Zogby is One of the Night's (Other) Big Winners

Certainly the news media is emphasizing that Obama is the night's big winner. But as I noted earlier, there were five pollsters who released polls in both states in the last few days and they didn't always make the same predictions. I've reproduced the figure from that post here, but added the actual results from tonight's voting (at least as they stand now, with 99% reporting in both states).

As you can see, Survey USA was the night's big loser, at least as far as pollsters go. They were estimating a good night for Clinton, including a 12% win in Indiana and a mere 5% win for Obama in North Carolina. On the opposite end, Zogby was the big winner. Zogby came closest to the North Carolina result by predicting a 14% victory for him in that state. Zogby was also the only pollster to have predicted an Obama win in Indiana, though Clinton ended up winning by about that margin. Of course, lest any pollster get a big head, these pollsters have been in the opposite positions (Zogby as the big loser and Survey USA as the big winner) in earlier primaries this year.

UPDATE: I should've added that Public Policy Polling was also fairly accurate relative to the other pollsters and, unlike Zogby, they had the winner right in both states.

UPDATE: Who needs polls? As one of our readers points out:

"Probabay the best prediction was Poblano at the 538 website, who's demographic model predicted Obama +18 in NC (final was +15), and Clinton +2 in Indiana (right on). Poblano also came very close in Penn."

May 6th Primaries: Live Blog

1:22AM: Looks like Lake County fell short for Obama and Clinton will have won Indiana by about 20-25K votes. This victory, however, is much smaller than most expected and Obama's margin in North Carolina was bigger than most estimates in the last few days. Clinton will possibly take a net gain of 1 or 3 delegates from Indiana while Obama will likely take a net gain of more than 10 delegates from North Carolina. All in all, Obama's big night allows him to cancel out the gains (in delegates and popular vote) that Clinton made in Pennsylvania two weeks ago.

Clinton has canceled her morning talk show appearances and is said to be contemplating what her next move is. Obama, in the meantime, is evidently planning to take on the air of a presumptive nominee.

This is a split decision that clearly favors Obama. Now, will large numbers of superdelegates begin falling in line behind his campaign? It may not take too long to find out the answer to that question.

Thanks for everyone who followed this blog tonight. According to Google Analytics, there were well over 1,000 visitors on this blog today and that number should increase since we are still waiting for the numbers to come in from Lake County.

1:00AM: Just noticed that the exit poll estimates have been revised (as they typically are to come in to line with where turnout is greater) and African American turnout in Indiana is now estimated to have been 18%. That is up from a 14% figure reported earlier. As I noted a few days ago, assumptions about African American turnout were driving the different poll figures we were seeing in the state. Blacks turned out at a greater rate than most pollsters expected in Indiana and that significantly affected the outcome.

12:40AM: The early voters in North Carolina went a little more strongly for Obama than did those who voted today, but the difference wasn't huge. With just a few precincts left, Obama won 59% of the early vote and 55% of the votes that came in today. As it stands now, 25% of the total vote came from early voters.

12:27AM: I think the networks are so focused on Lake County that they forgot about Monroe County. With 97% of the precincts in there, Obama won the county by a little over 5k votes. That is 1k more than the network sites are currently showing.

12:12AM: Tim Russert just said, pointedly, "We now know who the Democratic nominee is going to be and nobody is going to dispute it."

12:09AM: The CNN discussion right now is focusing on whether there is some conspiracy behind why it took so long for the Lake County returns to be reported. The Hammond (also in Lake County) mayor is on saying he turned in his vote hours ago.

11:59PM: Worth noting now that things has closed is that there are a small number of uncounted votes in Indianapolis and 1/3 of the vote still out in Bloomington.

11:55PM: Also worth noting is that Obama's margin (in total votes) in NC looks like it will be slightly bigger than her margin in Pennsylvania, wiping out any gains Clinton made in the popular vote gap two weeks ago.

11:49PM: It is important to note at this point that who wins or loses Indiana has little practical importance when it comes to delegates. It may mean the difference between one delegate. The only reason we care about this is perception...it will drastically effect how her candidacy is viewed by the media and superdelegates in the coming days.

11:45PM: Lake County is beginning to come in and it is already making a difference. Obama just cut Clinton's lead in half, from about 40k to about 20k.

11:32PM: Evidently the mayor of Gary is saying that there may very well be enough votes for Obama to take over the lead.

11:29PM: Turnout in the Democratic primary in NC has already topped 1.5 million. That is as many as voted for John Kerry in the 2004 general election. To get that kind of turnout in a primary is truly impressive.

11:04PM: Tim Russert reports that there are still 220k votes uncounted in Indiana. Clinton's lead is about 40k. If Russert is right, Obama needs to win about 60% of the outstanding vote to win Indiana. Certainly a possibility.

10:49PM: Clinton holds about a 40k lead over Obama and Lake County still has not come in. At this point, I'd say that it is unlikely that there are enough votes there for Obama to overtake her lead. Not impossible, but unlikely.

10:42PM: Oops, rest of Tippecanoe County just came in. It went 58% for Obama (he picked up about 5k votes there).

