DavidMRothschild on September 25, 2012 @ 3:12PM
There are five states that have a double digit number of electoral votes, where Obama and Romney each still have a decent 15 percent shot at winning: North Carolina (where Obama has a 31.8 percent chance), Florida (57.9), Virginia (73.6), Ohio (77.3), and Wisconsin (83.9). There is something striking when you look at a chart of the likelihood of any of them voting for Obama since August 1; there is a lot of movement, but very little crossing.
States respond to local shifts. Wisconsin's dip started just after their native son Paul Ryan joined the Republican ticket. We also know that the candidates' fortunes in states correlate with local economic trends, and that the huge differential across swing states in spending on advertisements and get-out-the-vote initiatives impacts the support and turnout in those states as well.
Yet trends still tend to be more national than regional or local. Year after year, granular predictions of the state-by-state outcome over the months preceding an election look similar to the above chart with few states cutting dramatically across the other states.
The likelihood of Romney winning North Carolina and Ohio is not much greater than the likelihood that he will win Ohio. Because the most likely way for Romney to win Ohio is a national trend that moves all of the states between North Carolina and Ohio into his column, picking off Ohio before states that currently favor Romney more is very unlikely. Conversely, the likelihood of Romney winning Ohio and losing North Carolina is almost negligible. (I am further examining the nature of these state-by-state relationships in a prediction game on my website, which readers are encouraged to play.)
This method of determining joint probability is called the ranking method and it has proven surprising difficult to beat this simple/transparent and reliable method. Drawing this method out to all states in the Electoral College, if you list all states from most likely for Romney to most likely for Obama, it is unlikely that any state moves more than a few points in rank. Thus, the quick and easy estimate for the likelihood of any candidate reaching 270 electoral votes is to figure out the state that flips the election if every state stays in their order. Below is the key section of the states 23-32 in that list:
If Romney takes Missouri he has 191 electoral votes, 206 electoral votes with North Carolina, onward until he crosses 270 with a victory in Ohio. Similarly, Obama has 243 electoral votes if he takes Nevada, 253 with Wisconsin, and he crosses 270 with Ohio.
The flip state is Ohio at 77.3 percent for Obama and 22.7 percent for Romney. This is a great, simple approximation for the likelihood of the election. It has tracked my more complex model of electoral victory, within 2 or 3 percentage points, for the entire election cycle.
This article is syndicated on the Huffington Post.
The downside of outside spending: Candidates are hard to shop for (Originally posted on Yahoo!'s "The Signal" Blog)
DavidMRothschild on September 25, 2012 @ 11:37AM
Mitt Romney's campaign understands that almost every reasonable scenario for victory includes winning Ohio, Virginia and Florida—a troika that, along with all the states safely in the Republican column, would award the challenger 266 electoral votes, four shy of the magic number. Campaign spending figures published by National Journal verify this in no uncertain terms. Since May 1, the Romney campaign and its allies have spent more on advertising in these three states than in all other competitive states combined. The same is true of the Obama campaign, whose clearest path to victory involves denying Romney any one of these battlegrounds.
Where the campaigns blow their overflowing fountains of cash is only half the story, of course, due to the torrents of outside spending flooding this campaign. Overall, the Republicans and their supporters have outspent the Democrats $256 million to $217 million since May. This is a little misleading, however, because of a simple economic fact: The marginal value of a campaign dollar is significantly higher if raised by the campaign than if raised by a super PAC.
Outside spending groups are not allowed to coordinate with campaigns, though they can coordinate with one another and operate in the same political reality. In an era of incredibly precise political targeting, however, outside spending that is not privy to the campaigns' precise strategies and messaging is not as effective. Consider the difference between spending $100 on yourself and having a friend buy you something for $100, especially if this well-meaning friend is not legally permitted to ask you what you want. Economists call this the "deadweight loss of Christmas."
DavidMRothschild on September 24, 2012 @ 6:05PM
We are very excited to introduce to you a new feature of PredictWise, called WiseQ Game. WiseQ Game - Elections 2012 is the first of a series of games that will allow you to place your own predictions on politics and economic indicators. Over the next few months the selection will expand to entertainment, sports, and finance. We have created this game for our users for a few reasons:
PredictWise is dedicated to providing the most accurate and meaningful predictions possible, on a real-time updating basis. We hope that these games will delight our readers and provide us with a new source as well that will let us both improve and expand on our predictions.
