Thursday, February 28, 2008

Wednesday, February 27, 2008

Quinnipiac PA poll

Do we really have to start talking about PA? I think/hope it will be over before then if Barack has a convincing win in Texas, and narrow win or loss in Ohio.

PA: Clinton 49, Obama 43

Tuesday, February 26, 2008

SUSA Ohio poll

Clinton 50%, Obama 44%

Note Clinton's huge advantage with earlier voters. Obama will have to work hard to overcome this with his GOTV operation.

New PPP Texas Poll

TX: Clinton: 48%, Obama: 48%

Rasmussen Ohio Poll

So according to Rasmussen Reports, Obama is now only down 5%
Clinton: 48%, Obama: 43%

Monday, February 25, 2008

Saturday, February 23, 2008

Rhode Island and Vermont polls

I don't believe anything from American Research Group, as they've been proven wrong so many times. I wonder why they are still even doing polls... But I suppose bad data is better than no data (at least we can decide what to ignore).

Rhode Island: Clinton 52%, Obama 40%

Vermont: Clinton: 34%, Obama 60%

Friday, February 22, 2008

Hard-to believe Texas and Ohio polls show Obama with significant leads in TX and OH

They are currently hard to believe because they don't agree with other polls so far. Recall that so far we've seen Obama about even in Texas and about 7-9 % down in Ohio. I'm guessing there must be some major sampling bias in these polls from "Decision Analyst"; or they are the only ones (so far) predicting a Wisconsin-like result for Obama.

TX: Obama 57%, Clinton: 43%
OH: Obama 54%, Clinton 46%

Here's to hoping!

Great texas projection

Looks like it will probably be tied in terms of delegates, though for a treat, see the best case Obama scenario

Words our next president should listen to

Not his famous "military industrial complex speech" that everybody sees as such an important warning signal today, but certainly in the same direction:

Full speech text:

These are words that our next president needs to understand. However, the next president also needs to understand that this is no longer "the best way of life to be found on the road the world has been taking", there is both a better way, and a better road.

Thanks Thom Hartman for playing this clip on your show!


Thursday, February 21, 2008

ABC/WP Texas & Ohio poll

If this holds up, Clinton would only trim the delegate lead by 10-15, and I'm guessing that a good showing by Obama in Vermont and Rhode Island would probably compensate for that.

Ohio: Clinton 50%, Obama 43%; Texas: Clinton: 48% to Obama: 47%

Newest Texas Polls, show Obama within striking distance

Constituent Dynamics: Clinton: 46%, Obama: 45%
IVR: Clinton 50%, Obama 45%

Wednesday, February 20, 2008

Analysis of Obama's performance vs. polls: Black voters

Recently, Barack Obama has been outperforming polls. The goal of this analysis is to look at how exactly how he is doing this. I will look at pre-election polls that have crosstabs (usually SurveyUSA) with race included vs. actual election exit polls courtesy of CNN. I am going to look at each group independently in different posts, and I am going to separate the analysis into three categories: Pre super Tuesday elections, Super Tuesday elections, and post Super Tuesday elections. I am picking a total of 5 states for this analysis:
  • Pre Super Tuesday: South Carolina
  • Super Tuesday: California, New Jersey, Missouri
  • Post-Super Tuesday: Wisconsin
I chose these because they represent a pretty good cross-section of the electorate. I did not analyze any other states, and perhaps the trends could be less or more significant if I had taken a more detailed look at other states. In this post, I'll look exclusively at Black voters.

Black voters
This is supposed to be one of Barack Obama's key constituencies, and it has been through the election so far. The question we should ask here, however is whether his support has solidified with Black voters over the course of the election.

