Alasdair Brown

Information Processing Constraints and Asset Mispricing

I analyse a series of natural quasi‐experiments – centred on betting exchange data on the Wimbledon Tennis Championships – to determine whether information processing constraints are partially responsible for mispricing in asset markets. I find that the arrival of information during each match leads to substantial mispricing between two equivalent assets, and that part of this mispricing can be attributed to differences in the frequency with which the two prices are updated inplay. This suggests that information processing constraints force the periodic neglect of one of the assets, thereby causing substantial, albeit temporary, mispricing in this simple asset market.

Information Acquisition in Ostensibly Efficient Markets

I use U.K. betting exchange data on Wimbledon tennis matches to investigate the Grossman and Stiglitz (1980) paradox. Risk-free arbitrage opportunities arise frequently during matches (as information arrives and asynchronously shifts prices), but seldom arise before matches (when there is little information to move prices). I find that on the few occasions that arbitrage opportunities do arise before matches, they last substantially longer than average. This suggests, in line with the paradox, that traders neglect to acquire information (i.e. carry out research, or watch markets) if they believe that markets are already efficient. This neglect, in turn, makes markets inefficient.

FORECASTING WITH SOCIAL MEDIA: EVIDENCE FROM TWEETS ON SOCCER MATCHES

Social media is now used as a forecasting tool by a variety of firms and agencies. But how useful are such data in forecasting outcomes? Can social media add any information to that produced by a prediction/betting market? We source 13.8 million posts from Twitter, and combine them with contemporaneous Betfair betting prices, to forecast the outcomes of English Premier League soccer matches as they unfold. Using a microblogging dictionary to analyze the content of Tweets, we find that the aggregate tone of Tweets contains significant information not in betting prices, particularly in the immediate aftermath of goals and red cards. (JEL G14, G17)

Have Betting Exchanges Corrupted Horse Racing

Betting exchanges allow punters to bet on a horse to lose a race. This, many argue, has opened up the sport to a new form of corruption, where races will be deliberately lost in order to profit from betting. We examine whether anecdotal evidence of the fixing of horses to lose—of which there are many examples—is indicative of wider corruption. Following a “forensic economics” approach, we build an asymmetric information model of exchange betting and take it to betting data on 9,560 races run in 2013/2014. We find no evidence of the widespread corruption of horse racing by the betting exchanges.

The Wisdom of Amateur Crowds: Evidence from an Online Community of Sports Tipsters

We analyse the accuracy of crowd forecasts produced on Oddsportal, an online community of amateur sports tipsters. Tipsters in this community are ranked according to the betting return on their tips, but there are no prizes for accuracy. Nevertheless, we find that aggregated tips in this community contain information not in betting prices. A strategy of betting when a majority predict an outcome produces average returns of 1.317% for 68,339 events. The accuracy of these forecasts stems from the wisdom of the whole crowd, as selecting sections of the crowd based on experience or past forecast accuracy does not improve betting returns.

Framing Effects and Market Selection: Evidence from Betting Markets

We collect data on 75 million GBP of tennis bets over a 6 year period to analyse whether participants in high-stakes environments recognise simple framing differences. The structure of this market means that we can place the same bet at the same time in two different ways. These two isomorphic bets are framed differently, and often priced differently. We find that bettors make frequent mistakes, choosing the worse of the two bets in 35% of cases. However, bettors who choose the inferior price earn higher returns from their bets, suggesting that their effort has been focused on fundamental information acquisition rather than bet execution. The net result is that market selection may, if anything, slightly favour those who are unable, or unwilling, to recognise framing differences.

Combining Prediction Markets and Forecasting Contests

Two popular methods for aggregating individual forecasts are prediction markets, where participants bet on the outcome of future events, and forecasting contests, where participants are ranked according to the accuracy of their forecasts. Can these methods be used in concert to produce more accurate forecasts? We analyse 1.79 million forecasts on oddsportal.com, a social network for sports tipsters. Tipsters are ranked according to the betting return on their tips. We find that an aggregation of these tips predicts sporting outcomes, after controlling for betting/prediction market prices. Rank-order forecasting contests, even without tangible prizes, are useful tools for eliciting crowd forecasts.

The Wisdom of Large and Small Crowds: Evidence from Repeated Natural Experiments in Sports Betting

Prediction markets have proved excellent tools for forecasting, outperforming experts and polls in many settings. But do larger markets, with wider participation, perform better than smaller markets? In this paper we analyse a series of repeated natural experiments in sports betting. The Queens Club Tennis Championships are held every year, but every other year the Championships clash with a major soccer tournament. We find that tennis betting prices become significantly less informative when participation rates are adversely affected by the clashing soccer tournament. Larger markets perform better, in part, because of the higher returns they offer for informed trading.

Anchoring and Manipulation in Speculative Markets: A Field Experiment

A common finding in laboratory studies is that subjects anchor on irrelevant initial cues when valuing assets. We run a field experiment to examine whether this heuristic can be exploited to manipulate prices in real markets. We provide early quotes in a series of horse race betting markets, and randomly vary whether these quotes are high or low. We find that subsequent prices are indeed distorted in the direction of our anchor, and that the effect spills over into correlated markets. Furthermore, we show that there are positive returns to some market manipulation strategies predicated on exploiting the anchoring heuristic.

Selection and incentives in contests: evidence from horse racing

ABSTRACT The designer of internal labour market promotion contests must balance the need to select the best candidate with the need to provide incentives for all candidates. We use an extensive data set from horse racing – where there is abundant variation in contest design features – to analyse if there are particular features that help to achieve these two objectives. We find that contests with higher prize money and fewer participants are the most successful at achieving the dual remit of selection and incentives.