Bradley Wascher is a professional elections analyst and an alum of extemp speaking, where he competed for 6 years. He was gracious enough to lend his time & write up an article for the Extemper’s Bible, where he combines his unique political insights and years of experience in extemp to provide stellar electoral analysis.
Every four years, the United States has a presidential election, and every four years, extempers have to talk about it.
Because the kick-up of the campaign always coincides with the beginning of the tournament calendar, it’s easy to become overwhelmed. So to help you get your bearings, here’s a how-to on extemping about elections.
As a disclaimer, this article is not a primer on the US government or an overview of its electoral history — there’d be way too much to cover here. Instead, think of this as a guide for responsibly and persuasively analyzing politics, along with some resources to steer you in the right direction.
The lens for viewing politics
One of the best pieces of advice I ever received as an extemper was this: politics is politics is politics is politics. Basically, it matters less if you know the minutiae of every person or place when you flip over a question; the real skill is developing an understanding of the broad rules of the game, so you can apply that theoretical knowledge to any current event thrown at you. It happens all the time in extemp. You might have to speak on a race that you previously hadn’t read much about. Well that’s okay, because as long as you remember the general norms dictating electoral politics, you at least now have a starting point.
In the same way, news is news is news is news. An extemper drawing a question at a tournament is not all too different from a political analyst being asked a similar question in an interview. Whether your audience recognizes it or not, they expect you to take a realistic and empirical stance when discussing these topics. They can detect malarkey — and when they do, they don’t judge it fondly.
To be clear, that doesn’t mean you should omit your personal opinions from the speech, or feel forced to advocate for things you dislike. But it does mean, for example, you should not assert that some prediction will undeniably come true when it very obviously won’t. After all, it’s a lot easier to build credibility and trust among an audience when you’re playing the role of the objective consultant rather than the unreasonable hack. You need to be believable.
Empiricism is especially crucial for questions about campaigns. No matter how much you love any Democratic candidate for president, they will probably not win Wyoming’s three electoral votes anytime soon, and the same goes for a Republican nominee and D.C. On the other end of the extreme, a candidate leading by one point in a poll doesn’t automatically imply they’re guaranteed to win it all.
The fact of the matter is, forecasting elections isn’t a matter of red or blue, black or white, yes or no. There’s a whole range of possible outcomes, and if you want your audience to buy your analysis, you should do your best to paint that gradient. Your side is not the only side, your answer is not the only answer, and your arguments are not the only arguments.
This all probably sounds like common sense. But during the frenzy-filled half-hour blur that is an extemper’s prep time, these basic truths can get taken for granted. In my experience, speakers most often fall short by failing to think probabilistically and forgetting to express uncertainty.
These two issues go together, and they’re just fancy ways of restating what I said above. Democrats probably can’t win in Wyoming, Republicans are unlikely to take D.C., and a candidate leading in the polls is far from a sure shot. Notice how each claim includes indefinite language — words like “probably” and “unlikely” are your best friends when describing an election. Candidate X has a very low chance of winning, but that doesn’t mean the probability is zero or never; Candidate Y’s odds are very high, but that doesn’t make the certainty one hundred or always. Ask anyone who’s ever been struck in the head by a baseball that flew out of the park: low-probability events do happen.
For this reason, the classic extemp question — “Who will win the election?” — is honestly lowkey misleading. Sometimes we have no clue who will win, and even in the cases where we’re pretty sure, that confidence can never be full. So in the process of defending the answer you chose for your question, express an appropriate amount of uncertainty and avoid absolute language. That’s true in any speech, regardless of topic area.
Places to look
While the bad habits I mentioned above can be tough to break, there luckily are a few resources to help you recalibrate.
First, every follower of elections should know the best practices for reading public opinion polls. There are too many guidelines to list here (read this article and this article for more information on that), but the most important quick tip is to cite polling averages, instead of only one poll, when possible. Due to the errors inherent to survey research, it helps having as many data points as possible to support your speech.
Second, familiarize yourself with race raters and election forecasters. Think of these folks as “professional extempers” — it’s literally their job to do this stuff. And their daily workflow even resembles an extemper’s prep time. In order to reach conclusions about how they think an election might go, these researchers consider a constellation of factors, from fundraising to polls to demographics to historical trends to candidate-specific assessments. They then present their predictions in an accessible and easy-to-follow fashion.
For insight on practically every federal contest, my colleagues and I publish race ratings at Inside Elections, and the experts at The Cook Political Report and The Crystal Ball make their own predictions as well. For statistical election forecasts that produce estimated win probabilities, I heavily advise checking out the methodology sections for a deeper understanding of how they work; the two most popular models this year are hosted by FiveThirtyEight and The Economist, with The New York Times’ Upshot and others having issued forecasts in prior cycles.
As helpful as these sites can be for extempers, they alone do not constitute entire points. To see what I mean by that, let’s go back to the question, “Who will win the election?” Although it can be enticing to find one poll or forecast and think you’ve finished a third of your speech already, you should avoid using a single topline by itself as the crux of an argument. That’s like saying something will happen later because it’s happening now, which of course isn’t always the case. The causality is also unclear — a president doesn’t get elected because they were leading in the polls.
Instead, and sorry for stating the obvious, it’s much more straightforward to cite these sources as evidence or warrants within points. The main arguments could be derived by thinking like a race rater or forecaster: construct reasons with falsifiable and empirical indicators, such as fundraising numbers or electoral history.
Then use the experts to reinforce why you’re right. For instance, a point broadly describing past demographic patterns in the A subpoint can be verified by fresh polling about that demographic in the B subpoint. A general and theoretical claim about incumbents’ strength receives credibility if you also name-drop particular political scientists who have dug into the data themselves. When it comes to adding other rhetorical tactics, topline estimates of a forecast or survey serve as great one-sentence summaries of your entire speech, so they’re strong in power lines at the end of points or significance restatements at the top of the conclusion.
By taking these steps, an extemper is not only improving their analysis — they’re giving their audience a realistic and accurate play-by-play of current events. And in an activity that requires judges to suspend their disbelief, being believable is one of the most powerful traits a speaker can have.