Download
how to forecast an election and how to win one countdown to election 2012 n.
Skip this Video
Loading SlideShow in 5 Seconds..
How to Forecast an Election, and How to Win One! Countdown to Election 2012 PowerPoint Presentation
Download Presentation
How to Forecast an Election, and How to Win One! Countdown to Election 2012

How to Forecast an Election, and How to Win One! Countdown to Election 2012

218 Views Download Presentation
Download Presentation

How to Forecast an Election, and How to Win One! Countdown to Election 2012

- - - - - - - - - - - - - - - - - - - - - - - - - - - E N D - - - - - - - - - - - - - - - - - - - - - - - - - - -
Presentation Transcript

  1. How to Forecast an Election, and How to Win One! Countdown to Election 2012 Leighton Vaughan Williams Professor of Economics and Finance Director, Betting Research Unit Director, Political Forecasting Unit Nottingham Business School Nottingham Trent University Leighton.Vaughan-Williams@ntu.ac.uk

  2. How Well Do Markets Aggregate Information? • How wise is the crowd?

  3. What are Prediction Markets?

  4. Betting on the outcome • Prediction markets are speculative (or betting) markets created or employed for the purpose of aggregating information and making predictions. Based on the ‘Efficient Markets Hypothesis’, the idea that markets accurately incorporate all relevant information.

  5. Polls or markets? • Predicting the winner of an election!

  6. Early Prediction Markets • The earliest data we have from prediction markets are those from organized markets for betting on the US Presidential election between 1868 and 1940. • As an example, the New York Times reported that between $500,000 and $1 million was wagered on the Curb Exchange in one day on the 1916 election and that “oil stocks were almost forgotten”. The total amount wagered in these markets in 1916 was about $200 million (at 2012 prices). • In this period between 1868 and 1940, the market failed to predict the winner on just one occasion.

  7. AN EARLY BRITISH PREDICTION MARKET • Brecon and Radnor By-Election, 1985. • Mori v. Ladbrokes

  8. ELECTION EVE • MORI: Labour to win by 18%. • LADBROKES: • Liberal candidate: 4 to 7 • Labour candidate: 5 to 4

  9. WINNERS: • The Liberal candidate. • Those who ignored MORI and backed the market favourite.

  10. Bush v. Gore, 2000IG Index v. Polls

  11. Outcome forecasts • IG Index: • 265-275 Bush • 265-275 Gore • Rasmussen: Bush by 9% • Result: • Bush 271; Gore 266 • Gore by 1%

  12. Opinion Polls v. Markets • Opinion polls, like all market research, provide a valuable source of information, but they are ONLY ONE source of information. • Other information includes: • 1. Local canvass returns • 2. On-the-ground inside information • 3. Forecasting models • 4. Opinions of professional ‘pundits’ (‘experts’) • 5. Focus groups • Betting markets aggregate all the available information.

  13. Follow the Money! • Betting websites also offer a more accurate prediction of electoral outcomes than opinion polls because the average voter doesn’t have any incentive to respond or tell a pollster the truth about their voting preferences, but when they are betting their own money they will tend to think hard about the choices they make using the best information available to them. • People who know the most tend to bet the most, but people who know only a little tend to bet only a little. So you can be pretty confident that a market deep with people’s hard-earned cash is giving you a quite accurate insight into what is going on. • The more money involved, the more efficient and accurate the market will tend to be. • It really is a case of “follow the money”.

  14. Moreover ... • ...Unlike polls, which are snapshots of opinion, betting markets are all about forecasting the eventual outcome. • The lower the transaction costs (there is no tax on betting in the UK) and the better the access to information (in never more plentiful supply due to the Internet), the more efficient we might expect betting markets to become in translating information today into forecasts of tomorrow. • Since the advent of zero tax low-margin betting exchanges, like Betfair, the accuracy of these markets has improved yet further. • These exchanges differ from traditional betting markets by eliminating the odds-setting bookmaker, instead providing the technology to match up the best offers to back and lay an outcome on offer from all the clients of the exchange. • This ensures that the margins implicit in the odds are lower than they have ever been. For all these reasons, betting markets today are likely to provide better forecasts than they have done at any time in history.

