If the protracted Democratic primary process was good for anyone besides John McCain, it was good for NBC Political Director Chuck Todd.
His understanding of the intricate calculus of Democratic delegate math, exit polling, and prognosticating made him a rockstar of cable news; his ability to take the complex and data-rich and make it consumable vaulted him to cult-figure status. On the Internet, the website Chuck Todd Facts takes a page from the Chuck Norris fanbase’s book, declaring such truths as “Contrary to claims by the Clinton campaign, Chuck Todd leads in the popular vote.” The blog “Viva Chuck Todd” mourned the end of the primary season wondering when his loyal fans would see him return to their television screens with analysis and commentary.
“We know Chuck will grace the sets of Hardball, Countdown, The Today Show and Meet The Press. But with what frequency? Who knows? The man certainly does need a vacation. But what about us? His loyal fans, the Chuckolytes if you will.”
Fear not, Chuckolytes. The next six months will be full of more data, polling, and electoral math than you can possibly imagine.
Now that Barack Obama has crossed the magic threshold of delegates to clinch the Democratic nomination (barring any Clintonian use of the dark arts), the amount of general election ballot matchup data available will take off. Dozens of pollsters will tout numbers declaring that one candidate is ahead of the other. The horse race will continue unceasingly, and without mercy.
Additionally, the rise of the paint-by-numbers map of the United States will begin in full force. Already, pollsters and pundits have begun predicting the relative “blue-ness” or “red-ness” of states. These maps, using varying algorithms and methods that blend polling, historical trends, and demographic information all purport to show how the candidates are doing in the tally that really counts – the electoral college.
Finally, those exit polls aren’t going anywhere. Given the protracted nature of the Democratic primary, we as number crunchers now have a lot of information about who is voting and what they care about, at least on the Democratic side. The slicing and dicing of the electorate based on the primary season’s exit polling will no doubt fill in the gaps between discussions of whether or not Michigan should be considered “indigo” or “periwinkle”.
Given all this, there are a few things the typical observer of politics needs to know that don’t take Chuck Todd to explain.
The national popular vote total doesn’t matter, and only electoral votes decide the Presidency. So why even bother with the national ballot match-ups?
True, the national popular vote total doesn’t decide the Presidency. However, this doesn’t make these polls useless. On the contrary, these ballot tests can do a lot to tell the story of the election. Has either candidate made gains over the last two weeks? Who was the momentum? What demographic groups are shifting significantly?
If you look at the poll as a snapshot in time, it’s not telling you much except what a hypothetical popular vote total (not the electoral college total!) would look like if a hypothetical election were hypothetically held today. If you look at it as a way of “predicting” anything, you’re taking it for more than it is. (In case you haven’t noticed, it’s a really long way until November.)
But if you look at the trends over time, watch the ebb and flow of support for candidates and evaluate what voters are changing or making up their minds, you can learn a lot. You can’t have trends over time if you don’t get a baseline, so polling this far out is necessary, if only to give context to the numbers that will come closer to election day.
Beyond simply gauging momentum, if someone is winning the ballot test by more than a slight margin, you should feel pretty comfortable saying that person would win a general election even with the electoral college. The possibility that someone can be up in the popular vote/ballot test by a margin of something like 53-47 and not also win the electoral college is negligible.
What about the electoral college maps and elector counts? How useful are they at this point?
Again, the predictive capability of anything this far out is negated by the simple unpredictability of politics. Bear in mind that between early June and early November, anything can happen. You never know when (another) video of your pastor saying crazy things is going to leak onto the Internet, or when Hillary Clinton is going to get teary eyed, or when you’re going to get Swift Boated.
Despite the limited “Crystal Ball” factor these polls may have, they also can help candidates and observers know where the major battles of the campaign will be fought. They stir up debate over whether or not, say, California is within reach of the McCain campaign or if Obama really can snatch Georgia away from the Republicans.
Be advised – these maps are computed in a number of different ways. The Evans-Novak Political Report uses good old-fashioned brainpower to create their current map, drawing heavily on historical precedent and an evaluation of the demographics of the state. Others, such as the one you’ll find on RealClearPolitics or FiveThirtyEight.com rely heavily or entirely on the polling being done in the states.
Whatever map you decide to put your faith in, make sure you understand how the creators decided what states to put in which candidate’s column. Maps that are plug-and-chug with polling data are only as good as the numbers they’re using. At this early stage in the game, there’s not a lot of solid statewide polling out there in many places. If a map has filled a state in as “red” or “blue” based off of one poll done with a small sample back in February, it’s probably going to be about as believable as a farm-state legislator’s defense of ag subsidies.
While the numbers – national and statewide — at this stage in the game are good for setting the stage, it’s foolish (though tempting) to take them for more than they are. Elections don’t happen in a static environment. There is a human element that precludes a lot of predictability, particularly in a six-month time frame.
Setting that aside, however, the numbers are useful in educating us on what is going on in the minds of voters, helping us train our eyes on what groups or states to focus on or on what the campaigns should plan to do. Best of all, they give us an excuse to see Chuck Todd back on the air. Chuckolytes, rejoice!
Kristen Soltis is the Director of Policy Research at The Winston Group, a political polling firm. She is also a proud alumna of Chuck Todd’s class at Johns Hopkins University, and would consider herself a Chuckolyte, though of the slightly less creepy variety. The views expressed In this column are her own and do not necessarily reflect the views of her firm or employer.
Source: AFF Doublethink Online | Andrew Stiles
Source: AFF Doublethink Online | Kathlyn Ehl