DISCLAIMER: I take and use the text and graphics from the Bickers and Berry Report for educational purposes only.
Of the few times I hear about this model, people say, “This model is not useful for close races.” To the contrary, the model was designed for close races.
From the report:
These national vote models are assumed to be reliable forecasts of who is likely to
win the general election. In most cases, this assumption is reasonable. It becomes problematic, however, at precisely the point that forecasts are most interesting: when elections are close. In tight elections, national forecasts can and have produced a “winner” different from the actual winner.
The Bickers and Berry Model does not rely on the popular vote at all or on national polls. Why should it? The presidential race is fifty one separate state elections (DC being considered a state by the model).
Consider the forecasts and ultimate outcome of the 2000 election. Each of the 2000 presidential election forecasts predicted vice president Al Gore to win amajority of the two-party popular vote, which he did, but none correctly predicted governor George W. Bush to assume the presidency (Campbell 2001). Never in US history have White House residents been determined through a national popular vote. Presidential elections are decided through contests in the states and the District of Columbia. The forecast model we developed explicitly models the presidential contest based on factors inherent to these 51 jurisdictions. This modeling approach allows us to make a projection of the Electoral College result, which popular vote estimates cannot.
I’m not certain, but I do believe the Bickers and Berry model was the only one that correctly predicted W. Bush winning the presidency but Gore winning the popular vote. There is no closer presidential race than 2000 and the model worked when every other forecast didn’t.
I am positive that few, if anyone, bothered to actually look at the Bickers and Berry model. The reason why is that I hear people say, “How can you have a model where you don’t talk anyone? How can you make a forecast without polling data?” If anyone downloaded the report from the University of Colorado website, they would have read the following:
Methodologically, our model stands as an alternative to those of Campbell (1992), Cohen (1998), DeSart and Holbrook (2003), Holbrook and DeSart (1999), and Klarner (2008), which, like ours, predict outcomes on a state-by-state basis. Most of these models are based on horse-race public opinion polls in each state in the one to two months prior to the general election. When well done, these polls provide likely
voter responses to many factors influencing support for candidates in a state, including economic conditions, as well as factors that we are unable to observe. However, this strategy is dependent on the timeliness and quality of publicly available polls asking horse-race questions in each jurisdiction. Alternatively, Campbell (1992) and Klarner (2008) use a battery of polling, political, and contextual variables to forecast the presidential vote, but each model includes just a single state-level measure of economic conditions.
Bold is my emphasis.
Of course, Bickers and Berry are aware of state poll based modeling. The thing is that state poll based modeling is nothing new. The Bickers and Berry Model, instead, doesn’t allow inconsistent state polling and allows the more consistent economic data.
In contrast to these other Electoral College models, our model includes measures of change in real per capita income, as well as national and state unemployment figures. Accounting for both changes in personal income and unemployment provides a more robust approximation of state economic well being and, thus, serves to model the impact of retrospective evaluations of the incumbent party’s stewardship of the economy.
The model is using state economic data as well as national economic data. Even if the state is economically ‘happy’, it might frown if the national economic average is down.
One of my mistakes in 2008 was to assume since Bush wasn’t running (Constitution won’t allow a third term) that there was no incumbent. The way how the model works is to look at the party control of the White House, not the man. Therefore, if a president was replaced in the middle of his term or had to retire after two terms, the model would see it all as consecutive control by one party.
We also control for the term that the incumbent party is seeking.We consider two possibilities: incumbent parties seeking a second presidential term and incumbent parties seeking a third or higher term in the White House. This occurred most recently in 2008 when senator John McCain ran for a third consecutive
presidential term for the Republican party. In our modeling approach, only one binary variable is required to capture these alternatives.
