Prediction market accuracy in the long run

Published on Apr 1, 2008in International Journal of Forecasting
· DOI :10.1016/J.IJFORECAST.2008.03.007
Joyce E. Berg16
Estimated H-index: 16
(UI: University of Iowa),
Forrest D. Nelson20
Estimated H-index: 20
(UI: University of Iowa),
Thomas A. Rietz21
Estimated H-index: 21
(UI: University of Iowa)
"Prediction markets" are designed specifically to forecast events such as elections. Though election prediction markets have been being conducted for almost twenty years, to date nearly all of the evidence on efficiency compares election eve forecasts with final pre-election polls and actual outcomes. Here, we present evidence that prediction markets outperform polls for longer horizons. We gather national polls for the 1988 through 2004 U.S. Presidential elections and ask whether either the poll or a contemporaneous Iowa Electronic Markets vote-share market prediction is closer to the eventual outcome for the two-major-party vote split. We compare market predictions to 964 polls over the five Presidential elections since 1988. The market is closer to the eventual outcome 74% of the time. Further, the market significantly outperforms the polls in every election when forecasting more than 100Â days in advance.
Figures & Tables
📖 Papers frequently viewed together
482 Citations
6,944 Citations
4 Authors (Joyce E. Berg, ..., Thomas A. Rietz)
174 Citations
#1Robin Hanson (GMU: George Mason University)H-Index: 24
A key question to ask about any social institution is how well it generates, aggregates, and distributes information. Speculative markets seem to do well at this, while familiar democratic institutions, relying in part on academic institutions, seem to fail in many ways. So perhaps we should consider “futarchy,” a form of government where betting markets become our primary common source on matters of fact. Democracy would say what we want, while speculators would say how to get it. That is, elec...
60 CitationsSource
#1Robert S. Erikson (Columbia University)H-Index: 52
#2Christopher Wlezien (University of Texas at Austin)H-Index: 47
Election markets have been praised for their ability to forecast election outcomes, and to forecast better than trial-heat polls. This paper challenges that optimistic assessment of election markets, based on an analysis of Iowa Electronic Market (IEM) data from presidential elections between 1988 and 2004. We argue that it is inappropriate to naively compare market forecasts of an election outcome with exact poll results on the day prices are recorded, that is, market prices reflect forecasts o...
111 CitationsSource
#1Vernon L. SmithH-Index: 103
41 Citations
#1Randall J. JonesH-Index: 8
#2J. ArmstrongH-Index: 5
Last. Alfred G. CuzánH-Index: 10
view all 3 authors...
Prior research offers a mixed view of the value of expert surveys for long-term election forecasts. On the positive side, experts have more information about the candidates and issues than voters do. On the negative side, experts all have access to the same information. Based on prior literature and on our experiences with the 2004 presidential election and the 2008 campaign so far, we have reason to believe that a simple expert survey (the Nominal Group Technique) is preferable to Delphi. Our s...
4 Citations
#1James E. Campbell (UB: University at Buffalo)H-Index: 27
In the 2004 presidential election, 121.1 million ballots were cast for the major party candidates. President George W. Bush received 51.24% of these votes to Senator John Kerry's 48.76%. As Bush prepares for his second term, we know how the candidates performed. How did the election forecasters do?
19 CitationsSource
#1Steven Kou (Columbia University)H-Index: 24
#2Michael E. Sobel (Columbia University)H-Index: 28
The dominant methodology for short-term forecasting of electoral outcomes uses trial-heat polls, where respondents report their current electoral preferences (not their election-day predictions). Election markets, where self-selected participants trade shares of candidates at prices predictive of election-day results, provide an alternative method that often produces more accurate forecasts. Consequently, increasing attention is being paid to this methodology. However, it is poorly understood an...
39 CitationsSource
The Iowa Electronic Markets are specially designed futures markets that appear to aggregate information efficiently to predict events such as election outcomes. Yet, in theory, perfect information aggregation is impossible. Further, the markets are populated by a nonrepresentative sample of mistake-prone and biased traders. That is, traders are prone to the behavioral anomalies predicted by behavioral finance. How can this be reconciled with market efficiency? Here, we take a first step by analy...
134 CitationsSource
#1Joyce E. Berg (UI: University of Iowa)H-Index: 16
#2Thomas A. Rietz (UI: University of Iowa)H-Index: 21
Valuations from “prediction markets” reveal expectations about the likelihood of events. “Conditional prediction markets” reveal expectations conditional on other events occurring. For example, in 1996, the Iowa Electronic Markets (IEM) ran markets to predict the chances that different candidates would become the Republican Presidential nominee. Other concurrent IEM markets predicted the vote shares that each party would receive conditional on the Republican nominee chosen. Here, using these mar...
195 CitationsSource
#1Charles R. Plott (California Institute of Technology)H-Index: 69
The academic literature, Wall Street commentary, and even daily news reporting reflect a belief that markets can anticipate events. The current movements in the stock market are interpreted as reflecting the likelihood that the Federal Reserve System will change interest rates. Futures markets are interpreted as reflecting the best estimates of things to come. Movements in individual stock prices are thought to anticipate earnings reports or the probability of a merger or buyout. If we want to k...