Pre Super Tuesday election: South Carolina
In this pre Super Tuesday marquee matchup, Barack outperformed pre-election polling very slightly by 4%.
  • In pre-election polling by SurveyUSA, Barack had a 73% to 18% edge over Hillary amongst likely black voters, for a gap of 55%
  • In post-election exit polls reported by CNN, Barack won the Black vote 78% to 19% over Hillary, for a gap of 59%
Super Tuesday election: New Jersey
In this 2/5 matchup, Barack significantly outperformed pre-election polling by 14%
  • In pre-election polling by SurveyUSA, Barack had a 76% to 22% edge over Hillary amongst likely Black voters, for a gap of 54%
  • In post-election exit polls reported by CNN, Barack won the Black vote 82% to 14% over Hillary, for a gap of 68%
Super Tuesday election: Missouri
In this Super Tuesday matchup, Barack significantly outperformed pre-election polling by 17%
  • In pre-election polling by SurveyUSA, Barack had a 74% to 22% edge of Hillary with likely Black voters for a gap of 52%
  • In post election polls reported by CNN, Barack won the Black vote 84% to 15%, for a gap of 69%
Super Tuesday election: California
In this Super Tuesday matchup, Barack significantly outperformed pre-election polling by 13%
  • In pre-election polling by SurveyUSA, Barack had a 72% to 25% edge of Hillary with likely Black voters for a gap of 47%
  • In post election polls reported by CNN, Barack won the Black vote 78% to 18%, for a gap of 60%
Post Super Tuesday election: Wisconsin
In this post Super Tuesday matchup, Barack significantly outperformed pre-election polling by 13%
  • In pre-election polling by PPP, Barack had a 72% to 25% edge of Hillary with likely Black voters for a gap of 47%
  • In post-election exit polls reported by CNN, Barack won the Black vote 78% to 18%, for a gap of 60%
Here's a chart showing the change over time

Latino voters, and White voters will be covered in later blog posts.

Tuesday, February 19, 2008

An actual Hawaii poll

Shows Obama with a 57.6 to 27.4 lead. Not sure how much value to put in this... I've never heard of the poll before.

Is Hillary's Hispanic Support Dropping?

Great report: Nationally, Obama: 50%, Clinton: 46%

Also, data from recent exit polls (courtesy of
  • VA: Obama: 54% Clinton: 46%
  • MD: Obama: 45% Clinton: 55%
Compare that to Feb 5th exit polls
  • NM: Obama: 36% Clinton: 62%
  • CA: Obama: 32% Clinton: 67%
An East:West Hispanic divide is also a possible reason for this, however.

New Poll numbers, 2/19/08

Primary matchups

General election matchups
Note, here is SurveyUSAs report card on the pollsters

Thursday, February 14, 2008

Texas momentum

How much Money has Obama raised in 2008?

That's a good question, but it's one that outsiders to the Obama campaign can't answer right now, because the Obama camp has stopped publishing the fundraising amount, and instead just listed the number of donations they've received so far this year. Not advertising the actual amount of money raised was probably a good move for the Obama team, as this just sparked the Clinton team to rally around "Look how much Obama's getting in donations... we need to match/beat that". Since the Clinton campaign will never be able to match the number of individual donations the Obama campaign has received, the number of people that have donated is probably a much better number to post.

However my brother asked me to do a projection based upon the number of donations this year, and I think that's a good idea so we can get a rough estimate. According to Time, Obama raised over $100 Million last year from about 470,000 unique donors. According to this blog Obama has an average donation size of about $250. If we do the math, it comes out to $117.5 million, so I suspect the actual average donation was smaller: something like $215 per donor.