10:40PM: Here are the counties that still have significant votes that haven't been reported:
Porter County (about 30% missing) going 58% for Clinton so far.
LaPorte County (about 50% missing) going slightly for Clinton so far.
Tippecanoe County (about 50% missing) going 60% for Obama so far.
Monroe County (about 50% missing) going 66% for Obama so far.
Hamilton County (about 10% missing) going 60% for Obama so far.
Lake County (100% missing) supposed to go for Obama.
Union County (100% missing) not much vote there, even when they do come in.

10:16PM: Key statistic on Lake County: about 25% of the population there is African American. Is that, and its closeness to Chicago, enough to produce the big margins that Obama needs there?

10:09PM: Just to give you a sense of how many votes could be in Lake County, in 2004, about 115K voted for John Kerry in Lake. If there are 100K votes there, Obama needs to win about 2/3 of them to have a shot at pulling out a win.

10:05PM: Tim Russert now says that Lake County may not be in until 11pm. Evidently extra ballot counters were brought in to count the votes (are we doing this the old fashion way?). In the meantime, Clinton is supposed to be teaching soon, but what can she say without having won anything yet?

9:55PM: Evidently Lake County will begin reporting at 10pm. That county is going to decide Indiana. But even if Obama comes up short, he appears to be the clear winner tonight, at least as much of a winner as one can be when they lose one and win one.

9:52PM: There was a lot of Zogby bashing going on by the readers at Pollster.com and other sites over the last week or so. But at this point, it looks like he is going to be closest to getting both NC and IN right.

9:42PM: Chuck Todd says the Obama campaign thinks that they will come up just short (by about 10-15k or so). If it ends up that close, then there is no way CBS could've confidently known that he was going to win the state an hour ago. Unless something odd happens and it doesn't tighten significantly, the call was a mistake, regardless of whether or not they end up being wrong at the end of the night.

9:25PM: Here is why Indiana is now too CLOSE to call:
Monroe County (Bloomington) is only 10% in.
Marion County (Indianapolis) still has about 1/4 of the vote not in.
Hamilton County (Wealth suburbs of Indianapolis) is only 1/3 in.
Lake and LaPorte Counties (Chicago area) are reporting nothing.

These should all be Obama areas. Clinton is up by less than 50k votes now. Are there enough votes for him in these counties? Quite possibly.

9:15PM: Obama just congratulated Clinton on "what appears to be her victory" in Indiana. Does he know something we don't or is he unaware of how things are tightening there?

9:12PM: Will CBS end up with egg on its face? MSNBC just changed its IN statement from "too EARLY to call" to "too CLOSE to call." The change is important, because it reflects that the vote there is tightening and still not a single vote in from Gary.

9:10PM: This is a rare night (at least compared to the last several election nights) where Obama gets to come out early to celebrate while Clinton sits back and waits out "her" state's results.

8:52PM: CBS was too early in calling Indiana. Only about half of the Marion County vote is in, and he is winning that by about 25K votes. If he wins the second half by the same margin, that cuts her lead in half. And not a single vote has come in from Gary yet. IF Clinton wins this state, it is going to be fairly close and I think the votes are out there for Obama to catch up. I'm not saying it WILL happen, but it certainly COULD and the early call from CBS was probably too soon.

8:48PM: The NC Board of Elections has a top-notch website complete with county maps and all sorts of other tools for looking at the primary results. The IN Secretary of State site struggles to calculate percentages. I'm just saying...

8:44PM: Wow, quite an exchange between Paul Begala and Donna Brazile on CNN. The long primary campaign appears to be wearing on even them.

8:29PM: On MSNBC, Howard Fineman just said that the Obama campaign now thinks that May 20th will be the day they wrap up the nomination because that is when they will have a majority of pledged delegates. Sound familiar? Well, I blogged this very point on March 31st. This is what I wrote then:

"Based on the delegate totals that Obama's campaign thinks he will win in the upcoming states (from the memo that the campaign inadvertently sent to a Bloomberg reporter), Obama would pass this milestone on May 20th, after picking up 28 delegates in Oregon and 23 in Kentucky. ... If he does clinch this on May 20th, might Clinton call it quits at that point? Certainly she would be facing increasing calls to get out of the race if she stayed in. In addition, on the 20th, she will likely have won Kentucky but lost Oregon. Thus, she would be able to leave the campaign on a day that she carried a state. The only states voting after the 20th are Puerto Rico and then Montana and South Dakota. She could win Puerto Rico, but she is likely to lose Montana and South Dakota and none of these states will do much to change the delegate count. If she left the race on May 20th, she could do it on a relatively high note and on her own terms, whereas anything after that may make it look like she was forced."

8:26PM: The NC Board of Elections breaks out the early vote differently from those voting today. It appears as though they've counted a little more than half of the early vote so far, and Obama leads with 61%. His percentage among election day votes counted so far is 54%.

8:18PM: Good thing Obama was in North Carolina this afternoon. He wouldn't have been able to do this in Indiana this afternoon. Bars in Indiana are closed while the polls are open. Wouldn't want anyone drinking and voting.