  15. US Presidential Election 2004 • State-by-state predictions: 50 out of 50.

  16. Who Will Win? • Asked by The Economist magazine, to call the winner and the seat majority in the 2005 British General Election over two weeks out. • My prediction of a 60-seat majority for the Labour Party, which I repeated in a debate with Peter Kellner, Chairman of YouGov, on the BBC Today programme was challenged in a BBC World Service debate with the chairman of the MORI polling organization. • He wanted to bet me that his figure of a Labour majority of over 100 was a better estimate. I declined the bet and saved him some money. • The Labour majority was a little over 60 seats.

  17. US Senate 2006 Exchanges: All correct

  18. US Presidential Election 2008 • Prediction markets • BETFAIR state-by-state predictions: 49 out of 50 (called Indiana wrong).

  19. Indiana • Polls closed at 11.30 pm (UK time) in Indiana, a key state which McCain almost certainly needed to win to secure the Presidency. • McCain was marginal favourite to win Indiana on Betfair. • 11.45 pm: Obama becomes favourite to win Indiana, attracting significant sums to win from traders. • By this time CNN was calling just 1% of precincts in Indiana. • So what caused the shift to Obama on Betfair? • In retrospect, it seems that professional traders had latched on to the detail in the few published results. • Importantly, this shows the power of prediction markets in assimilating and processing new information very rapidly.

  20. Early Precinct Results • Stueben: Kerry 34%, Obama 42% • De Kalb: Kerry 31%, Obama 38% • Knox: Kerry 36%, Obama 54% • Marshall: Kerry 31%, Obama 50% • Only the most well-informed had accessed these results by 11.45, and knew what they meant, i.e. a big swing from Republican to Democrat since 2004, but Betfair traders were among them. Minutes later, the swing was confirmed in Vigo County.

  21. The election was called at 4am, but Betfair watchers knew before midnight! • At 4am, California was declared, giving Obama the final few electoral votes required to win the Presidency. • At 2.30am Ohio was called by most news networks. • Before midnight, the knowledge that Indiana was going to Obama, or at least that McCain would at best claim a small win there, was enough to indicate to Betfair watchers that the election was all but over. At 12.23 am, McCain was available at 25 to 1. • Meanwhile, Fox News declared that Indiana was over-polling for Obama because it shares a border with his home state of Illinois! • It was well past 3am when Fox News called the election for Obama. • Betfair 1, Fox News 0.

  22. 2010 UK General Election • It was the debates that lost it for the Conservatives!!! • Before the first debate, the markets predicted a Tory overall majority. • After the first debate, the markets never predicted anything other than a hung parliament! • While the polls swung all over the place, the markets barely flickered after that first debate in predicting a hung parliament with the Conservatives the largest party. • I commented on this at the time and forecast the outcome as polling and election betting analyst for Channel 4 News.

  23. How to Forecast an Election: Polls or Prediction Markets?

  24. Polls or Prediction Markets?Latest Research • Making use of a vast data set to conduct a forecast comparison exercise between polls and betting/prediction markets. • Assessing the two groups based on the same criteria: Accuracy based on past performance. • Accuracy: How often a forecast correctly predicts the election outcome.

  25. Sources of pre-election forecasts • 1. Opinion polls as collated by Pollster, Polling Report and Real Clear Politics, websites that collects historical and current polling information surrounding elections. • 2. Price data from online betting/prediction markets. • 3. Including price data from Betfair.

  26. Opinion polls • 19,277 observations from 394 different elections ranging from Presidential elections in 2004 and 2008 both at the national and state levels, Senate, Governor and House election and also the Democrat and Republican candidate selection primary elections.

  27. Betfair • Data on every matched contract for each party and candidate, in a range of elections, over the 900 days prior to the 2008 US Presidential election. • Similar data sets for the 2004 US election. • Well over 100,000 matched bets.

  28. Results • The analysis suggests that betting/prediction markets, and in particular Betfair, provide the most accurate forecasts.

  29. How to Forecast an Election • Look to the betting markets not to the polls. • Look in particular to Betfair. • Adjust upwards the win probability of the favourite, just a touch.

  30. US Election 2012 • Obama has been the market favourite for the last several months, but money has flowed toward Romney at key moments in the race. Reaction to the first presidential debate increased Romney's odds of winning the election, and just before the vice-presidential debate, on Oct. 11, Obama’s chance of winning was down to 62 percent from a post-convention high of around 80 percent. • Obama has enjoyed boosts from moments such as President Bill Clinton’s well-received Democratic National Convention speech and Romney’s much-criticized comment at a fundraiser that 47 percent of Americans are too dependent on government and see themselves as “victims”. • Following the third debate, Obama’s chance of winning stood at 68 to 69 percent — up from 65 percent before the debate.