According to, nearly 437,000 people have donated so far this year, which is nearly as many has have donated last year, and we're only in the middle of the first quarter. According to the figures from the Obama campaign, they raised $32 million in the month of January from 250,000 donors, which is an average donation of $128 per donor. I would guess that donation amounts have increased since then, just based upon repeat donations due to Obama's success in the recent primaries. However, I don't have any hard data on this, so I'll just provide some rough estimates given a few scenarios:
  • Low estimate: Average donations stayed at $128/donor: ~$56 million
  • Medium estimate: Average donations went to $175/donor: ~$76.5 million
  • High estimate: Average donations increased to $215/donor: ~94 million
As a point of reference, I donated about $200 this year, and my brother donated around $400 total (probably most of which happened this year). Given this highly representative sample, let's assume a Very high estimated average of $300/person:
  • Very high estimate: Average donations increased to $300/donor: $131 million
Hillary Clinton's campaign is advertising that they raised $13 million since Feb 5th, I bet when the Q1 fundraising figures come out, we'll find that Obama either doubled or tripled Clinton's figure. Clinton's January fundraising figure was $13.5 million -- so assuming a modest 50% increase from that based upon her new underdog status, let's assume she raises $20.25 million per quarter, that's $60.75 million for Q1 2008. If we ignore the numbers above, and conservatively assume Obama's $32 million per month holds up, he'll have $96 million from the first quarter, compared to Hillary's $60.75 million. Using the estimates above, given that we are 50^ through the quarter, we can try doubling those figures:
  • Low Q1 estimate: Average donations stayed at $128/donor: ~$106 million
  • Medium Q1 estimate: Average donations went to $175/donor: ~$150million
  • High Q1 estimate: Average donations increased to $215/donor: ~$188 million
Even using the low estimate, assuming things stay on pace, Obama would raise nearly as much in the quarter as he raised all of last year.

Now the lines have *really* crossed

Rasmussen shows Obama leading nationally by 12%

Initial Ohio Polls

However, most of the polling was done before the MD, VA, DC primaries.
So clearly, Obama has some work to do in Ohio. I expect these numbers to change quite a bit in the coming weeks, and I would guess that by March 4th, we'll see Obama within 5-8% of Clinton.

Wednesday, February 13, 2008

The lines have crossed

Obama now leads in popular vote including FL


Even counting FL, Obama leads

Interesting tally today on First Read, noting that even counting the votes of Florida and Michigan, Barack Obama is leading Clinton in popular votes. And even with Clinton's (hypothetical) net Florida advantage of 38 pledged delegates, Obama would still be ahead of Clinton in delegates. Here's the overall vote:

Total Vote %
Obama 9,373,334 50%
Clinton 8,674,779 46%
Others 726,095 4%

With Florida
Total Vote %
Obama 9,942,375 49%
Clinton 9,531,987 46%
Others 984,236 4%

With Florida and Michigan
Total Vote %
Obama 9,942,375 47%
Clinton 9,860,138 47%
Others 1,249,922 6%

Tuesday, February 12, 2008

Barack leads in NC, Clinton leads in OH

This is a Survey USA poll, and these have been pretty good so far. Still a significant time before either of these primaries happen. One thing of note though is that Ohio has early voting, and I don't think North Carolina does.

It remains to be seen what the impact of the races today and next week will be. Obama has very good shot of winning 10 primaries/caucuses in a row, so it's hard to imagine a case where this wouldn't cause favorable movement in his direction.

Great poll of potomac primary

I would love it if all polls gave this level of detail and projection. It has voting breakdowns by district, and delegate projections per district. It's really awesome.

Obama with a 11% lead in WI

White vote (84% of the result): Obama: 49%, Clinton: 41%
Black vote (12% of the result): Obama: 62%, Clinton: 24%

Final Obama MD & VA Polling Numbers

VA: Average of polls: Obama + 17%: Obama 55.0% Clinton: 37.3%
MD: Average of polls: Obama + 22%: Obama 55.0% Clinton 32.7%

Looks to be a very good night for him tonight based upon this.

Monday, February 11, 2008

Can Missouri tell us anything about Ohio?

Right now there is no polling data on the democratic primary in Ohio that is set to happen on March 4th, 2008, so I'm hoping that we can use other factors to help us determine what the outcome will be. My brother tells me that Ohio will go for Obama because he has a lot of natural constituencies, but I want to do at least a little bit of analysis to see if we can get something more than a gut reaction here. I was looking at demographics on Wikipedia and some past election data on uselectionatlas, and I have come to the conclusion that Missouri could be a bellweather in predicting how Ohio is going to vote. I'll detail the reasons below.

