Preface to 2008 Edition

Academic Press published the first edition of this volume in 1994. Our goal was
to collect the important papers and many new articles for an up-to-date and
comprehensive volume on racetrack efficiency. The ensuing years have pushed
the volume into classic status. The volume went out of print at the same time
that various hedge fund-like racing syndicates were using these ideas as an
integral part of their research that led to hundreds of millions in profits for
the most successful groups and about US$10 billion in total gains. When the
volume went out of print, it became a cult item. Its scarcity, its content, and the
dreams of new investment groups drove the price of the volume on eBay and
Amazon to thousands of dollars. Hence, the volume was only available to a
few. We appreciate the classic and cult status but want to make the book more
accessible to students, researchers, academics and others. So, we are pleased
that World Scientific is producing this 2008 edition for this purpose. Except for
this Preface, the 2008 edition is identical to the original.
This 2008 edition is coming out simultaneously with the Handbook of
Sports and Lottery Markets edited by Donald B Hausch and William T Ziemba
published by ElsevierNorth Holland, 2008. That volume complements this 2008
edition by providing new, original surveys of the racetrack, sports betting and
lotto investment field. Together they provide an up-to-date treatment of the field
of racetrack efficiency plus the key historical and current papers.
There are a number of new developments since 1994 that have altered the
racetrack investment markets, These affect the academic research surveyed in
this 2008 edition and the Handbook volume as well as the professional and
amateur application of these results to racetrack markets across the world.
Since circa 2000, we have seen the following important changes in racetrack
markets.
1. Rebates on wagers are available to essentially all bettors. These rebates
began as simply a reward to large bettors for quantity purchases, similar to
discounts when one buys in bulk such as a case of San Pellegrino offered at
food stores and other places. However some rebaters have extended these
discounts to all bettors. The tracks provide a signal of the pools to the
rebaters at a low price. Then the rebaters put the bets of their clients into the
track pools, with the rebaters and the bettors sharing the difference between
the regular track take and the signal price. So bettors, instead of paying a
track take of 13-30% at the track on various wagers, actually face a net track
take of about 10%. This lower track take for syndicates disadvantages small

...
Xlll

Electronic
Electroniccopy
copyavailable
availableat:
at:https://ssrn.com/abstract=1469950
http://ssrn.com/abstract=1469950
xiv PREFACE TO 2008 EDITION

bettors, though, since their effective track take becomes more than the posted
track take.
2. The extent of off-track betting from other racetracks and other betting
sources has expanded greatly and is now about 87% of all wagers at a typical
race track (see Hausch and Ziemba, 2008). There are also delays i n adding
these monies to the host track pools, which can lead to substantial
differences between the odds posted at the time the race begins and the final
odds that determine the actual payoffs. Indeed, about 50% of all bets are not
shown in the pools until after it is too late to bet at the host track and the
horses are running (see Ziemba, 2008). Hence, a 4-1 horse at 1 minute Lo
post time could end up paying 5-2. This uncertainty is challenging for any
system that incorporates the public’s odds. An example is the Dr Z place
and show system, which is discussed in two Hausch and Ziemba (one with
Rubinstein) papers in Section V of this volume and which estimates a
horse’s win probability based on the public’s win odds. Ziemba’s (2008)
update of this earlier work demonstrates the challenge, but also shows that,
with rebates, the system may still yield profits.
3. The 1999 introduction of Betfair and other betting exchanges allows bettors
to lock in prices for win, place and show and other wagers by betting
not against the house but against other bettors. The house, say Betfair of
London (the biggest by far), then takes a 1-5% commission on all net
winning wagers. So bettors can lock in odds rather than bear the risk of
changing odds in the on-track pasimutuel betting system. Equally important,
the betting exchanges allow short as well as long bets. Hence, bettors can
wager against over-priced horses just as hedge fund managers can short
over-priced stocks, stock index futures, currencies or precious metals. This
feature also allows for long-short hedging akin to long-short equity hedge
funds. While these exchanges are not legal in the US, they have had an
impact on some important US horse races. See Snowberg, Wolfers and
Zitzewitz (2008) and Smith and Vaughan Williams (2008) for a discussion
of prediction markets in politics and other areas and betting on myriad sports
and other events, respectively.
4. The figure below shows the return from wagers on different odds-levels.
The steeper of the two lines is from Ziemba and Hausch ( I 986) and is based
on data from the 1980s and earlier. The flatter line is from Ziemba (2004)
and is based on Equiform data from 1997 to 2004. Both lines are based on
data of more than 300,000 horses, and both exhibit the favorite-longshot bias
where favorites are underbet relative to the true odds and longshots are

