The PollyVote team has completed its first survey of elections experts to forecast the 2012 presidential election.  In the mid December-early January survey, 15 academics from a variety of colleges and universities predicted that President Obama will eke out a narrow victory over his Republican opponent. 

The trimmed mean forecast is that Obama will garner 50.7% of the major-party vote. The experts are far from certain of their prediction’s accuracy, however, as might be expected with 10 months to go before the election.  Nevertheless, most experts’ assessments of Obama’s chances are quite similar, with 12 of the 15 predicting an Obama victory.  Forecasts of the Obama vote range from a high of 52% to a low of 48%. 

This is the third presidential election campaign in which an expert panel has been formed by the PollyVote team to forecast the election outcome.  The first surveys, those in 2004, did not begin as early as the current survey, but those for 2008 did.  The second 2008 survey was conducted in December 2007 and early January 2008, resulting in a 47.7% mean forecast for incumbent Republican party candidate John McCain.  McCain in fact received 46.3% of the major-party vote, for a mean forecast error of 1.4% for the experts. 

As evident in the January 2008 column in the Figure below, for this long-term forecast the experts were more accurate than other competing methods during this period, which all incorrectly predicted larger shares of the vote for McCain.  Mean errors were 2.6% for campaign polls, 2.50% for the Iowa Electronic Market vote share contract, and 1.8% for Ray Fair’s econometric model (the only such model generating forecasts this early in the campaign).       

The PollyVote team plans to conduct surveys monthly until fall 2012, when the experts will be polled more frequently. 

Forthcoming in Foresight

In March 2004, Scott Armstrong, Alfred Cuzán, and Randy Jones launched PollyVote.com (Cuzán et al., 2005). The goal of this project was to use the high-profile application of election forecasting to demonstrate advances in forecasting methodology. The primary focus was the “principle of combining”, in which one can achieve large reductions in forecast error by combining forecasts from different methods that draw upon different data. Armstrong (2001) has summarized evidence from 30 empirical comparisons. He found that combining nearly always reduced the error from the typical individual method. The average error reduction was 12.5 percent.

Computation of the PollyVote

During the past two U.S. presidential elections, in 2004 and 2008, Polly the Parrot, the mascot of the PollyVote, provided predictions of the two-party popular vote-shares by averaging forecasts within and across four component methods: trial-heat polls, forecasts from the Iowa Electronic Market (IEM), quantitative models and experts’ forecasts.

Combining within component methods was used to aggregate forecasts based on a similar approach and similar information. It is, however, more powerful to average across methods, especially when the methods draw upon different information. Using equal weights, the final PollyVote forecast is computed by averaging forecasts across the four component forecasts.

Past performance

With this simple averaging procedure, the PollyVote has provided highly accurate forecasts of election outcomes. In 2004 Polly posted updated forecasts every two or three days.  An automated system developed in 2008 enabled Polly to provide daily forecasts.  In both elections PollyVote forecasts were made beginning more than half a year before Election Day. Each single PollyVote forecast -- even those generated several months before the election -- has been correct in predicting the election winner.  In the final forecasts released on Election Eve, Polly missed the candidates’ actual two-party vote-shares by 0.3 percent in 2004 and 0.7 percent in 2008 – an average error of only 0.5 percentage points.

Graefe et al. (2011) summarize the predictive performance of the PollyVote for the past five U.S. presidential elections, reporting retrospective analyses of the 1992, 1996, and 2000 elections, in addition to the ex ante forecasts for 2004 and 2008. They found that the PollyVote forecasts reduced error by 52% to 58% compared to forecasts of the typical randomly chosen poll, model, or expert. Compared to the IEM, which is essentially a means of aggregating information from disparate sources, the PollyVote reduced error by 10%.

PollyVote 2012

To forecast the 2012 presidential election, Polly added a new component, models that are based on the index method. This method, which has a long history in forecasting and decision-making, is useful for identifying the best of several options. It draws upon prior evidence to determine which variables are important and how they affect the outcome. The method works well when there are many important variables and good prior knowledge (Armstrong & Graefe, 2011).

The first index model used for presidential election forecasting was the “Keys to the White House” by Allan Lichtman (2010), which first appeared before the 1984 election. As noted by Nadeau and Lewis-Beck (2012) in this issue, the Keys forecast the election winner by assessing how well the party in the White House has governed the country. Since the Lichtman index first appeared, other index models have been developed that predict the election outcome based on information about candidates’ biographies or voters’ perceptions of candidates' ability to handle issues facing the country. Detailed information on each of the models can be found at PollyVote.com.           

Indexes are a proven forecasting method and draw upon different information than traditional approaches to election forecasting. Therefore, including them as a separate component of the PollyVote is expected to further increase Polly's forecast accuracy.

Year-ahead forecast for 2012

On January 1, 2011, almost two years before Election Day, the PollyVote was launched to forecast the 2012 presidential election. Since then, the daily updated forecast has consistently predicted Obama to win the two-party popular vote, with a vote-share forecast ranging from about 54% shortly after the death of Osama bin Laden in May, to almost 50% in early November.

