclosure: I don't gamble on politics, but I don't mind if you do), and I thought it would be worth comparing the best buys and sells.

Since we don't yet know which Democrat is going to be nominated, I have assumed a 75% chance of Obama as the nominee and a 25% chance of Clinton, which are essentially the same prices that have been forecast by Intrade. I've then produced a blended estimate of the Democrats' chances of carrying each state based on this 75/25 split, and compared it to the most recent trade on Intrade.

As of right now, it looks like the best buying/selling opportunities are as follows. The percentage listed are the fivethirtyeight.com estimate that the Democrat wins, followed by the most recent Intrade trade on the Democratic contract in that state.

1. North Dakota: buy Democrat (35.8% versus 10.0%; 25.8% discrepancy). Even if you don't quite believe the Survey USA poll that has Obama ahead in this state -- our regression model is a little skeptical -- this contract looks cheap at $10.

2. Pennsylvania: sell Democrat (53.5% versus 75.0%; 21.5% discrepancy). The polling averages here show a dead heat against either Democratic opponent. Will six weeks of scotched earth campaigning help or hinder those percentages?

3. Indiana: sell Democrat (9.2% versus 30.0%; 20.8% discrepancy). I'm not sure why Intrade is so sanguine about Democratic prospects here. Obama has an outside chance to take the state, if he can crush McCain in the NW part of the state that is tethered to Chicago, but it's only an outside chance, and Clinton has no chance at all.

4. Mississippi: sell Democrat (4.7% versus 24.5%; 19.8% discrepancy). It looks like the Democratic contract that was executed at 24.5 percent was a one-off -- you can't currently buy a contract at that price -- so I wouldn't read too much into this one. It's one of the worst states for the Democrats, even relative to elsewhere in the South.

5. Michigan: sell Democrat (67.8% versus 84.5%; 16.7% discrepancy). Boy, we're starting to see a pattern here, are we not? Intrade may be placing too much emphasis on the results of the 2004 election.

6. Florida: sell Democrat (26.8% versus 43.2%; 16.4% discrepancy). As I've written, it's certainly a swing state if Clinton is the nominee, but Obama has not polled at all well here and has plenty of better places to compete.

7. California: buy Democrat (96.4% versus 80.0%; 16.4% discrepancy). It's not like this state is getting any less Blue. McCain may make some pretense of competing here, but it looks like he'd be wasting his resources.

8. Colorado: buy Democrat (67.4% versus 51.5%; 15.9% discrepancy). This one entirely depends on Obama winning the nomination, who polls much stronger than Clinton here.

9. Virginia: sell Democrat (41.3% versus 57.0%; 15.7% discrepancy). The Intrade contract would be priced fairly reasonably for Obama, but Clinton is a long-shot here.

10. Kentucky: sell Democrat (3.5% versus 18.5%; 15.0% discrepancy). Obama is not competitive in this region, and states like Tennessee are better bets for Clinton.

There's More...

Contract Post

FAQ

With the addition of the FAQ -- which will be continually available both from the link at the top of this column and at the bottom of this page -- this site should be considered fired up and ready to go. We are no longer in Beta Test mode.

There's More...

Contract Post

Regions

Along the left-hand sidebar, you should see top-level results for each of the 50 states plus the District of Columbia, grouped into eight geographic regions.

The regions are meaningless other than as a way to organize the presentation of the data. However, they are similar to official regions as designated by the US Census Bureau, with a couple of exceptions:

I group Maryland, Delaware and the District of Columbia into the "Mid-Atlantic" region along with New York, New Jersey and Pennsylvania. The Census Bureau considers them part of the South Atlantic region.

The West South Central and East South Central regions are grouped together into a region I call "Deep South". I also include West Virginia in this region. West Virginia is an extemely difficult state to characterize. Geographically, it is not all that close to the other states in the Deep South (although it is contiguous with Kentucky and thereby the rest of the region) but it seems to fit better there than anywhere else given its demographics and political culture.

What remains of the South Atlantic Region -- Virginia, North Carolina, South Carolina, Georgia and Florida -- I have dubbed the "South Coast" region. These states have pockets of liberalism
that other Southern states don't, and should generally be more competitive in November.

The other states are grouped into the exact same regions that the Census Bureau uses, but I use a somewhat more evocative set of names. "East North Central" becomes "Great Lakes", and "West North Central" becomes "Prairie". "Mountain" becomes "Interior West". "Pacific" and "New England" stay as is.

...as you should see, these regions form a relatively coherent set of electoral units, although almost every region has an exceptional state or two.

There's More...

Contract Post