Electronic
Electroniccopy
copyavailable
availableat:
at:https://ssrn.com/abstract=1469950
http://ssrn.com/abstract=1469950
 PREFACE TO 2008 EDITION xv

overbet. However, the extent of the bias is clearly weaker in recent years, so
the expected return on longshots is higher and the expected return on
favorites is lower. Further, the positive expected return on extreme favorites
is no longer available. Betting exchanges appear to partly explain this recent
shift. Snowberg and Wolfers (2008) show similar results based on horses in
647,903 races, which are all races run in the US from 1992 to 2001.

Fa, 100
2
2
Y 80
u
m
Q
x
m
L 60

Since the original papers were published, some have been extended. Some
of these extensions are in Vaughan Williams (2003, 2005), two books we
strongly recommend, and Hausch and Ziemba (2008). A few of them are now
discussed.
Reference 3 1 in the annotated bibliography was published as Lo and Bacon-
Shone (2008). They provide a discount model that is a simple approximation
to the Henery and Stern models for computing ordering probabilities superior
to those obtained by the Harville formulas. They show the accuracy of the
approximation and the method is applied in Lo, Bacon-Shone and Bushe (1995).
The MacLean, Ziemba and Blazenko paper was extended by MacLean and
Ziemba (2006) and an updated table and discussion of the good and bad
properties of the Kelly criterion appear in Ziemba and Ziemba (2007).

Electronic copy available at: https://ssrn.com/abstract=1469950

xvi PREFACE TO 2008 EDITION

The crosstrack betting paper by Hausch and Ziemba has evolved into a
procedure for going long on horses with rebates at the host track and short on
Betfair or other betting exchanges plus long/shorts across betting exchanges.
An extension of Ziemba and Hausch ( 1 987) on the Dr Z place betting system
in England arises because of the different way place pools were computed in
England and other locales by splitting the net bets among the collecting horses.
Hence, once about 24.5% is bet on any horse, the payoff is at its minimum. We
noted this, but Jackson and Waldron (2003) extended Hausch and Ziemba’s
system for locks to exploit this rule. See also Edelman and O’Brian (2004) for a
more general analysis of these tote arbitrages.
We hope that the readers of this 2008 edition and the Hausch and Ziemba
(2008) Handbook find these papers stimulating and useful.

Donald B. Hausch, Madison

Victor S. Y. Lo, Boston
William T. Ziemba, Vancouver
February 2008

References
Edelman, D. C. and N. G. O’Brian (2004). Tote arbitrage and lock opportunities
in racetrack betting. European Journal of Finance, 10: 370-378.
Hausch, D.B. and W.T. Ziemba (Eds.) (2008). Handbook of Sports and Lottery
Markets. Elsevier.
Jackson, D. and P. Waldron (2003). Parimutuel place betting in Great Britain
and Ireland, in L. Vaughan Williams (Ed.), The Economics of Gambling,
18-29.
Lo, V. and J. Bacon-Shone (2008). Approximating the ordering probabilities of
multi-entry competitions by a simple method, in D.B. Hausch and W.T.
Ziernba (Eds.), Handbook of Sports and Lottery Markets. Elsevier.
Lo, V., J. Bacon-Shone and K. Bushe (1995). The application of ranking
probability models to racetrack betting, Management Science, 41:
1048-1059.
MacLean, L.C. and W.T. Ziemba (2006). Capital growth: Theory and practice,
in S.A. Zenios and W.T. Ziemba (Eds.), Handbook of Asset-Liability
.Management, Volume I: Theory and Methodology, pp. 429473. Elsevier.