8:15PM: Evidently CBS called Indiana for Clinton. I would imagine that the other networks will follow soon. But still no vote in from Gary.

8:00PM: Here is the comparison between early voters and the racial and gender composition of the NC exit polls:

	% of Early Voters	Exit Poll
Men	38.7%	43%
Women	60.8%	57%
White	56.5%	62%
Black	39.9%	33%
White Women	33.2%	35%
White Men	23.1%	28%

7:55PM: 19% in North Carolina are voting in their first primary...they favor Obama 68-28%.

7:50PM: Clinton holds a comfortable lead in IN right now, but only 11% of Marion County (Indianapolis) is in and nothing in yet from Lake County (Gary).

7:47PM: Some talk on MSNBC right now about whether Superdelegates will start flocking to Obama (and calling on Clinton to leave the race) if Obama wins as big as it appears in North Carolina tonight.

7:40PM: The early exit polls are showing a 55-41% lead for Obama in NC. Usually, his lead shrinks, but I'll be curious to see if it does this time. We already know that the exit polls are overestimating the percentage of early voters who were white women, which would seem to inflate Clinton's support.

7:35PM: The initial exit polls indicate that blacks were 33% of the NC Democratic electorate. Will be interesting to see if that number creeps upwards closer to the 40% figure for early voters. But the work that the Obama campaign and other groups did in getting African Americans out early in NC is one of the most under-told stories of the campaign.

7:30PM: At the minute the polls close, most networks call NC for Obama. That's big for him, since it likely means a big victory. No unpleasant surprises for Obama tonight.

7:29PM: Apparently Romney and Huckabee were both on the IN Republican ballot. What will a strong second place showing mean for Huckabee's VP chances? What does languishing in 4th do for Romney's? Probably nothing at all!

Candidate	Vote	%
John McCain	42,744	76.4%
Mike Huckabee	5,990	10.7
Ron Paul	4,354	7.8
Mitt Romney	2,841	5.1

7:26PM: Question: will the networks be able to call NC right away? If not, how long will it take?

7:21PM: 21% of those voting in the IN Democratic primary were voting in a primary for the first time ever. That group went for Obama 59-40%.

7:11PM: According to early exit polls, 14% of IN voters were black. That is slightly higher than most pollsters were estimating, but lower than the Howey-Gauge poll that showed Obama ahead. (See here)

7:03PM: Networks say Indiana is too EARLY to call. Suggests Clinton wins by 5-10%?

6:59PM: MSNBC notes that exit polls show that 37% of early voters are white women. That figure is actually higher than the actual number, which is 33%.

6:50PM: The New York Times, perhaps inadvertently, has put the following exit poll data from Indiana online:

% of total		Clinton	Obama
14	17-29 years old	42	58
23	30-44 years old	43	57
34	45-59 years old	51	49
29	60 years old and older	67	33
85	Less than $100,000	52	47
15	$100,000 or more	52	48
17	Today/In the last three days	62	38
82	Sometime last week or earlier	50	50

Based on these figures (particularly the income figures) it looks like Clinton is up 52-47% in the early exit polls. Keep in mind, as well, that Obama usually does better in these than he ends up doing in the actual vote totals.

Obama did well among younger voters, but the 60+ crowd remains a tough group for him.

6:33PM: Interesting that the networks are expected to wait to call states until everyone has voted, yet election boards in Indiana are already reporting their vote totals. Ah, the ethical questions raised when your state spans time zones.

6:17PM: In case you hadn't noticed.

6:00PM: The polls just closed in most of Indiana. However, the areas near Chicago are on central time, so they have another hour to vote.

Some things to think about tonight:

1) Beware of early vote totals, particularly in NC. We already know that Obama is going to win big among early voters in NC. In Texas, the early voting returns were the first to come out, so Obama held a lead for the first few hours after the polls closed there. Eventually, Clinton overtook Obama based on the vote from the day of the primary.

2) For the same reason as above, be especially skeptical about the early exit polls. These have tended to favor Obama, but will they in NC when so many Obama supporters voted early?

3) The memo accidentally released by the Obama campaign after Super Tuesday projected that he would win Indiana 53-46% and North Carolina 53-45%. The IN result seems doubtful, the NC prediction may be closer to the mark.

4) What will African American turnout be in NC and IN? In NC, 40% of early voters were black, but pollsters have been estimating that African Americans would make up about 30-35% of the NC electorate. If it is closer to the 40% figure, it should be a blowout for Obama. Closer to the 30% figure, and it is a tight race. In IN, black turnout will also be very important. Most polls were estimating that African Americans would make up about 10-12% of the Democratic electorate. If that number climbs higher, the chances of an Obama upset increase.

Final Democratic Delegate Estimates for North Carolina and Indiana

It looks like all of the pollsters have weighed in with their final polls at this point. Therefore, I'll go ahead and make the final poll-based predictions on how the delegates will be allocated according to tonight's primaries. As always, I use the Pollster.com averages to generate these estimates.