As of mid-November 2011, the combined PollyVote forecast predicts Obama to win 51% of the vote. Thus the forecasts of the four component methods currently available suggest a close race. While the combined poll forecast shows a slight advantage for Obama (50.5%), the IEM (49.6%) and the average of five econometric models (49.5%) put him slightly behind the Republican candidate. Only the average forecast of four index models predicts a clear victory for Obama (54.4%).

These forecasts will, of course, continuously change during the run-up to the election. In particular, the selection of the Republican candidate may have a big impact. In addition, econometric and index models may not have been released yet, and results from other indicators will be updated as new data becomes available. Current polls and updated IEM prices are published almost daily – and generally become more accurate closer to Election Day. Finally, Polly is waiting for her fifth component method, the experts’ forecasts, which will be added as soon as available.

Summary

The PollyVote combines forecasts within and across virtually all methods that are commonly used in election forecasting. This is beneficial because large gains in accuracy can be achieved by combining forecasts from different methods that draw upon different data. Instead of choosing a single forecast, Polly relies on a combination of several forecasts in order to avoid large errors. Polly thus demonstrates through election forecasting the validity of the "principle of combining" for forecasting problems more generally.

References

Armstrong, J. S. (2001). Combining forecasts. In: J.S. Armstrong (Ed.), Principles of Forecasting: A Handbook for Researchers and Practitioners. Boston: Kluwer Academic Publishers, pp. 417-439.

Armstrong, J. S. & Graefe, A. (2011). Predicting elections from biographical information about candidates, Journal of Business Research, 64, 699-706.

Cuzán, A. G., Armstrong, J. S., and Jones, R. J., (2005).  How we computed the Pollyvote. Foresight, 1, 51-52.

Graefe, A., Armstrong, J. S., Jones, R. J. and Cuzán, A. G., Combining forecasts: An application to election forecasts. APSA 2011 Annual Meeting Paper. Available at SSRN: http://ssrn.com/abstract=1902850.

Lichtman, A. J. (2010). The keys to the White House: Forecast for 2012, Foresight, 18, 33-37.

Nadeau, R. and Lewis-Beck, M. S. (2012). Does a presidential candidate’s campaign affect the election outcome? Foresight, 24, XX-XX.

According to the Time-for-Change forecasting model, which has correctly predicted the winner of the popular vote in every presidential election since 1988, Barack Obama has a good chance of winning a second term in the White House next November. The main reason for this is that the Democratic Party has only held the White House since 2008. That makes Obama a first-term incumbent, and first-term incumbents rarely lose.

The Time-for-Change model is based on three predictors: the growth rate of the economy in the second quarter of the election year (Q2GDP), the president’s net approval rating in late June or early July of the election year (JUNEAPP) and the time-for-change factor (CHANGE), which takes on the value of zero if the president’s party has been in office for one term and one if the president’s party has been in office for two or more terms. The model predicts the incumbent party’s share of the major party vote (PV).

Each of the three predictors is assigned a weight based on a regression analysis of the 16 presidential elections since World War II and the results are as follows:

PV = 51.7 + (.11*JUNEAPP) + (.54*Q2GDP) – (4.4*CHANGE).

The three predictors are highly statistically significant and the model explains 89% of the variation in the incumbent party’s share of the major party vote. The weight assigned to the CHANGE factor in the model indicates that after controlling for both the growth rate of the economy and the incumbent president’s approval rating, a first-term incumbent like Barack Obama can expect to receive an additional 4.4% of the major party vote compared with a candidate seeking to extend his party’s hold on the White House beyond eight years. That explains why first-term incumbents rarely lose. In the past century only one first-term incumbent has been defeated — Jimmy Carter in 1980. (Remember, George H.W. Bush lost after one term too, but he took over for Ronald Reagan, who had served two previous terms. Had Bush been reelected, the Republicans would have held the White House for 16 straight years.)

Could Barack Obama be the next Jimmy Carter? There is clearly a chance that he could, if the condition of the U.S. economy deteriorates in 2012 and/or the president’s approval rating slips further into negative territory. But because of the first-term incumbent advantage, the model gives Obama a good chance of winning a second term even with fairly modest economic growth next year and an approval rating in the low- to mid-forties, as the following table shows. The table presents conditional forecasts of the Democratic share of the major party vote in 2012 depending on the growth rate of the economy in the second quarter and Barack Obama’s net approval rating at mid-year.

Chart 1: Conditional forecasts of Obama’s share of major party vote

Based on the president’s average net approval rating in recent national polls, between -10% and -5%, and the estimated growth rate of the U.S. economy during the second quarter of 2011, 2.5%, Obama would be expected to win approximately 52% of the national popular vote, enough to almost certainly guarantee him a majority in the Electoral College. But the president clearly has little margin for error. A double-dip recession and/or a substantial decline in his approval rating could easily put him below 50%.

