Several people have asked about using data to make predictions so before the Tuesday that could decide it all here is a “Make Your Own Prediction” guide.
If you have ever sat in on a campaign headquarters on election night, this is exactly what they do. Most candidates know if they have won or lost before the networks “call” the election because they know which precincts they have to win and by what margin and what the margin needs to be in the precincts they will probably lose. In more sophisticated campaigns they do their own polling right along with the networks.
For Tuesday let’s focus on the two states everyone agrees Hillary Clinton must win to stay viable: Texas and Ohio. The conventional wisdom has shifted a bit in the last week, especially as the Clinton people have tried to spin it. Several weeks ago, the pundits predicted that Clinton would not only need to win, but win big in order to rake in enough delegates to over come Obama’s lead. This past week the Clinton people have been saying that all she needs to do is win because that would demonstrate her viability.
Right now the Clinton campaign is playing very coy with the numbers game, refusing to say what they would consider a viable margin of victory. That could spell trouble. If you hear this from the Clinton people on Tuesday night, expect the worst, for it will mean that Clinton has decided to extend the fight. That will set off a battle for the superdelegates along with whether the delegates Clinton won in Florida and Michigan should count. This will not be good news for the Democratic Party.
I am attaching below a spreadsheet you can actually print out and fill in as the night goes on, sort of like a political box score. Actually, you will need to print the entire page, but you can use the rest to put your beer and pizza on. Tonight we’ll do Texas, tomorrow Ohio.
TEXAS
To begin with the Democratic Texas primary is a weird creature, part primary and part caucus. Texas gets 228 delegates for the national convention: 126 will be chosen by the voting primary, 67 will come from caucuses held after the polls close, and 35 are superdelegates. In essence you get to vote twice. This dual system favors the committed and those with stamina. Imagine standing in line to vote and then going to a caucus later that evening–a caucus that, if our experience in Minnesota is any example, could last well into the night. Since these are only precinct caucuses the delegates elected there get to move on to the senate conventions and then the state convention. You really will not know who wins those delegates until the state convention.
So for purposes of election night we will focus on the primary, not the caucuses. Because of the dual system, the Texas vote will close at 7:00 p.m. That could have an impact, because people at work will either have to take time off or vote when they get home, which could clog up the polls. It will also mean we need to be even more suspicious of early data than usual, because the vote–and the data–most likely will come in one huge surge.
The latest polling data for Texas show a dead heat.
Source: RealClearPolitics
The poll is a composite of all the major polls. The purple line is Hillary Clinton; the green Barack Obama. Nothing here we didn’t already know–Clinton is in a dive and Obama is peaking. However, note two things: Clinton appears to have bottomed-out, while Obama’s upward surge has actually dropped in recent days, with a slight up-tick in the last two days. Given that Clinton’s line has stayed relatively flat, that uptick can only mean one thing–the undecideds are making up their minds. The total for each candidate is 91%, which means almost 10% of Texas is still making up its mind.
Turnout
That gives us our first three pieces of data to watch. First, we will want to know if the turnout is heavy. That usually means more than the Party faithful are voting. In this election and in Texas, the Party faithful lean Clinton, so if the turnout is heavy that should help Obama. The next two pieces of data help further nail that. We will want to know how many Independents are voting. These have been Obama’s bread and butter throughout the campaign. If they make up a large percentage of voters, lean towards Obama.
The other notable thing about Texas is that it is an open primary, so Republican cross-over voters could play a role. The GOP cross-over in Wisconsin went Obama and the GOP’s dislike for Clinton is long-standing, however if the GOP really wants to cause trouble for the Democrats they would vote for Clinton because it would extend the nominating process. I tried to find some preliminary data on this one but could not find anything reliable. So right now it is a bit of a wild card.
Now we get into the demographic data that the networks typically ignore when they are announcing their exit preliminary exit poll data in favor of answers to questions like, did the color of Obama’s tie impact your vote? Here there are several crucial factors to watch.
We all know the Latino turnout will be critical for Clinton. In California they made up almost 1/3 of those voting. An average Latino turnout would range from the high teens to low 20s. If it goes over 30%, Clinton should be in good shape. All this assumes she will retain the loyalty of Latinos, which has been slipping. That makes a high turnout even more important.
The second key voting group will be African Americans. Obama will need a high turnout here to offset Clinton’s expected margin with Latinos. Census data show African Americans make up 11.9% of voting Texans. If the turnout among black voters exceeds this, it should be a good sign for Obama. He will need to win over 90% of them to help his cause.
Young voters have always gone for Obama. Census 2000 data showed 22% of potential Texas voters fall into the 20-34 year-old age group. This group historically under votes, so if they turn out in numbers close to their population percentage it is another plus for Obama. Watch particularly the votes from college precincts.
In contrast, the one group that has stayed with Hillary Clinton is older voters, especially voters over 60. In Texas 11.9% fall into that cohort with another 4.3% over 55. If thee voters turn out it will be a good sign for Clinton.
The most volatile group has been women. In the beginning of the campaign they were Clinton voters, but lately that has not been true. In his winning streak Obama has won a majority of female voters in just about every demographic category except women over 55. Texas is almost evenly split between men and women–50.4% of all Texans are female.
With groups like EMILY’s List putting a big effort into boosting Clinton, I’m going to go out on a limb and say a larger turnout of women should favor her because they know if she loses the chance of having a woman as President is lost. If the percentage of women gets into the mid-fifty percent range or higher, it should be a good sign for Clinton.
I have labeled the final column nonvoters, but what I really mean are the less-educated and those living in poverty. These voters need to turn out if the Democrats are going to win in November. In Texas 17% of eligible voters make under $15,000. According to CNN data, 11% of people making under $15,000 voted nationally in 1996 and were the income level that gave Clinton his highest percentage 59%. Thirty percent of the voters in 1996 had a high school education or less and also gave Clinton 59%. In Texas almost 39% of eligible voters have a high school degree or less.
Without either of these groups, Clinton might have been in trouble. So as you watch Texas, remember these two groups. They tend to have low turnouts in primaries, so if half of them vote that would be a good sign for November.
So there is your scorecard for Tuesday. Have fun. I will be writing away Tuesday night right along with you so feel free to stop in and toss in your own views and maybe all of us can add our own predictions to the mix.
See you then.
CHART COURTESY OF THE STRANGE DEATH OF LIBERAL AMERICA
Print Posted by: liberalamerican