As you can see, the polls suggest that Clinton will take a net gain of 4 delegates from the Indiana primary while Obama would take 9 from his victory in North Carolina. Overall, Obama would take 5 more delegates from today's primaries than Clinton.

Of course, the polls have varied significantly in both Indiana and North Carolina and certain pollsters sometimes seem to favor particular candidates. As a result, we can also put together a best-case scenario for each candidate. Five survey firms have polled both Indiana and North Carolina during the final week of the campaign. The margins those firms give to Obama in each state are plotted below.


What is notable about this figure is that four of the five pollsters fall on a diagonal line across the scatter plot (only Insider Advantage falls off this line). This indicates that the pollsters who see Obama doing better in NC also see him doing better in IN, and vice versa. You wouldn't necessarily expect this to be the case unless these firms are using assumptions that systematically advantage one candidate in both states.

Zogby would have to be the favored pollster of Obama supporters. He is the only pollster (of these five) who projects an Obama win in Indiana and he also has Obama winning by the largest margin in North Carolina (14%). If Zogby is correct, Obama would take a net gain of about 20 pledged delegates from today's primaries. More importantly, the calls for Clinton to leave the race would likely intensify.

On the other hand, Survey USA has to be the Clinton campaign's favored pollster. They see Clinton winning by 12% in Indiana and only losing North Carolina by 5%. If this happened, Clinton would likely win two or three more delegates than Obama today.

So, what will it be? Will the results follow the Pollster.com averages and produce 5-10% margins for Clinton in Indiana and Obama in North Carolina? Will Obama or Clinton have a better day than expected, as forecast by Zogby or Survey USA? We only have a few more hours to wait.

And don't forget, I'll be live blogging this evening as the results come in. Should be interesting.

UPDATE: Check out Mark Blumenthal's excellent post on the IN and NC polls here.

Final Pre-Election Analysis of North Carolina Early Voters

I noticed this morning that the North Carolina State Board of Elections has updated their early voting data and added a few more early voters. In all, they now report that 495,828 North Carolinians have voted early in advance of today's primary. The vast majority of these early voters (406,125) are casting their ballots in the Democratic primary.

While the number of those voting early in the Democratic primary has increased slightly since I posted on Sunday, the race, gender and party balance of these voters has remained the same.

	% of Early Voters
Men	38.7%
Women	60.8%
White	56.5%
Black	39.9%
White Women	33.2%
White Men	23.1%
Democrat	84.1%
Unaffiliated	15.8%

I've already blogged about how the composition of these early voters would seem to strongly favor Obama. So let's look at some other information about early voters.

The congressional district with the most early voters is district 4. This district includes both Chapel Hill and Durham and has a significant African American population (about 21%). The district with the second most early voters is the 11th district, which is Heath Shuler's district in the western most tip of the state. This district has a much smaller minority population and is often viewed as a more conservative district in the state.

Some other statistics of early voters:

# of Early Voters in Top Cities
Charlotte 39,216
Raleigh 23,881
Greensboro 20,596
Durham 19,276
Fayetteville 14,728
Winston Salem 14,182
Chapel Hill 13,628
Asheville 11,051
Wilmington 10,071

Precincts with most Early Voters
Nashville (Nash County) 1,099
Pittsboro (Chatham County) 1,096
West Williams (Chatham County) 1,073
Precinct 44 (Durham County) 1,041

And your totally pointless statistic:

Most popular first names of Early Voters
James 10,577
Mary 8,953
William 8,337
John 7,871
Robert 7,616

Check the blog tonight as I will be live-blogging as the returns come in. Polls close in Indiana at 6pm (though since parts of the state are in the Eastern Time Zone, they effectively close at 7pm). Polls close in North Carolina 30 minutes later.

Updated Democratic Superdelegate Predictions: Introducing New Variables and Subtracting Others

If you read Carl Bialik's story about my superdelegate predictions, you'll recall that he mentioned that one factor that I could consider adding was which candidate the other committed superdelegates in the same state were supporting. This was easy enough to create, so I added it and when I did, it was a very significant predictor. This isn't too surprising since superdelegates from the same state are often facing the same considerations when making their decision. In fact, this new variable was so important that it washed out the effects of other predictors that I had been using, such as the percentage of the state's population that belonged to a union, the percent living in urban areas, the percent with a college degree, and the per capita income in the state. Therefore, I am revising the methodology for the predictions to simplify the model and, hopefully, give it greater predictive power at the same time.

I now include just 5 variables to predict who superdelegates will endorse. These variables are the gender of the superdelegate, the presidential vote in the superdelegate's state or congressional district (for House members), the percentage of the state's superdelegates who are supporting Clinton, whether Clinton or Obama won the state's primary/caucus, and whether the superdelegate made their endorsement before or after Super Tuesday. The last variable accounts for the fact that people who didn't line up behind Clinton early, when she was the front runner, are far less likely to endorse her now. Indeed, this proves to be a very important predictor in the model (and failing to include it changes the predictions substantially as I'll show below).