According to wikipedia, Missouri and Ohio have very, very similar demographics. The table below illustrates this:

Demographics in Missouri vs. Ohio
(includes ~2% hispanic)

So the question is: Does similar demographics in the two states mean that voters will vote the same way in the primary election? I don't think this can be answered right now, especially given Obama's broad base of support in various parts of the country. However, we can look at how the two states voted in the past and try to use this as a predictor.

Missouri and Ohio have voted the same way in the presidential elections since 1960, with similar margins

Presidential elections
So in every election after Kennedy vs. Nixon, both Missouri and Ohio voted for the same candidate, who was also the winner. Interestingly, the margins were actually very similar, as the table below illustrates.

Matchups in General elections 1964 - 2004

1964: Johnson (J) vs. Goldwater (G)
J: 61.05%
J: 64.05%
J: 62.94%
1968: Nixon (N) vs. Humphrey (H)
N: 43.42%
N: 44.87%
N: 45.23%
1972: Nixon (N) vs. McGovern (M)
N: 60.67%
N: 62.29%
N: 59.23%
1976: Carter (C) vs. Ford (F)
C: 50.08%
C: 51.10%
C: 48.92%
1980: Reagan (R) vs. Carter (C)
R: 50.75%
R: 51.16%
R: 51.51%
1984: Reagan (R) vs. Mondale
R: 58.77%
R: 60.02%
R: 58.90%
1988: Bush (B) vs. Dukakis (D)
B: 53.37%
B: 51.83%
B: 55.00%
1992: Clinton (C) vs. Bush (B)
C: 43.01%
C: 44.07%
C: 42.91%
1996: Clinton (C) vs. Dole (D)
C: 49.23%
C: 47.54%
C: 47.38%
2000: Bush (B) vs. Gore (G)
*B: 47.87%
B: 50.42
B: 49.97%
2004: Bush (B) vs. Kerry (K)
B: 50.73%
B: 53.30%
B: 50.81%
*Note: in 2000, Bush won the election with less of the popular vote than his opponent, Al Gore. However, Bush won the popular vote in both Ohio and Missouri.

Analysis: Really, I want to answer the following questions:
1) Does Missouri vote like Ohio in the general election
2) Does Missouri vote like the US in the general election
3) Does Ohio vote like the US in the general election

In order to tests these things, I ran 3 independent paired two tailed t-tests with an alpha of .05 with the following results:
1) Missouri vs. Ohio: p=.1952, No significant difference. Average margin of absolute difference 1.41%
2) Missouri vs. US: p=.0442, Missouri votes significantly different than the US. Average margin of absolute difference = 1.65%
3) Ohio vs. US: p=.4311, No significant difference. Average margin of absolute difference 1.17%

So the conclusion from this is that Missouri doesn't vote differently from Ohio significantly in presidential elections, but Missouri does vote significantly different than the US. Ohio doesn't vote significantly different than the US in presidential elections (actual this result showed that Ohio is more likely to closely mirror the presidential outcome than Missouri is likely to mirror Ohio's result).

Primary elections
It's really difficult to use previous primaries to help predict this race, simply because the previous contests have happened at different points in the election where the race might have already been decided, and therefore the voting would have been different if we had a situation like we have here in 2008. So until I can figure out a method to make these results meaningful, I'm just going to leave them out.

Using demographic data and absolute margin of difference to predict Ohio
Model #1: Using Missouri to model Ohio
If we just make the assumption that the electorates are the same between the two states, and will vote the same in Ohio as they did in Missouri with a margin of error of the absolute difference +- 1.41%, then it's very easy to predict the outcome:
  • Missouri 2008 primary: Barack Obama 49.40% vs. Hillary Clinton 48.18%
  • Model #1 Ohio 2008 primary projection: Barack Obama 50.81% to 47.99% vs. Hillary Clinton 49.59% to 46.77%
Model #2: Using US election data so far to model Ohio
We've already determined that using the actual vote percentages from the entire US vote have historically been a better predictor of the Ohio vote than using Missouri for this. Given this, the second model just uses the current vote percentages in the states that have voted so far and the absolute historical difference between the US presedential vote and the Ohio presidential vote of 1.17%.
  • US vote percentages so far: I don't have a good source for this, but I think I saw on TV that Obama has a lead of about 250,000 votes out of 16 million cast, or 1.56% total. So just to plug in some numbers, let's say Obama is at 48.56%, and Clinton is at 47.00%
  • Model #2 Ohio 2008 primary projection: Obama: 49.73% to 47.39% vs/ Clinton: 48.17% to 45.83%
Model #3: Using demographic data we've seen so far to predict the outcome in Ohio
I question the accuracy of this kind of approach without fully understanding the demographics. We've seen Obama appeal to the whitest of white voters (Maine), so I can't begin to understand how we could use this as predictive. Revisit this