  31. Where are we now? • Conventional wisdom is that the betting markets are pointing to the re-election of the President, while the polls apparently have had Mitt Romney as the narrow favourite since the first debate. • So what’s the evidence?

  32. Markets and polls • Betting markets – bookmakers, spread betting operators, Betfair converging at a probability of re-election of between 70% and 75%. • RealClearPolitics (RCP) average of national polls shows Romney leading for the last few days at up to a percentage point. • But RCP only publishes a selection of polls, excluding a number of very respectable polls for reasons best known to themselves. • Taking a more broad-based average of the national polls, such as published in Pollster, still has it close, but in the President’s favour. • An analysis of all the state polls shows Obama more comfortably ahead, including in the key swing states of Ohio, Virginia, Wisconsin, Colorado, Iowa, Nevada and New Hampshire.

  33. The state of play • Combining the national and state polls and giving them a similar weighting reveals Obama to be polling close to 2 per cent ahead of Romney – more in the state polls, less in the national polls. • Depending on the statistical analysis used to convert polling advantage into probability of winning, current polling evidence would seem to give Obama approaching an 80% win chance of winning (based on national and state polls) to over a 90% chance of winning (based on the state polls alone). • So the betting markets are in fact currently more conservative in their evaluation of President Obama’s chances than are the polls. • Why so?

  34. Why are the markets rating Obama’s chances a little less than are the polls? • 1. Favourite-longshot bias: For whatever reason, markets tend to under-estimate the true likelihood of the favourite winning, and vice-versa. It happens in horse racing, in football, in tennis, in boxing, in cricket – you name it - and in politics. • 2. The markets may be allowing for uncertainties a model can’t fully allow for – the current weather crisis and its potential effects, the numbers in the forthcoming jobs report, are examples. • 3. The markets may be wary that one of the outlier polling organisations might just possibly be right, and the polling consensus might just possibly be wrong. I term this the ‘Rasmussen effect’, after the polling company that has leaned Republican by a few points in the national and state polling for some time now. This was hugely in evidence in the 2010 congressional elections, and this is the pollster that had Bush up 9 per cent against Gore on election day 2000 – out by about ten points! • But what if Rasmussen is right, and most of the other polls are wrong?

  35. What if Rasmussen is right? • If Rasmussen is right, the race is very tight in the key states and Obama is much less of a favourite than the polling consensus or the markets suggest. • So Rasmussen must be right and most of the other polls wrong, and Obama must lose Ohio, and he must not win elsewhere to compensate. That would provide a route for Mr. Romney. All other routes are regarded as far more unlikely by the markets, though not impossible.

  36. What is the most likely outcome of the election? • Pretty much however you cut the data, the answer is the same. It’s a win for the President. • As a forecaster, now hope for the best, because even a candidate with a 75% chance of winning will lose 1 time in 4! • But polls don’t decide elections, and neither do the betting markets. • People decide elections, and that’s why anything can happen, and sometimes does. • But don’t bet on it! Perhaps!

  37. And How to Win One? • Example: The betting markets saw perhaps their best triumph of 2004 in Florida. Even though a number of polls put Kerry ahead in that state, or said the race was too close to call, the betting markets consistently showed Bush would win Florida comfortably. • If the Democrats had paid as much, or more, attention to the markets as the polls, I am convinced that the election result could have been different. They could have downsized their effort in Florida and focused their efforts more on other swing states where betting sites showed the race was much closer. • The stark conclusion is that political operatives need to pay more attention to the markets in planning their election strategy. If those close to Mr. Kerry had listened to this advice in 2004, the senator from Massachusetts would in my opinion have become (and perhaps still be) President Kerry. Any sophisticated campaign needs to make sure that it doesn’t repeat that mistake.

  38. In conclusion ... • ... Recent years have witnessed groundbreaking shifts in the way in which betting is taxed, regulated and perceived by economic theorists. • This means that betting markets will become more than just a major part of our future. Properly utilized they will be able to tell us what that future is likely to be! • We seem therefore to have created, almost by accident, a ‘high-tech’ crystal ball that taps into the accumulated expertise of mankind and makes it available to all. • The challenge now is to make the best use of it. Tonight is part of that!

  39. Prof. Leighton Vaughan Williams leighton.vaughan-williams@ntu.ac.uk