Based on this new model, I have now updated the superdelegate predictions. As always, information on the superdelegates is provided by the Democratic Convention Watch site. In the figure below, I present the distribution of unpledged superdelegates based on the probability of supporting Clinton: Superdelegates who are between 40% and 60% likely to vote for Clinton/Obama are labeled as "unclear." There are 61 superdelegates in this range. There are 139 unpledged superdelegates who are at least 60% likely to vote for Obama; just 41 unpledged superdelegates are at least 60% likely to vote for Clinton. These predictions suggest that unless something dramatically changes, Obama will be able to cut into and even overtake Clinton's superdelegate lead in the coming weeks and months.

The estimates for each unpledged superdelegate are listed here. Note that I am now generating predictions for superdelegates in NY, AR, and IL, which I was not doing previously. Not surprisingly, all unpledged superdelegates in NY and AR are estimated to go for Clinton while all unpledged IL superdelegates are predicted to support Obama.

-----------------------------------------------------------------

FOR THOSE INTERESTED:

I do want to return to a point I made above. It matters quite a bit if you include a variable in the model that accounts for when a superdelegate made his/her endorsement. This variable captures whether a superdelegate endorsed before Super Tuesday or if they endorsed after (or have not yet endorsed). This variable is intended to capture the dynamic aspect of the race that led many superdelegates to endorse Clinton before Super Tuesday, but then caused more to flock to Obama after Super Tuesday. But what happens if you ignore this factor? The figure below presents predictions from a model that removes the variable accounting for when a superdelegate made his/her decision.

As this figure clearly indicates, the predictions change dramatically when you don't account for the timing of a superdelegate's decision. In this model, 72 unpledged superdelegates are in the "unclear" range, 71 are at least 60% likely to endorse Obama and 98 are at least 60% likely to endorse Clinton. Thus, ignoring the dynamics of the race tends to favor Clinton. However, it is important to note that even in this scenario, Clinton would likely not pick up enough superdelegates to overtake Obama's overall delegate lead.

Analysis of Early Voters in North Carolina: Obama in the Lead?

I have now posted several times on the composition of early voters in North Carolina. My last post analyzed those who had voted early through Thursday (May 1st). Since then, there has been a huge spike in the number of early voters. The figure below shows the number of early votes by date, with over 80,000 cast on Friday (May 2nd) and nearly 60,000 cast on Saturday (May 3rd).


In total, 486,786 valid early votes have been cast at this point with 398,635 (82%) of these being cast in the Democratic primary. Assuming these numbers don't change, about 13% of registered Democrats cast their votes early for this primary while 5% of unaffiliated North Carolinians voted early in the Democratic primary. It is hard to know just how much of the total vote that early voters will represent, but in 2004, about 1.5 million voted for John Kerry in the general election. Perhaps that gives us a sense of what a large turnout primary will look like. If we assume that turnout in the Democratic primary will be in the 1.5 - 2 million range, then between one-fourth and one-fifth of the electorate has already voted. If turnout is going to be less, then early voters represent an even larger share.

What do these early voters look like? Below, I present the gender, race, and party registration figures for those voting early in the Democratic primary.

	% of Early Voters
Men	39%
Women	61%
White	57%
Black	40%
White Women	33%
White Men	23%
Democrat	84%
Unaffiliated	16%

The most remarkable figure to note here is that 40% of early voters were African-American, a figure that is up from 37% earlier this week. It is important to note that most of the polls being conducted in North Carolina are assuming African-American turnout of about one-third; if it is closer to 40%, then this clearly will advantage Obama.

Clinton's most loyal supporters, white women, make up about one-third of early voters. The "swing" group of white men make up nearly one-quarter of early voters in North Carolina. Finally, voters registered as "unaffiliated" make up 16% of the electorate so far.

If we assume that Obama wins 85% of the African-American vote, just 30% of white women and 40% of white men and then we split the remaining 6% of the early voters (those in the "other" category) evenly between Obama and Clinton, then Obama is currently leading Clinton among early voters by a margin of 56-44%. In terms of raw numbers, that would give Obama a lead of somewhere close to 50k votes. And this estimate is probably slightly on the conservative side since Obama could very well win 90% or more of the black vote and could do a bit better among white men and women (indeed, reader "x curmudgeon" notes that surveys show early voters going 63-31% in favor of Obama).

So, Obama is almost certainly ahead by a significant amount at this point. The question is whether the composition of the electorate voting on the day of the primary will look like early voters. If it does, then Obama should have a very good day. But if African American turnout on the day of the primary doesn't match that among early voters, then this race could end up being much closer.

Update: dean4ever has posted a similar analysis today.

Who is Winning the Early Vote in North Carolina?

As of Friday morning, 271,388 voters had cast valid ballots for the Democratic primary in North Carolina. So which candidate is faring better among these early voters?

The votes themselves won't be counted until the polls close on May 6th, but we can get a sense of who is ahead based on the race and gender of those who have already voted. As has often been noted during this campaign, race and gender tend to be highly correlated with vote choice. African-Americans have voted for Obama at rates upward of 85-90%. On the other hand, about 65-70% of white women have supported Clinton. White men have been a little less predictable. Obama won 58% of the white male vote in Virginia, 43% in Pennsylvania, and just 38% in Ohio.