1) You follow predictions and we want to give you an opportunity to "comment" in the most interesting and meaningful way possible. Go into the game and wager that the current odds should be higher or lower for a particular event! Can our readers "beat" our aggregated predictions?
2) We spend a lot of time aggregating the available data in the world, but we see limits to that data. This project will provide us with new a data source that will allow us to create even more accurate and meaningful predictions.
3) We also follow markets and prediction games closely, and we are really excited about some of the market design innovations we have created including simple front-end wizards for making predictions and really complicated back-end market makers to keep track of all of the predictions.
Democrats likely to retain Senate as four critical races shift in their favor (Originally posted on Yahoo!'s "The Signal" Blog)
DavidMRothschild on September 24, 2012 @ 1:31PM
After weeks in which the fate of the Senate simmered at nearly even odds of flipping for Republicans or remaining in Democrats' control, the outlook has shifted dramatically in the Democrats' favor. The incumbent party now has an 80 percent chance of retaining its majority, according to the Signal's prediction model.
The break is largely due to critical races in Missouri, Wisconsin, Massachusetts and Virginia, all of which have favored the Republican at some point in the past month and now favor the Democratic candidate.
2012 Congress - Likelihood of Party Control
Can Romney still win? Of course—just not the way things are going (Originally posted on Yahoo!'s "The Signal" Blog)
DavidMRothschild on September 21, 2012 @ 8:51AM
This is not where the Romney campaign wanted to be three weeks after Tampa.
The bounce in the polls that President Barack Obama netted coming out of the Democratic National Convention might have vanished by now, as those postconvention bumps tend to do, if not for the bad press that has pelted Republican challenger Mitt Romney nearly every day since the Democrats returned from Charlotte. First, Romney's response to the death of the American ambassador to Libya was widely viewed as inappropriate, even within his party. Just as he was recovering from that stumble, Mother Jones magazine released a leaked video of Romney disparaging the work ethic of 47 percent of Americans.
What everyone wants to know, of course, is whether historians will look back on these episodes as the effective end of Romney's chances at the presidency, to which we answer: Of course not. This is not just because the news media will tire of this story line by the end of the week, though that's part of it. ("Assignment editors: Now is the time to order up those 'Romney's coming back' pieces," one columnist quipped on Tuesday.) Mostly, it is the fact that new information and new story lines will relentlessly pile on for the next 47 days.
DavidMRothschild on September 19, 2012 @ 1:34PM
The effect of the hidden video of Mitt Rommey addressing fundraisers about the 47% of Americans that do not pay federal income taxes was about 2.5 percenage points. We are not sure if the polls will really be affected in the short-run, but the prediction markets anticipate a long-term cost to Romney. The video will play in advertisments, dominate the debate over the next few days and weeks, and may play a major part in debates. Even if it does not move polls tomorrow, it will, likely, make a difference on Election Day.
DavidMRothschild on September 18, 2012 @ 11:50AM
The headline would not surprise regular followers of the election; both aggregated polls and expectations created from polls have been telling that story. What is impressive is the trend and magnitude of the movement in last few weeks (i.e., since the end of the Republican National Convention). Below I have charted the likelihood of victory for Obama from my forecast that uses a combination of fundamental data, polling data, and prediction market data.
As an incumbent candidate with mediocre economic indicators, Obama's likelihood of victory hovered at just over 50 percent likelihood a year before the election. A full year before the election the uncertainty in the prediction rests in a year's worth of events: from the nomination of the Republican candidate, to economic indicators, to any of the multitude of unforeseen events that can occur in a year's time.
Obama's rise during the first few months of the 2012 occurred for two reasons: Romney's vulnerability and strong economic indicators. First, my model always had Romney as the most likely Republican nominee, but struggling to secure his party's nomination made him appear weak heading into the general election. Second, economic indicators were surging during this period, 275,000 jobs were created in January, inviting the possibility of a quick and decisive economic recovery.
The economic indicators took a turn for the worse in the spring and Obama's numbers collapsed along with them. Research that I conducted with Patrick Hummel makes three key observations about economic indicators and elections. First, it is trends not levels that are important; if economic levels were key, than Obama, or anyone who took over in January 2009 when we lost 818,000 jobs, would have no chance at reelection. Second, the impact of economic indicators crescendos around the second quarter of the election year; they continue to be meaningful after that, but there is a diminishing return to their impact. The storyline created by the first and second quarters was a tepid recovery and unless something shocking happens in the next few indicators, that story is hard to shake. Third, state-by-state indicators are of paramount relative to national numbers; thus, it is very relevant that key states such as Ohio and Virginia are doing much better than the national average.
As the summer turned towards Labor Day, Obama gained ground every day that Romney did not. Time is kind to incumbents and leading candidates. For example, Romney continues to struggle with historically low favorability and every day that he does not turn that around is one less day for him to make it happen.
The conventions and the nomination of a vice-presidential candidate are a key opportunity for the trailing candidate to shake up the race. The Republicans did not exceed expectations and the Democrats did not fall below expectations. That juxtaposition alone is sufficient for the Democratic nominee to trend upward and gain in likelihood of victory.
The remaining uncertainty in my prediction is not what will happen if the election were held today, but in what will happen over the next few weeks that could impact the election. The burden is now completely on Romney and/or Ryan to move the needle in their four debates, or orchestrate some other major event. If nothing major happens, than Obama will be reelected.
Where does this leave the overall election for me as we head into the Republican National Convention -- I have Obama with a 58.9 percent likelihood of reelection, or the exact same likelihood as his chances in the pivotal swing state of Ohio.
This article is syndicated on the Huffington Post.
DavidMRothschild on September 12, 2012 @ 3:48PM
There are four trends to consider when thinking about the general election so far:
As Romney encountered stiffer-than-expected resistance in his march to the nomination, he began to appear vulnerable as a general election challenger. Economic indicators were looking up in the first quarter of 2012, fortifying Obama's position. If that trend had continued into the second quarter, this would not be a close election.
The jobs numbers, released the first Friday of every month, hit the skids in the second quarter. Obama's re-election likelihood tumbled with them. April, May and June were the three lowest job growth months this year.
From early July to the present, we see the influence of Father Time, who is unkind to challengers of either party. By definition, an election is the incumbent's to lose, and his or her opponent has a finite amount of time to successfully make the case for a change in leadership.
Last, we see that the conventions were a net boost for Obama. Romney netted a historically low poll bounce in an average convention and the unfortunate presence of Hurricane Isaac. Obama gained a modest bounce in the polls from his own convention highlighted by President Clinton's policy-oriented address.)
Markets sour on both Romney and Obama speeches, but conventions boost Obama (Originally posted on Yahoo!'s "The Signal" Blog)
DavidMRothschild on September 07, 2012 @ 1:46PM
While the purpose of the political conventions was to draw clear distinctions between President Barack Obama and challenger Mitt Romney, the two men remain identical in one key respect: When they open their mouths before a national audience, their odds of winning in November suffer.
As you see here, the likelihood of Obama's re-election jumped about 1 percentage point immediately after Romney's address last Thursday evening, and fell slightly after Obama's turn last night. (The model for predicting the outcome of presidential elections relies on many contemporary and historical factors, but the fine-tuned reactions you see here are driven by prediction markets.)
DavidMRothschild on September 05, 2012 @ 6:41PM
There were 10 lead changes over the course of the primary season in the polls, but the foresight of our prediction model allowed us to hold steady with eventual nominee, Romney.
Sources: Betfair, Intrade, and RealClearPolitics
As we enter the second night of the Democratic Convention, Obama and Romney are tied in both HuffPost’s Pollster and RealClearPolitics’ aggregated voter-intention poll averages. With foresight over the events that are yet to come in the election cycle, our prediction model concludes that Obama is 62.1 likely to win reelection, despite the tie in the polls.
There is a huge value to polls in understanding where the race stood if the election were held on that day; that is what polls provide. It was both fun and insightful to see Perry, Cain, Gingrich, and Santorum have their time as the anyone-but-Romney candidate. Yet, most people just want to know who is going to win; for that that, we encourage you to follow our predictions.