157 CitationsSource
#1James R. CampbellH-Index: 1
#2James C. GarandH-Index: 28
PART ONE: INTRODUCTION Forecasting US National Elections - James E Campbell and James C Garand PART TWO: FORECASTS OF THE 1996 AMERICAN NATIONAL ELECTIONS Polls and Votes - James E Campbell The Trial-Heat Presidential Election Forecasting Model, Certainty, and Political Campaigns Bill and Al's Excellent Adventure - Alan Abramowitz Forecasting the 1996 Presidential Election Of Time and Candidates - Helmut Norpoth A Forecast for 1996 The Future in Forecasting - Michael Lewis-Beck and Charles Tien ...
103 Citations
Cited By231
#1Oliver MerzH-Index: 1
#2Raphael Flepp (UZH: University of Zurich)H-Index: 5
Last. Egon Franck (UZH: University of Zurich)H-Index: 30
view all 3 authors...
Abstract This paper examines whether individuals’ decision making is affected by fast-sounding horse names in a betting exchange market environment. In horse racing, the name of a horse does not depend on the horse's performance and is thus uninformative. If positive affect towards fast-sounding horse names is present, we expect less accurate prices, i.e., winning probabilities and lower returns due to the increased demand for these bets. Using over 3 million horse bets, we find evidence that th...
#1Andrea Albertazzi (University of Essex)
#2Friederike Mengel (University of Essex)H-Index: 17
Last. Ronald Peeters (University of Otago)H-Index: 55
view all 3 authors...
Theoretical and experimental literature have provided mixed insights on the ability of financial markets to perfectly aggregate private information into asset prices. We conduct an experiment designed to benchmark information aggregation in markets. In our lab experiment, we randomly assign subjects to different institutional environments, either a market or a Becker–DeGroot–Marschak mechanism. We find evidence that market interaction is worse for information aggregation. The difference between ...
#1Yaser Abolghasemi (UW: University of Waterloo)
#2Stanko Dimitrov (UW: University of Waterloo)H-Index: 10
This is the peer reviewed version of the following article: Abolghasemi, Y, Dimitrov, S. Determining the causality between U.S. presidential prediction markets and global financial markets. Int J Fin Econ. 2020; 1– 23., which has been published in final form at This article may be used for non-commercial purposes in accordance with Wiley Terms and Conditions for Use of Self-Archived Versions.
#1Mirta GalesicH-Index: 33
#2Wändi Bruine de Bruin (SC: University of Southern California)H-Index: 44
Last. Tamara van der Does (SFI: Santa Fe Institute)H-Index: 5
view all 9 authors...
The ability to 'sense' the social environment and thereby to understand the thoughts and actions of others allows humans to fit into their social worlds, communicate and cooperate, and learn from others' experiences. Here we argue that, through the lens of computational social science, this ability can be used to advance research into human sociality. When strategically selected to represent a specific population of interest, human social sensors can help to describe and predict societal trends....
6 CitationsSource
#1Spyros Galanis (City University London)H-Index: 7
#2Stelios Kotronis (City University London)H-Index: 1
The ability of markets to aggregate information through prices is examined in a dynamic environment with unawareness. We find that if all traders are able to minimally update their awareness when they observe a price that is counterfactual to their private information, they will eventually reach an agreement, thus generalising the result of Geanakoplos and Polemarchakis [1982]. Moreover, if the traded security is separable, then agreement is on the correct price and there is information aggregat...
2 CitationsSource
#1Dian Yu (SJTU: Shanghai Jiao Tong University)
#2Jianjun Gao (SUFE: Shanghai University of Finance and Economics)H-Index: 11
Last. Tongyao Wang (SJTU: Shanghai Jiao Tong University)
view all 3 authors...
The past several decades have witnessed a growing enthusiasm to implement the market as an information aggregation tool. However, some deep-rooted pricing anomalies frequently observed in various markets loom large, questioning the rationality of such eagerness. One of these anomalies is the favorite-longshot bias, and sometimes its opposite, the reverse favorite-longshot bias. We propose an equilibrium model consisting of traders with cumulative prospect theory-based utility and heterogeneous b...
This thesis contains three chapters employing controlled economic experiments. The first chapter studies to what extent laboratory measures of cheating generalize to the field. I present a novel measure that allows for individual level observations of cheating, and I relate decisions made in this laboratory task with individual choices taken in the field, where subjects can lie by mis-reporting their experimental earnings. According to this new measure, no correlation of behaviour between the la...
#1Wenlong Wang (Massey University)
#2Thomas Pfeiffer (Massey University)H-Index: 29
Decision markets are mechanisms for selecting one among a set of actions based on forecasts about their consequences. Decision markets that are based on scoring rules have been proven to offer incentive compatibility analogous to properly incentivised prediction markets. However, in contrast to prediction markets, it is unclear how to implement decision markets such that forecasting is done through the trading of securities. We here propose such a securities based implementation, and show that i...
#1Ahrash Dianat (University of Essex)H-Index: 1
#2Christoph Siemroth (University of Essex)H-Index: 6
How should corporate prediction markets be designed to improve manager decisions? We conduct a lab experiment to investigate an important corporate prediction market setting: A manager needs information about the state of a project, which workers have, in order to make a state-dependent decision. Workers can potentially reveal this information by trading in a corporate prediction market. We test two different market designs to determine which provides more information to the manager and leads to...
1 CitationsSource
#1Ke Wu (ETH Zurich)H-Index: 5
#2Spencer Wheatley (ETH Zurich)H-Index: 11
Last. Didier Sornette (ETH Zurich)H-Index: 107
view all 3 authors...
Exploiting the near-experimental conditions provided by the GBPUSD exchange rate during the Brexit vote of 2016, we quantify a significant delay of the market price in reflecting the increasing pro...