But what is the heart of the model? Listen to this. I believe this is the ‘secret sauce’ to its longetivity:
The heart of our forecast centers on the third set of independent variables. We use two basic measures of economic conditions: unemployment levels and change in real income per capita. Unemployment is measured in two capacities. First is the national unemployment rate. The second is the corresponding unemployment rate in each state. Operationally, we use the U3 measure of unemployment, which is the “headline” unemployment figure most often reported by the media.
To observers of presidential elections, you’ve heard that a candidate’s home state matters and the state where the convention is held also matters. How does the model incorporates these idiosyncrasies?
The fourth category of independent variables is included to capture state-to-state idiosyncrasies of a given presidential contest. In doing this, we include binary variables identifying the home state of the Democratic and Republican presidential candidates in each contest. As shown later in the text, we find evidence that some candidates perform better, on average, in their home states. Therefore, we also include binary variables to identify the home states of the candidates in the last election. The inclusion of these variables is necessary because, absent such controls, the lagged two-party vote percentage will over predict the current vote for that party’s candidate in any state in which the prior election featured a major party candidate who hailed from that state.
Now here is something that surprised me. Listen.
In earlier iterations of the model, we also included binary variables to identify
vice presidential candidate home states, the states in which the nominating conventions were held, and governor partisanship (see also Powell 2004). Despite frequent media speculation that such things play a role in the final outcome, no statistically significant effect of any of these binary variables in the models that incorporated these variables or subsets of these variables is found.
Hah! The Bickers and Berry Model DOES NOT factor in the vice presidents’ home states, the states the conventions were held, or the partisanship of the state governors. Bickers and Berry even go so far to say that despite the media hype over these elements, there is no statistically significant effect of them. These variables were part of the older versions of the model but are not included today. Interesting.
In the details of the model, Bickers and Berry say that the Republican Party, when the incumbent, and the Democratic Party, when the incumbent, lose votes due to different economic variables. For every point of unemployment, the Democratic incumbent loses 3.3 points. However, the Republican incumbent suffers when the income level drops.
“Skip to the good stuff. What about the 2012 election? Certainly, Obama has advantages.”
Indeed, he does. Here are the advantages Obama has:
To be sure, he enjoys some advantages. First, Obama’s successful campaign in 2008 gives him a substantial leg up. He can lose some states that he carried four years ago without losing the election. Second, a prominent second-term incumbency advantage should prove advantageous.
But Bickers and Berry warn Obama is in electoral trouble:
Still, the big issue is the fragile economy. With an unemployment rate in excess of 8%, Obama is about two-and-a-half points beyond the break-even point for a Democrat running as the in-party candidate. This situation translates into slightly more than an eight-point reduction in his two-party vote, wiping out virtually the entire bump accruing to an incumbent seeking a second term. Moreover, as the country continues to rebound from the largest recession in generations, whether voters will ultimately judge the economy in relative or absolute terms is unclear.
Note the language. A Republican would not have written it as a question of ‘as the country continues to rebound’. The Republican would have written ‘as the recession grew worse’. I’m pretty certain Bickers and Berry are Democrats. “Whether voters will ultimately judge the economy in relative or absolute terms is unclear.” There is much hope in that statement.
The 2000 election is mentioned again. I do remember hearing about the University of Colorado predicting the popular vote and electoral vote split before the election occurred. This model was famous long, long ago.
The 2000 election is of particular interest because no forecasting model published in advance of that election correctly predicted GeorgeW. Bush as the winner. Because our model is predicated on the notion that during close elections, the Electoral College winner may not win the popular vote, it is critical that our forecast classify this election accurately. In 2000, the model correctly classifies 47 states, most notably Florida, which Bush was estimated to win with 51.2% of the two-party vote. The state’s final certified result awarded Bush a razor thin majority of the two party
vote at 50.004%. The only states incorrectly classified in 2000 are Pennsylvania, West Virginia, Arkansas, and Louisiana. Despite these inaccuracies, the model expected Bush to win 274 electoral votes to Gore’s 264, an error of a mere two votes.
