One of the best accepted phenomena in betting is the favourite-longshot bias, where bettors undervalue favourites and overvalue ‘longshots’.
Yet, whilst many in the betting industry are aware of the favourite-longshot bias, working out exactly why this occurs is less clear. 60 years of academic analysis has yet to get to the bottom of it, but one thing is for sure- the answer likely lies in psychology of the punters themselves.
Behavioural psychology indicates that the favourite-longshot bias is likely a result of one of two things:
1. Punters are risk seekers
A ‘risk seeker’ is a person who values the ‘happiness’ they would get from winning a dollar more than the happiness they would lose from losing a dollar. People that bet on horses, buy lottery tickets would fall into this category. If you are happy to flip a coin to see who buys the next round of beers, then this is likely you!
At the shallow end of the risk seeking scale you have people that are willing to place small bets on favourites, while at the other end you have the extreme risk seekers- those that are happy to accept the lower probability of winning offered by longshots in return for higher potential gains. People that bet on large accumulators or ‘parlays’ would fall into this category as they are happy to accept a much larger house-edge for higher potential returns
Of course If we look at the general population, you would see a mix of risk seeking and risk adverse people and as such it is unclear whether humans as a whole suffer from this bias, or just those that bet. But we know that bettors, by definition are more risk seeking than most and are thus more susceptible to the favourite-longshot bias.
2. People assign a higher probability to events they remember happening
If you have ever been dealt a royal flush in poker, count yourself lucky- the odds of that happening on any hand is about 1 in 30,000. But if it has happened to you, you likely remember it well. The same is true of winning longshot bets. Simly put, you are likely to remember a higher proportion of your 100-1 wins than your 100-1 losses.
Remembering rare events better than each instance of a common event presents a problem for punters due to what is known as the ‘availability heuristic’. The availability heuristic is essentially a mental shortcut that relies on immediately recallable examples when evaluating a topic or choice. It operates on the notion that if something can be easily recalled, humans will believe it is more likely to happen again, than things that cannot be as easily recalled.
The consequence of the availability heuristic is that a person who more easily recalls their 100-1 wins than their 100-1 loses is likely to systematically overestimate the frequency that they picked an longshot to win in the past. This would then lead to an overestimation of the punter’s perception of their ability to pick longshot winners, and make a longshot bet appear more appealing.
Proof of the Favourite-Longshot Bias
Whilst it’s impossible to prove directly that people are biased towards longshots, we can use the difference in returns on favourites and underdogs as a good indicator of this bias. The way this works is as follows:
- In order to ensure a balanced book, bookmakers will set odds that take into account how punters will bet. If a bookmaker has even odds you can expect that there is a roughly even distribution of bets, and if there is a strong favourite can expect a higher proportion of bets on the favourite.
- Rational punters will make bets only that they expect to win money on. If the market thinks a team winning is good value (ie a $10.00 bet has more than 10% chance of winning) then they will bet on it. If enough punters do that the odds will shorten for that team winning and they will pay less than $10.
- Thus if the favourite-longshot bias were true, we would expect bets on longshots to be worse value than favourites (ie have a lower expected return).
A large number of studies prove exactly that. But one of the best is a paper by Andrew Zeelte (2012) that looks at over 40,000 Men’s tennis matches between 2003-2011. His findings are converted into decimal odds and tabulated below:
|Less than $1.11||98%|
|$1.11 - $1.25||97%|
|$1.25 - $1.43||95.20%|
|$1.43 - $1.67||93.60%|
|$1.67 - $2.00||92.90%|
|$2.00 - $2.50||92%|
|$2.50 - $3.33||88.80%|
|$3.33 - $5.00||82.10%|
|$5.00 - $10.00||75.40%|
|Greater than $10.00||31.50%|
Here we can see the dramatic difference between the payoff of short priced favourites vs underdogs. Whilst betting on teams that are paying less than $1.11 will only lose you a small fraction of money, you can expect less than a third of your money back if you place bets only on longshots of over $10.
Now that we know of the favourite-longshot bias, can we exploit it?
Not really, but we can hope to avoid it.
The easy assumption to make is that if punters are more inclined to back longshots, bookmakers will set odds low to take advantage of this (and therefore we should stick to favourites). However, even if you were to back only short priced favourites you should still expect to lose money- so a smart sports bettor should take any bet where you have the have the expectation to win, regardless of odds.
That said, next time you place a bet, you should be conscious of the favourite-longshot bias. History proves that your 20-1 outsider is more likely to actually be a 30-1 chance, and the low-return favourite may be the better value bet.