This race looks to be a very tight one with a slight edge to Obama based upon this analysis. Even without doing this analysis I doubt very much that either candidate will blow the other one out here, which is bad news for both candidates, but perhaps worse news for Clinton. Missouri turned out to be not as great as I had hoped in being a predictor, as the US election seems to be superior, but it is better than nothing. One limitation is that we don't know how much past voting behavior in presidential elections can predict voting in primary elections, but since Ohio has an open primary, I suspect they will be similar.

In a follow up post, I'll try to use model #1 (Difference from Missouri) and model #2 (Difference from US) and see how well it fares against contests that have already been decided

Obama won every county in Washington

This is very impressive:

Looking good for dems to gain in congress in 08

Obama with huge leads in MD and VA

He leads by 22% in Virginia:

...and 23% in Maryland (my home state, yay!):

Given that where huge leads were observed, Obama typically outperformed his expectations, I would not be surprised to see a 25% margin in each, and a 40-50% margin in DC.

The lamest campaign ad ever!

Can you say pandering to the youth vote? And not just pandering, doing a terrible job in pandering. It's not like this was a group of people that just decided to get together and make this, Hillary actually paid for this.


National polls, and Poblano's multiple regression analysis

After super tuesday, Clinton seemed to surge slightly in national polls, perhaps reflecting the surge of support she got in some of the larger states. Now that super tuesday is behind us, we see the lead closing again. Gallup has only a 2% gap between Clinton (46%) and Obama (44%):

Rasmussen has this at 5%:

Of course, what really matters are the state polls, and we have no data for important states such as Ohio or Texas, but Poblano came up with an imperfect, but pretty good regression model that shows a 10% Obama victory in Ohio, and a 10% Clinton victory in Texas.


Saturday, February 9, 2008

Washington: 5% lead for Obama among registered democrats, 30% lead amongst those likely to caucus

This shows Obama's power when it comes to caucuses. His supporters are just much more dedicated on the average, and willing to caucus.


The actual caucus is today, so we'll see how accurate these are.

Friday, February 8, 2008

Maryland poll gives obama a 19% edge over Clinton

I think his margin of victory will be quite a bit higher than this:

Current data suggests obama is the best candidate against McCain

Recent polling suggests that Barack Obama is the candidate who is best equpipped to beat John McCain in November.

This shows the results of several recent polls that compare matchups of Clinton vs. McCain and Obama vs. McCain

As you can see, Obama is favored to win the nomination in most polls, and Hillary is projected to lose. Actually there is almost a 6% absolute average difference between the margin of victory from these polls as the chart below indicates:

This is a strong argument that I hope the Obama campaign makes repeatedly. Saying you believe in polls doesn't play well with the American people, but there's a clever way to frame this without even mentioning a poll.

New polls out show Obama with a double-digit lead in Virginia

Looks very promising in what I think is the toughest state on 2/12 for him.

SurveyUSA (O: 59 C:39)
InsiderAdvantage (O:52, C:37)

As I mentioned previously Obama will do very well in the next several states, and he could emerge before the Ohio and Texas primaries with a 100-150 delegate lead.

Washington govenor endorses obama

Thursday, February 7, 2008

Both Obama and Clinton abstain from economic stimulus package


I know they were in the chamber today, so they *could* have voted on this if they wanted to, so I wonder why they didn't. I suppose it's because they believe the plan they proposed was better? Sounds to me like Clinton is voting present on this one, along with Obama.