The reader asks, “How accurate is this model?”
In 2008, the model performed similarly well, classifying 48 states correctly and missing Obama’s actual Electoral College vote total of 365 by just five votes. For comparison, Klarner’s (2008) median estimate of Obama’s electoral vote total was 346. The average error rate during the past four election cycles is about four states. Including all eight election cycles, the successful classification rate is nearly 90%, with an average deviation from the actual Electoral College result of 21.3.
Bickers and Berry Model came closer to the 2008 actual results than most models out there except for those that used internal campaign data *cough* *cough*.
But the reader asks, “Where is the largest inaccuracy in this model?”
The state-by-state forecasts for the elections of 1980 and 1992 have the most errors, with nine and 10, respectively.Of course, in these two elections independent candidates performed well. John Anderson received about 6.5% of the popular vote in 1980 and Ross Perot received nearly one out of every five votes cast for the presidency in the 1992 election.
The limit to Bickers and Berry Model is when third parties are present. 1980 and 1992 have sizable third party votes which flipped states that wouldn’t normally have been flipped. Since there doesn’t appear to be any huge third party presence in the 2012 election, the model should be more accurate than usual.
The impatient reader then demands: “Enough of this! What is the 2012 forecast?”
…the states we predict President Obama will carry include a substantially reduced set than those he carried in 2008.5 This is supported by the fact that no states won by McCain are predicted to flip to Obama. What is striking about our state-level economic indicator forecast is the expectation that Obama will lose almost all of the states currently considered as swing states, including North Carolina, Virginia, New Hampshire, Colorado, Wisconsin, Minnesota, Pennsylvania, Ohio, and Florida. Three other states that might be viewed as swing states—Michigan, New Mexico, and
Nevada—are predicted to stay in Obama’s column. Our forecast is that the president will receive 213 Electoral College votes, putting him well short of the 270 needed to win reelection.
“But what about the national vote?”
Finally, although not our primary objective, we can use
state-by-state vote projections to generate a forecast of the
national popular vote as well.
Moving to 2012, our forecast is that Obama will receive 47.14% of the two-party popular vote. Using confidence intervals around each individual state forecast and aggregating to a national popular vote as earlier described, our model projects
a likelihood of 77% that Romney will receive a majority of the ballots cast for the two major parties.
“How accurate has the forecast been for the popular vote?”
The popular vote predictions generated for each of the eight most recent election cycles are incredibly accuratewith an average deviation from the actual result of a mere 0.6%. In every election, except for 2008, the popular vote estimate is within one percentage point of the true result. In 1980, 2000, and 2004, the forecast error rate is less than 0.3%. The greatest deviation from the actual popular vote occurred in 2008 when our model overestimated the Obama vote by 1.3%.
The bold was my emphasis. Many polls overestimated Obama’s performance in 2008. Gallup had a 11 point D advantage and the final result was a 7 point D advantage.
“I heard there is a caveat about states around the 50% mark. What does that mean?”
The second caveat, which ties back to the first, is that a substantial number of cases depicted in figure 1 where the 90% confidence band around the state’s prediction includes the 50% mark. This indicates that the two-party vote could plausibly flip to the other side of the 50-50 line on which some of these states are currently predicted to land.
What this means is in the state prediction, largely determined by economic data, is that any state that is on 50% could easily flip one way or the other. What states are these?
(In the update at labor day, the economic data got worse which flipped New Mexico to Romney in the final analysis. The data above is not the final update (and the final update is a month before the election).
But let’s say Obama does better than expected by this model. He might get the states in the 49% and 48% range:
Going any further I suspect would be going too far. We will stop only at two percentage points moving away from 50%. Since the margin of error cuts both ways, what if Romney does better than the model suggests? Keep in mind that with the latest update, the model showed worse economic data. If the economic data kept getting worse, which appears to be the case (or else why is the government hiding the jobs report), then Romney might win these additional states: