James Surowiecki's Disruptive Way Of Predicting Outcome Of Football Matches [Study this Insighful Sport Betting Thesis]
Research Shows A Simpler and Accruate Way to Predict Outcome of Football Matches - Better or Comparable to the well known "Aggregate Betting Odds" Model. This method is called: "The Wisdom of Crowds."
I recommend the study of this thesis to:
- Sport Betting Bookmakers and stakerholders
- Scholars of Sport Marketing and management study
- Inquisitive Fans who loves to expound their knowledge.
The Wisdom Of Crowds model of predicting the outcome of football matches resurface again at the school of sport business studies, Erasmus University Rotterdam. after James Surowiecki submitted and defended his thesis in 2004. A new group study submitted that the model truly a Simpler Way to Predict Football Matches, Better or Comparable To Aggregate Betting Odds
This thesis analyzes the prediction accuracy of the wisdom of crowds on the outcomes of football
matches. A simple prediction model, based on player valuations reported on the football statistics
website ‘www.transfermarkt.de’, is drafted and used to predict the league results of the Premier
League, Bundesliga and Primera División over the period 2013-2016. The resulting prediction
accuracy is compared with that of four benchmark mmodels:
ELO ratings,
Aggregated betting odds,
Home team wwins
Pure chance.
My analysis shows that the wisdom of crowds model is a better prediction method than "Pure chance" and the "Home team wins" model, is able to compete with "ELO ratings."
but is clearly a worse forecasting method than the model based on aggregated
betting odds. Furthermore, I investigated the presence of biases that potentially could influence the
valuations of users on ‘www.transfermarkt.de’. I find that cognitive biases are likely to influence user
responses and therefore could undermine the prediction accuracy of the corresponding model.
This could explain why the wisdom of crowds does not outperform Aggregated betting odds. Other
factors might distort the validity of user stated player valuations; further research should therefore
try and focus on examining the impact of these confounders on the valuations of Transfermarkt. Below is the Introduction of the thesis:
Introduction
On May 2nd 2016, the English football club Leicester City FC became the champions of the
English Premier League. At the beginning of the season in August, odds of 5000/1 were offered
for betting on Leicester City becoming the champions, amusingly the same odds were offered for
Elvis Presley being found alive in 2016. This gives an indication on how low the probability of
Leicester City becoming champions was deemed by the English bookmakers. But Leicester City
did win the title, which meant that English bookmakers had to pay out £25 million, the biggest
loss in British history on a single sporting market (Rayner & Brown, 2016).
The question that arises is whether the bookmakers could have foreseen Leicester City
becoming the champions and if they could have adjusted their odds more adequately? By
predicting match outcomes, bookmakers can calculate exactly how much money they will win by
setting the odds at certain levels. By this they have control over the bets that are placed. If the
bookmakers end up being wrong with their predictions and for example set the odds for a certain
outcome too high, they can lose a large amount of money. It is clear that bookmakers could
benefit from more accurate prediction models.
Sport outcome prediction has been studied extensively in the literature. In a study by Hvattum
and Arntzen, ELO ratings, which are originally used to estimate chess players’ strengths, are
used to predict football outcomes (2010). The authors found that their methods based on ELO
ratings can be used to predict football matches, but are not able to outperform the methods
using market odds. In a different study, linear regressions were estimated that allow for
predictions of matches played in the English Premier League (Louzada, Suzuki, & Salasar,
2014).
The aforementioned studies all base their models on factual objective data. But it is also
possible to make predictions using subjective data. A means of doing that is by making use of
the wisdom of the crowd. The idea of the wisdom of the crowd is that a group of people can
collectively decide on the right solution, while being more accurate than a few experts. The
wisdom of the crowd is not a new concept; it has been around as early as the 18th century
(Sunstein, 2006). A somewhat more recent example is a publication by Sir Francis Galton
(1907). He asked a crowd of people at an ox auction to estimate the weight of a certain ox.
Using all these estimations, he was able to construct a precise estimation of the ox’s weight. He
called this concept ‘vox populi’ or the wisdom of the crowd. The crowd’s average judgement
converged to the right solution. Since then four requirements for the wisdom of the crowd to
emerge have been formulated by Surowiecki (2004).
The wisdom of the crowd has been used to help predict sport outcomes. It has been found that
the mere recognition of tennis player names has an added benefit to predicting models (Herzog
& Hertwig, 2011). A second example is that crowd estimated market values have been used to
predict international football matches (Peeters, 2016).
Besides sport outcome prediction, the concept of crowd wisdom has been studied in many other
different fields. It has for example been applied to political science to forecast US presidential
elections (Murr, 2015). It can also be found in studies on the rationality of group decisions versus
individual decisions under risk (Baillon, Bleichrodt, Liu, & Wakker, 2016). Study shows that
decisions made by groups are more rational than decisions made by individuals (Charness &
Sutter, 2012). Furthermore, research has been done on crowd wisdom and prediction markets.
In the study by Wolfers and Zitzewitz it was found that crowds perform well in information
aggregation tasks (2004).
A good source of subjective football data is Transfermarkt (Transfermarkt, 2016a). This
community finds its origin back in 2000 when its first version went online. Since then
Transfermarkt has seen a large grow in the last couple of years resulting in a more extensive
database and even localized versions of their website (Transfermarkt, 2016b). Transfermarkt
allows its users to make estimations of each player’s market value. These crowd estimations are
combined into a single market value.
The question that arises is whether these crowd estimated market values can predict domestic
football matches and whether the wisdom of the crowd effect takes place with these market
values. As mentioned earlier, this could be very beneficial for parties using match prediction
models. But it could also give us another insight on the usability of the wisdom of the crowd. This
leads to the following research question:
RQ: Can crowd estimated market values help predict domestic football match outcomes?
This research focusses on the English Premier League. This league is the most watched football
league in the world with a TV audience of 4.7 billion back in 2013 (Ebner, 2013). It also a league
where a great amount of money is circulating as can be seen by the record British TV rights deal
worth £5.14 billion (Gibson, 2015). It has also been studied in previous research on match
prediction. Focusing on the same league allows me to compare findings of my research with the...
In case, you desire to study the full Thesis, Just WhatsApp me to upload it for you: +2348098772556
Best work you have done, this online website is really cool with great facts.
ReplyDeleteBandarQQ