Electronic copy available at: https://ssrn.com/abstract=1469950

 PREFACE TO 2008 EDITION xvii

Smith, M.A. and L. Vaughan Williams (2008). Betting exchanges: A

technological revolution in sports betting, in D.B. Hausch and W.T.
Ziemba (Eds.), Handbook of Sports and Lottery Markets. Elsevier.
Snowberg, E. and J. Wolfers (2008). Examining explanations of a market
anomaly: Preferences or perceptions?, in D.B. Hausch and W.T. Ziemba
(Eds.), Handbook of Sports and Lottery Markets. Elsevier.
Snowberg, E., J. Wolfers and E. Zitzewitz (2008). Prediction markets: From
politics to business, in D.B. Hausch and W.T. Ziemba (Eds.), Handbook of
Sports and Lottery Markets. Elsevier.
Vaughan Williams, L. (2003). The Economics of Gambling. Routledge.
Vaughan Williams, L. (2005). Information Efficiency in Financial and Betting
Markets. Cambridge University Press.
Ziemba, W.T. (2004). Behavioral finance, racetrack betting and options and
futures trading, Mathematical Finance Seminar, Stanford University,
January 30.
Ziemba, W.T. (2008). Efficiency of racetrack, sports and lottery betting markets,
in D.B. Hausch and W.T. Ziemba (Eds.), Handbook of Sports and Lottery
Markets. Elsevier.
Ziemba, W.T. and, D.B. Hausch (1986). Betting at the Racetrack. Dr Z
Investments Inc, San Luis Obispo, CA.
Ziemba, W.T. and D.B. Hausch (1987). Dr Z’s Beat the Racetrack. William
Morrow.
Ziemba, R.E.S. and W.T. Ziemba (2007). Scenarios ,for Risk Management and
Global Investment Strategies. Wiley.

Electronic copy available at: https://ssrn.com/abstract=1469950

 This page intentionally left blank

Electronic copy available at: https://ssrn.com/abstract=1469950

 Preface in 1994 Edition

Racetrack betting interests not just gamblers and those seeking casual
entertainment, but also academics across many disciplines. Gamblers and
investors certainly enjoy the sport, but making profits is their challenge.
Academics are interested in this came challenge. Their approach is to determine
whether the racetrack betting market is efficient and, if not, whether positive
risk-adjusted returns are possible. In addition to market efficiency, active
research areas include methods for analyzing racetrack data, determining
probabilities of complex events, explaining gambling behavior, and developing
betting strategies. Racetrack betting is of particular interest because, beyond just
allowing a forum for theoretical questions of efficiency and gambling behavior,
abundant and easily-accessible data from racetracks around the world permits
extensive empirical consideration of these questions.
Studies related to racetrack betting have appeared in the academic literature
for over 40 years. The number of articles in this area has been growing recently.
A need for a book-length treatment of racing efficiency research covering
its economic, psychological, financial, statistical and mathematical aspects
has been felt for quite some time. This volume of collected papers is the
result of collaboration with many colleagues in a variety of disciplines and
from countries around the world. This volume is intended for two audiences:
academics across many fields who are interested in the subject and those of the
general public who wish to know more about the academic research on racetrack
betting.
Following an introduction, the papers in this volume are divided into
seven sections: psychological studies, utility preferences of racetrack bettors,
economic and mathematical insights, efficiency of win markets and the favorite-
longshot bias, place and show anomalies, efficiency of exotic wagering markets,
and research in the Commonwealth and Asia.
We are pleased to be part of this distinguished series on Economic Theory,
Econometrics, and Mathematical Economics, and we thank Academic Press for
the opportunity to publish this volume.

Donald B. Hausch, Madison

Victor S. Y. Lo, Vancouver
William T. Ziemba, Vancouver

xix

Electronic copy available at: https://ssrn.com/abstract=1469950