A note on Nate Silver’s Model

In an article in the Nov. 6, 2011 Sunday New York Times Magazine, election guru and pundit Nate Silver of FiveThirtyEight.com argues, based on a statistical model of U.S. presidential elections, that Barack Obama faces a very difficult battle for a reelection if his opponent is Mitt Romney and if economic growth does not improve markedly in 2012. Silver’s model includes three predictors — the president’s approval rating one year before the election, the growth rate of GDP in the year of the election and the ideological extremism of the opposing party’s nominee based on a rating scale developed by the political scientist Marty Cohen and his colleagues in their 2008 book on the presidential nominating process.

Based on these three factors, Silver presents some estimates of Barack Obama’s chances of winning a second term depending on his current approval rating, the growth rate of the economy next year, and whether the Republican Party nominates Mitt Romney or a more conservative candidate. Obama does not fare well against Romney under most plausible economic growth scenarios but does somewhat better against other possible GOP candidates such as Herman Cain or Rick Perry.

But there are several problems with Silver’s model. First, it isn’t really a forecasting model because the growth rate of the economy during the year of the election won’t be known until long after the election is over. In addition, the measure of the opposition candidate’s extremism is highly subjective. Many of these candidates in past elections did not have a voting record so the opinions of presidential historians were used to measure their extremism, but obviously different historians can reach different conclusions about a candidate’s extremism. This raises the possibility that the ratings for past elections could have been influenced by the results of those elections or that Silver’s own ratings of the 2012 GOP candidates could have been influenced by poll results showing Romney doing better than Cain or Perry against Obama. Moreover, it’s not even clear that this variable works. Although it has a significant effect in Silver’s own model, when it is added to the Time-for-Change model it is not statistically significant and the explanatory power of the model increases by less than one percentage point.

More importantly, Silver’s model may underestimate Barack Obama’s chances of winning a second term in the White House because it does not take into account the advantage enjoyed by first-term incumbents. And that advantage, as we have seen, is quite substantial.

Note: This article was first published here.

Allan Lichtman posted a year-ahead forecast of his Keys model, which can be accessed here. Although the Keys now consider Occupy Wall Street (in the Social Unrest key) and Solyndra (in the Major Scandal key), the coding of the Keys did not change compared to the last udate in June.

Obama still wins 10 of the 13 Keys and is therefore predicted as the clear election winner. That is, even if the economy would dip into another recession in 2012 (which would change the coding of Key 5), the forecast of the Keys model is in favor of Obama. Note that this is contrary to many econometric models that incorporate economic measures such as GDP growth as predictor variables. For example, Ray Fair's model predicts Obama to lose the election if the economy is going into another recession.

The Keys to the White House is a system for predicting the popular-vote result of American presidential elections, based upon the theory of pragmatic voting. America’s electorate, according to this theory, chooses a president, not according to events of the campaign, but according to how well the party in control of the White House has governed the country. If the voters are content with the party in power, it gains four more years in the White House; if not, the challenging party prevails. Thus, the choice of a president does not turn on debates, advertising, speeches, endorsements, rallies, platforms, promises, or campaign tactics. Rather, presidential elections are primarily referenda on the performance of the party holding the White House.

Retrospectively, the keys model accounts for the outcome of every American presidential election since 1860, much longer than any other prediction system. Prospectively, the Keys to the White House has correctly forecast the popular vote winner of all seven presidential elections from 1984 to 2008, usually months or even years prior to Election Day. For more information on the Keys model, click here.

A new issue poll, conducted by Washington Post / ABC News between October 31st to November 3rd, reveals that voters are tied when it comes to the question of who would do the better job in handling the most important issue: the economy.

This information is used by two models:

• The forecast of the PollyIssues model, which predicts the election outcome based on how voters expect the candidates to handle all issues, remained unchanged, predicting that Obama will achieve 55.0% of the popular vote. This is because Obama still owns the "economy" issue, despite decreasing support by voters.
• The forecast of the PollyBigIssue model, which predicts the election outcome based on how voters expect the candidates to handle the most important issue (i.e., the economy), now has Obama at 53.1% of the popular vote - 0.4 percentage points lower than before the release of this poll.

Polly added the electoral-cycle model by Helmut Norpoth to its combined econometric model forecast. The electoral-cycle model predicts the election outcome based on electoral history. In particular, it uses the vote of the two most recent elections as predictor variables in a linear model, which has been estimated based on data from all elections since 1828. As described by Norpoth, the model reveals the following pattern:

• There is a boom-bust cycle of electoral fortunes
• The cycles are irregular in length (period) as well as width (amplitude)
• The average length of a cycle is estimated as 5.2 terms
• The White House party can expect to stay in office 2.6 terms 

Alfred Cuzan provided Polly with an updated forecast of the fiscal model. The fiscal model is a variation of Ray Fair's model and includes a "fiscal policy" variable that measures the change in federal outlays / GDP. According to this model the incumbent loses support if fiscal policy is expansive or expansionary relative to the economy.

Plugging in the latest assumptions for the economic development that Fair used in his latest update and averaging the forecasts of the two variants of the model, the fiscal model predicts Obama to achieve 46.6% of the popular two-party vote. Click here to read the short note on this model update.