So far, the early voters in North Carolina break down as follows:

	% of Early Voters
Black	38%
White Women	34%
White Men	24%
Other	6%
Total	271,388

Let's assume that Obama is winning 85% of the African-American vote, just 30% of white women and 40% of white men. We'll split the remaining 6% of the early voters (those in the "other" category) evenly between Obama and Clinton. Under these assumptions, Obama is currently leading Clinton among early voters by a margin of 55-45%. In terms of raw numbers, that would give Obama a lead of somewhere between 27,000 and 28,000 votes. And this estimate is probably slightly on the conservative side since Obama could very well win 90% or more of the black vote and could potentially do a bit better among white men and women.

So, Obama is almost certainly ahead by a significant amount at this point. The question is whether the composition of the electorate voting on the day of the primary will look like early voters. If it does, then Obama should have a very good day. But if African American turnout on the day of the primary doesn't match that among early voters, then this race could end up being much closer.

Race and the Divergent Indiana Polls

UDPATED (5/2, 12:29pm): I have replaced the earlier American Research Group survey with the one released today and added the Downs Center poll that was just released.

I spent 5 years in Bloomington, IN while I got my PhD from Indiana University so I can tell you first-hand that the state is not the most racially diverse place. Yet, as I poured over the polling released over the past week in advance of the Indiana primary I came to realize that the relatively small African American population in Indiana may play a fairly significant role in the Democratic nomination race.

There have been several polls released out of Indiana during the past week. Surveys by Howey-Gauge and Research 2000 put Obama in a statistical dead-heat with Clinton, while surveys conducted by Public Policy Polling, Survey USA, TeleResearch, and American Research Group. One explanation for these divergent results may be the assumptions that pollsters are making about black turnout in the state. The figure below plots the percentage of each polling firm's sample that was black and the margin for Obama.


Note that most of the pollsters are using samples that are about 10-12% black, while Howey-Gauge stands out for its sample which was one-fifth African American. (Unfortunately, I was unable to track down the racial makeup of the Research 2000 poll, which also showed a statistical dead-heat). It is not surprising to find that the one survey organization who assumes much higher black turnout would be the one organization showing a small lead for Obama (rather than a significant deficit).

But which polling firm is making the right assumptions? African Americans make up about 9% of Indiana's population, but what percentage of the Democratic electorate might they comprise? Unfortunately, the Indiana presidential primary has not mattered for quite some time, so there is really no baseline for understanding what African American turnout would be in a typical presidential primary. Therefore, the best I can do here is use exit polls from the 2004 general election to get a sense of what share of the Democratic electorate is comprised of African American voters.

According to the exit polls, about 17% of Hoosiers who voted for John Kerry in 2004 were black. This figure is closer to the Howey-Gauge estimates than those being used by the other pollsters. On the other hand, blacks made up 15% of all Indiana voters who identified as Democrats in 2004, and 13% of those identifying as Democrats or independents. This figure would fall closer to the what the other pollsters are estimating for African American turnout in this primary.

Ultimately, we don't really know what to expect in terms of African American turnout. If it is around 10%, then it looks like Clinton may win by 5-10%. If black turnout approaches 20%, then Obama may be able to pull out a narrow victory that would go a long way to ending the race. Could African Americans in Indiana really play such a decisive role? Possibly.

Democratic Delegate Estimates for North Carolina and Indiana

With the May 6th primaries fast approaching, there have been a slew of new polls conducted in Indiana and North Carolina. But how do these polls translate into predictions of how many delegates each candidate might receive in the May 6th primaries? As usual, I'm using the Pollster.com poll averages from each state to estimate how many delegates Obama and Clinton will receive in each state. These estimates are presented here:

Based on these estimates, Clinton would take a net gain of 8 pledged delegates from Indiana while Obama would take a net advantage of 17 delegates from North Carolina. Overall, this would give Obama a net gain of 9 pledged delegates from the May 6th primaries.

My final estimate for the Pennsylvania primary was an 84-74 delegate split in favor of Clinton. Right now, MSNBC is showing an 83-73 split from Pennsylvania with two delegates still to be allocated. This means that, at most, Clinton will have taken a 12 delegate edge from her victory in Pennsylvania and, if the current estimates are correct, Obama will make most of that up on May 6th. As if the point needs to be belabored any further, barring something extraordinary, it is impossible for Clinton to catch up to Obama in pledged delegates at this point.

What else do these estimates mean? Well, I'm correct, Obama will have about 1,590 pledged delegates in his column by Wednesday morning. If you do not include Florida or Michigan, a candidate needs 1,627 to clinch a majority of the total pledged delegates. He will be about 30 delegates away from that mark. If you add in superdelegates, Obama presently stands at 1,739 (according to MSNBC) with 2,024 needed to capture a majority. By Wednesday morning, he should be less than 200 delegates away from that magic number.

How do the North Carolina Polls Compare to the Composition of Early Voters?