What's funny to me is that she earlier in the day railed against McCain for missing the vote, when she missed the vote as well. This is hillary putting her foot in her mouth for sure.

The actual vote counts

From the article:

"Mrs. Clinton: 7,427,700, or 48.83 percent;

Mr. Obama: 7,369,798, or 48.45 percent;

John Edwards: 411,740, or 2.70 percent."

So, a 57,911 vote difference between Obama and Clinton with over 15 million votes counted. This does not include all of the votes yet, and it doesn't include Alaska. It's really amazing how close this race is!

Delegates: Explained

From dailykos, a very good summary of what delegates are, and how superdelegates play into things.

What the race will look like on Feb. 13th, 2008

So right now there's a dearth of polling data for the upcoming contests, and previous data this political season has been spotty at the very best. Since there's not much data now, and the data we have has sucked (especially for somebody whom I'll never trust again: John Zogby), I might as well just make some up. Given this, I'm making up my own projections for the coming up race.

So below find my projections for the Feb 9th to Feb 12th races. Note these aren't based upon much data... all we have recently is the Survey USA poll out of Washington from Feb 4th 2008 showing Obama with a 13% lead over Clinton (53% to 40%, with 7% undecided). Since Survey USA has actually been one of the more reliable polls this time around, I'm going to trust them, and assume a 55% to 43% final margin in Washington. For the rest of the data I simply am using some intuition, and general knowledge about the electorate and knowledge of how the past contests have gone.

The chart below shows projected vote percentages in each state. Again, this isn't based upon much data yet, though I do plan to revisit it when more data is available. Instead, it's based upon mainly understanding the demographics of the areas.

Rationales for each state:
  • Washington: Latest poll by Survey USA showed Obama with a 13% lead over Clinton.
  • Louisiana: According to Wikipedia, the Black population Louisiana is somewhere around 34%. So this is even more than Georgia, and Georgia went for Obama 67% to 31% for Clinton. I was conservative in giving Obama 60% of the vote. In actuality, this may turn out to be a huge win, perhaps with Obama winning 65-70%.
  • Nebraska: I don't have a good sense of this state, but I think we can assume it will be like the other caucus states and go for obama heavily. I am just assuming it will be like nearby caucus states such as Kansas and North Dakota, where Obama won 74% and 61% of the vote respectively.
  • Maine: I have no idea here, but I suspect it will go to Obama. But just to be conservative, I gave Clinton the benefit of the doubt here with a 49% to 48% victory
  • Virginia: I would guess VA would go for Obama also by a thin margin given the number of college towns and the large Black population. However, the black population is not nearly as large (about 20% of the electorate) as states like Maryland, Georgia, Louisiana (all around or above 30%), so I'm just giving the benefit of the doubt to Clinton again with a 52% to 47% victory. Note: Updated to a 59% to 40% victory for Obama based upon newer polling data.
  • Maryland: Should be pretty solid for Obama given that it has very similar demographics to Georgia, and perhaps a more progressive electorate in general than some of the southern states.
  • DC: Very solid for Obama. I'm giving him an 85% to 15% victory in DC. This is because 60% of the population of DC is Black, and Obama will probably win 85% of these votes. I might have overstated the margins here, but I doubt by very much.

The below chart is a *very* naive approach to calculating the number of delegates coming out of these states. I just took the number of available pledged delegates and multiplied it by the proportion of the vote that the candidate received. Very simplistic, but I don't have the interest in analyzing each voting district.

This shows the number of delegates won overall when all is said and done in these states.

This is a stacked graph showing delegates already pledged, projected from these upcoming primaries/caucuses, and superdelegates. A note about superdelegates though: I think including these is very suspect by the TV networks, as these delegates can change their minds whenever they want. Also, as far as I know (correct me if I'm wrong) superdelegates votes will not even count if the race is decided before the convention. So even though I think it's misleading, I'm including it for the sake of no information hiding.

Below, I included the actual raw data I used.

More to come later.

Thanks a lot to Poblano from daily kos ( for the inspiring me to work on this.


Blog Archive