Is finding value in overlooked outliers a feasible betting strategy? How could such an outlier be found? This article discusses the opportunities and difficulties presented by searching for overlooked teams in the betting market.
Outliers in theory
A topic I’ve written about before is the importance of information in betting. Prices are made up of the sum of information available to the market, so how can an individual bettor gain an edge?
Uncovering overlooked outliers is an intuitive concept for aspiring bettors, and makes sense in theory. Bookmakers cover a vast range of sports and events that they must be spread thinly, versus a bettor who can focus on one.
In general the market is efficient but an optimistic bettor tells himself he only needs to find one opportunity whilst the bookmaker only needs to overlook one aspect of a fixture to provide one.
In a market proven to be highly efficient how realistic is this?
The case for outliers: Smart people miss things
The world has provided us with examples of very smart people doing seemingly not very smart things. Often something that seems obvious in hindsight is seemingly missed by the majority of market participants, even when they are strongly incentivised to do so.
Famously, former NBA player Jeremy Lin was a possible victim of a lot of seemingly smart people doing a not-so-smart thing.
Lin, despite being an NBA level talent, somehow received no Division 1 basketball scholarship offers. He then starred as a walk-on at Harvard only to go undrafted in the NBA draft. This despite statistics that pointed towards him being a high value pick.
In Michael Lewis’ book “The Undoing Project” Houston Rockets GM Daryl Morey talks about how he opted against drafting Lin even though he scored highly enough on the Rockets’ player models to be a possible first round pick. This was because Lin was considered to be “unathletic”. This turned out to be an example of smart a man doing an unsmart thing.
“A year after the Houston Rockets failed to draft Jeremy Lin, they began to measure the speed of a player’s first two steps: Jeremy Lin had the quickest first move of any player measured. He was explosive and was able to change direction far more quickly than most NBA players”.
Lin was incredibly athletic. Preconceived notions based on his appearance and unconventional path to the NBA masked this from intelligent people incentivised to make the best decisions possible when recruiting basketball players, even when they had the data to prove Lin’s ability in front of them.
Since the probability of an event occurring is driven by the sum of information available to that market (with some sources of information known to be more valuable than others) is it possible such inefficiencies present, until recently, at the top level of sport are there to be found in the World of betting?
Finding an outlier: Did the market miss Kosovo?
Kosovo became a FIFA member in May 2016, allowing them to enter their national team into official competitions. Their first official competitive fixtures played out how largely how we would expect team ranked as one of the worst in the World by FIFA to play out. They had one surprise draw versus Finland in their first ever game followed by nine consecutive defeats, including a thrashing by Croatia.
A glance at Kosovo’s record and World ranking would cause many people to rank them as one of the worst teams in Europe.
That glance would have missed out some vital information, however. For a new team, that was Europe’s youngest and yet to play an official game with true home advantage, it was decidedly unfair to rank Kosovo alongside the likes of Malta and Andorra.
One glance at the talent in the squad was enough to see that they had a higher ceiling than simple statistics suggested. This became especially apparent when their UEFA Nations League group was drawn:
|Team||Transfermrkt valuation of squad (millions)|
Despite their significant market value edge over the rest of the teams, Kosovo were available as high as 2.75 to win the group behind favourites Azerbaijan.
This was reflected in the odds of the initial matchups in the group. In their home games against the Faroe Islands (Pinnacle odds):
Kosovo to win odds vs. Faroe Islands 1.61 (home)
Azerbaijan to win odds vs. Faroe Islands 1.53 (home)
Based on market value alone Kosovo were very similar to the Hungary team that hosted the Faroe Islands during World Cup qualifying. Hungary were priced at 1.33 to win that match despite in theory possessing a squad no stronger than Kosovo’s.
In this case it is possible Kosovo were a Jeremy Lin style outlier. Without any further inspection they looked and ranked like a bad team, just as a glance at Lin gave scouts a false impression about his athletic ability.
Could the difference in price between Kosovo and similar or inferior teams be due to the market underestimating the strength of their team? Or was it simply a case of Kosovo’s relative lack of experience being built into the market price?
The case against outliers: Why Kosovo’s potential value is impossible to judge
The case for Kosovo being an overlooked outlier is reasonably compelling. A fresh team priced below their ability levels intuitively feels like a strong value proposition.
The major issue is that it is impossible to judge whether there was value present here, let alone the extent of that value. The same scenario would need to be rerun thousands of times to get a statistically significant judgement of their valuation, even if they are successful in the short run.
In a similar way, whilst it is very easy to say Jeremy Lin was overlooked at college level, his success is not a big enough sample size to suggest there are other Lin like prospects of the same level now working office jobs.
In fact, if he had picked up a major injury in college or flunked out of the NBA prior to Linsanity we would not be talking about him at all. Kosovo, like Lin, could just be classic examples of survivorship bias.
Selection bias is impossible to avoid here then. The outliers we remember are the ones who succeed -we forget about the Kosovos that lose or the Lins that never make it.
Stabbing in the dark: Working with what we have
Whilst acknowledging the difficulty involved with assessing the value of outliers, I feel it would be unadvisable for bettors to ignore them completely.
In the case of Lin the resulting data showed he was indeed an overlooked player. This was demonstrated by the fact that the Rocket’s approach to scouting players, which showed Lin was a miss, was soon adopted by the rest of the league.
NBA teams were forced to reassess their valuation of players in a competitive market, suggesting undervalued players like Lin had been overlooked previously and the Rockets had indeed found an edge.
The existence of a possible oversight the betting market may have made on a team like Kosovo this is more difficult to determine, despite their positive results.
However, even if it is almost impossible to quantify what kind of edge a bettor can expect on a team like Kosovo that does not necessarily mean such an edge does not exist.
This does make it very difficult to incorporate such a bet into a staking strategy since it is difficult to ascertain what kind of edge you have.
Whilst this makes returns difficult to model, finding outlier events and teams similar to the Kosovans, such as the Mayweather vs. McGregor bout, could represent a viable but frustratingly unquantifiable and irregular betting strategy. Even in an overall efficient market.
Unfortunately, in sports as in betting, if such edges exist they soon disappear. As Morey puts it “once every team knows about a market mechanism, it’s gone”.
Any bettors that feel they have found a potentially lucrative outlier would be wise to exploit it whilst it still exists.