In the last two days, surveys of North Carolinians have shown a 12 point Obama lead (Public Policy Polling), a 10 point Obama lead (American Research Group) and a 5 point Obama lead (Survey USA). So far, we aren't seeing the widely divergent polling results that we had about a week before the Pennsylvania primary. On the other hand, there is a big difference between a 5 point Obama lead and a 12 point lead. After all, the former will be seen by pundits as a Clinton win while a double-digit margin would be enough for Obama to claim victory.

For the surveys leading up to Pennsylvania, I blogged about comparisons between the demographic compositions of the samples drawn by different pollsters. I present some similar comparisons here between the last three polls taken in North Carolina. However, North Carolina's Board of Elections offers us an additional interesting tool in evaluating the composition of these surveys' samples--a tally of actual early voters. As of Tuesday morning, over 144,000 voters had cast early ballots in the Democratic presidential primary. The racial, gender, and party balance of those voters is presented in the table below.


For the most part, the composition of the pollsters' samples appear to be fairly close to that of those who have voted early so far. However, there are some small differences. For example, each polling organization has African Americans comprising one-third of their sample, but so far they have made up 37% of early voters. In addition, 59% of early voters have been women, but the pollsters samples are comprised of slightly fewer women. On the whole, when it comes to race, gender, and party registration, American Research Group appears to have a sample that most closely mirrors the composition of early voters. However, it is important to note that the differences across polling organizations are not major.

One final important point to make is that early voters are not necessarily a representative cross-section of the eventual electorate. Research suggests that those who take advantage of early voting tend to be of higher socioeconomic status than the regular electorate. Only 4.5% of registered Democrats have voted early so far while about 1.8% of those registered as unaffiliated have early voted in the Democratic primary. Bottom line: the early voting figures may be a guide to the eventual composition of the electorate, but right now they represent just a small share of everyone who will eventually vote in this primary.

Answering Questions About the Superdelegate Predictions

Since I've begun generating the Superdelegate predictions, I've gotten several questions from readers about how I generate the predictions or why various things are missing from the model. For some of the questions, you might find answers in Carl Bialik's recent blog posting about this site. I've also taken a stab at three questions below...

1) What about John Edwards (or other missing superdelegates)?

We originally left out special superdelegates, like John Edwards (CORRECTION: John Edwards is NOT a superdelegate), who are in their position because they are a former president, speaker of the House, etc. The issue with these delegates is that unlike active politicians, they don't have an obvious constituency, which makes their behavior far less predictable.

If you were assume that John Edwards' constituency is the people of North Carolina, then the model would likely predict that he would support Obama (as it does for other male politicians from North Carolina). However, he is a bit of a free agent since he doesn't really have to answer to any particular group, and that makes it far more difficult to model his behavior (or that of other superdelegates like him).

CORRECTION: Thanks to Matt from the Democratic Convention Watch site for pointing out that John Edwards is NOT a superdelegate. Being a VP candidate on a losing ticket does not qualify you for a vote. Since we took our superdelegate list straight from the Democratic Convention Watch website, Edwards has never been in our dataset, so this has not affected our predictions.

2) Why is race not also included in the model? I would expect that it would have similar predictive power to sex.

When my research assistants originally collected the data for these models it was very hard to find information like race on some of the superdelegates who are not governors or members of Congress. Many of these superdelegates were not well known and google searches revealed little to us about their race or ethnicity. This remains true about at least some of the superdelegates. Therefore, we had to exclude this variable from the model. If there is a central source of information that would allow us to fill in the holes on the race of superdelegates, let me know and we will add it to the model.

For what it's worth, I re-ran the model with the superdelegates that I did know the race of and found that the race of the superdelegate was NOT a statistically significant predictor of who a superdelegate chose to support.

3) Does your methodology take into account the fact that a large number of pledged superdelegates made their commitments (primarily) to Clinton prior to anyone believing that there would be a viable alternative candidate?

We use a two-stage model that allows us to account first for the factors affecting whether a superdelegate has endorsed at all and then estimates which candidate that superdelegate endorsed (if he/she has endorsed). To some extent, this should pick up some of the fact that potential Clinton endorsers were more likely to have already endorsed while more potential Obama supporters may be waiting to make sure he is going to actually pull out the nomination. Nevertheless, in recent works I have begun considering a different approach, by attempting to model when superdelegates made their decisions (for example, before or after Super Tuesday) and include that as a factor in the model. I may do that in the next iteration of the predictions.

Thanks for the questions, and let me know if you have more.

Early Voting for North Carolina Presidential Primary

The North Carolina State Board of Elections does something very cool--they put all of their early voting data up on their website for easy download. So, of course I just had to download the data this morning and do a little crunching to see what types of people are voting early for the May 6th primary.

Based on the data I downloaded this morning, 144,440 voters have cast valid early ballots so far in North Carolina. Of that number, 117,655 have requested a Democratic ballot for the primary (only 26,371 have requested a Republican ballot, with the rest taking an unaffiliated ballot). Based on party registration figures included in the data, 68.6% of early voters are registered Democrats, 16.1% are registered Republicans, and 15.3% are registered as unaffiliated. 84% of the early voters who are registered as unaffiliated voted in the Democratic party. These voters make up 15.8% of those who have already voted in the Democratic primary with Democrats making up the rest.

The great thing about this data is that it includes the race and gender of those who have voted early, so we can take a look at a few different demographics. The table below presents the racial and gender makeup of early voters in North Carolina so far:

	% of Early Voters
White	59.10%
Black	37.20%
Male	40.40%
Female	59.10%
White Men	25.20%
White Women	33.70%

It certainly bodes well for Barack Obama that over one-third of the early voters are African American, since this group has supported him at such a high rate. White women, on the other hand, have been a very strong group for Hillary Clinton, and they also make up one-third of the early voters in the state.

I'll present some more information from these data as the week goes on.

Updated Democratic Superdelegate Predictions

UPDATED AT 9:45PM ON APRIL 26TH

I have now had a change to update the superdelegate predictions. As always, I use information about the superdelegates who have committed to a candidate to generate predictions for the remaining unpledged superdelegates. I exclude superdelegates from DC and the territories because we lack complete data from those areas, and from IL, NY, and AR because superdelegates in those states have nearly unanimously cast their support for their native son/daughter. As always, information on the superdelegates is provided by the Democratic Convention Watch site. You can find more about the methodology I use here. Since I began generating these predictions, 106 superdelegates have announced their endorsements, and we have been correct on 74 of these. Thus, overall, the models have been correct 70% of the time.

In addition to adding the newest endorsements to the model, I have also added information about the add-on superdelegates who have been designated at this point. We are unable to predict how most add-ons will vote since we don't know who they are yet. But I added those who have been selected to the model (most add-ons have already endorsed one of the candidates).

In the figure below, I present the distribution of unpledged superdelegates based on the probability of supporting Clinton:Superdelegates who are between 40% and 60% likely to vote for Clinton/Obama are labeled as "unclear." There are 77 superdelegates in this range. There are 129 unpledged superdelegates who are at least 60% likely to vote for Obama; just 38 unpledged superdelegates are at least 60% likely to vote for Clinton. These predictions suggest that unless something dramatically changes, Obama will be able to cut into and even overtake Clinton's superdelegate lead in the coming weeks and months.

The estimates for each unpledged superdelegate are listed here. These estimates show that among Obama's most likely endorsers are Rep. Dennis Moore (KS), Rep. Michael Michaud (ME), and Rep. Tom Allen (ME). Clinton's most likely endorsers include Reps. Jerry McNerney, Susan Davis, and Lois Capps (all from CA).

UPDATE: I updated this information to reflect that fact that my original data had Patrick Lynch as undeclared (when he has, in fact, endorsed Obama).

Michigan and Hillary Clinton's Popular Vote Claim

There has been much discussion since the Pennsylvania primary of the Clinton campaign's claim that she is now ahead in the popular vote. Of course, this only happens when you figure in Michigan and Florida into the count, and only when you do it in a very specific way.

According to MSNBC, the popular vote with every state included except Michigan (i.e. every state where Obama was at least on the ballot) breaks down like this:

Obama: 15,016,607
Clinton: 14,822,400

If you include all of Clinton's Michigan vote and nothing in Michigan for Obama (who was not on the ballot there), then you get a Clinton lead (this is the metric being promoted by the Clinton campaign):

Clinton: 15,150,551
Obama: 15,016,607

Of course, most people are looking at this metric with a great deal of skepticism. Of the 593,837 Democrats who turned out to vote in the Michigan primary, 55% (328,151) cast their vote for Clinton. But what would have happened if all the candidates' names had been on the ballot? Fortunately, we have exit polls from Michigan which can give us some insight here. On the exit poll survey, voters were asked who they would have voted for had every candidate's name actually been on the ballot. Here are the results:

Clinton: 46%
Obama: 35%
Edwards: 12%

So, what happens if re-allocate the Michigan vote accordingly? In Michigan, the vote would have broken down as follows:

Clinton: 273,165 votes
Obama: 207,843 votes
Edwards: 71,260 votes

Thus, had Obama's name been on the ballot, Clinton's margin in the state would have been much smaller. Of course, there is no really good metric for measuring the vote in Michigan. Even in this scenario we have to assume that turnout wasn't suppressed by the fact that Obama's name wasn't on the ballot. Yet, you can imagine that many Obama supporters (and some Clinton supporters) may not have bothered to turn out to vote given that they knew that their votes were not likely to count. Nevertheless, this metric probably comes closest to capturing the actual preferences of those who did turn out to vote in Michigan.

So, how big a difference does this make in calculating the popular vote nationwide? If you add in the Michigan vote using the reallocation based on the exit polls, then you get the following national count:

Obama: 15,224,450 votes
Clinton: 15,095,565 votes

That would give Obama a lead of more than 120,000 votes nationwide, a lead that would be difficult for Clinton to overcome in the remaining states.

Post from Public Policy Polling on their Pennsylvania Poll

The folks at Public Policy Polling posted last night to discuss why their survey showing Obama beating Clinton in Pennsylvania was so off the mark. They note that their over-representation of voters in the 18-45 age group was a big factor in throwing off their results, something I noted before the primary.