- Luck and the Paradox of Skill. How to become a profitable bettor?
- People and robots vs. chance: Who wins?
- When does betting become a game of skill?
Betting is centered around the pursuit of positive expected value to achieve steady profits. This endeavor requires gamblers to weigh the roles of randomness and fortune in the outcomes of events. This piece delves into the debate of whether becoming a profitable gambler hinges on luck or skill. Continue reading for further insights.
The essence of Squares & Sharps, Suckers & Sharks was to examine the reasons behind the profitability of some gamblers over others. Mirio Mella highlighted that certain professions, such as actuaries, financial traders, and professional gamers, might naturally lend themselves to gambling success. This piece, however, broadly investigates if success in the gambling world is a result of luck or a skill that can be cultivated.
Luck and the Paradox of Skill
The majority of outcomes in the gambling market are determined by chance, an unwelcome truth for many bettors. Indeed, casino offerings like roulette and craps are purely chance-based, following straightforward probability laws. But in sports betting, with its unpredictable outcome probabilities, there seems to be room for profitable strategies, right? Technically yes, as some individuals might excel in gathering and analyzing information more efficiently than others, but this advantage is practically limited to a minority due to what's known as the paradox of skill.
Generally, in lower-scoring sports with more participants, luck plays a more significant role.
When prediction becomes an arms race with increasingly sophisticated methodologies for forecasting sports outcomes, absolute skills might improve across the board. However, relatively, we might not advance much.
Even as absolute forecasting skills enhance, the gap between the top and bottom narrows; given that gambling involves both skill and luck, with diminished variance in relative skills, luck's influence becomes more pronounced. Considering the cost of participation (the bookmaker's margin), most will likely face losses over time.
Betting robots vs. Chance
The football prediction service Botprediction.com demonstrates the randomness in betting by running 40,000 soccer prediction robots for the total goals market, likening them to a large cohort of bettors.
“Statistically, there are always big losers and big winners in that group. Our prediction software enables you to follow those few winners with win rates above 70%”
A few years back, I analyzed these robots' performance distribution over a five-week span. This is what it looked like.
The outcome was akin to a random coin toss. The idea of following a high-performing robot and expecting continued success is flawed. There's no consistency in performance from one period to the next. Robot success rates merely revert to the mean, illustrating that outcomes are entirely down to chance.
Real people vs. Chance
Replacing robots with real individuals, one might expect a difference, as humans don't make random choices.
Furthermore, I analyzed a large sample of bettors (6,044) posting picks (1,073,029 in total) on the sports betting community Pyckio.com. Using the t-score to compare betting records against market expectations based on our odds, the performance distribution was as follows.
The normal distribution of betting performances points to randomness. While there are better and worse bettors, their distribution mirrors that of coin-toss decisions, despite each claiming sophisticated forecasting methods.
A skilled but unfortunate poker player might still face losses after 100,000 hands, nearly two years of full-time play.
Rather than being inherently good or bad, most bettors' outcomes are influenced by luck. To further illustrate, I divided the records of 249 bettors with over 1,000 picks into two halves and correlated their performances.
If skill beyond luck were evident, we'd expect consistent performances. However, a scatter plot of these correlations revealed only 0.19% of variance in second-half performance could be attributed to first-half performance, indicating that nearly all variations were due to chance. This doesn't deny the presence of skilled bettors but suggests they're obscured by randomness.
An R2 of 0.0019 means that just 0.19% of the variation in the second half performance of these 249 bettors could be explained by the variation in the first half performance. The implication is that all the rest is explained by chance. Collectively, these bettors were not demonstrating consistent performances; like Bot Predictions robot they were largely just regressing to the mean. This is not to argue that there were no skilled bettors at all; rather, what few there might have been were hidden amongst the noise of chance.
The Luck-skill continuum
Michael Mauboussin's "The Success Equation: Untangling Skill and Luck in Business, Sports, and Investing" introduces a luck-skill continuum for various activities. Pure chance games like roulette and lotteries sit at one extreme, while skill-dominated games like chess occupy the other. Most team sports lie in between.
This suggests the unpredictable nature of sports betting.
Typically, lower-scoring sports with more participants see a greater influence of luck. What about gambling? Mauboussin ranks activities like investing, betting, and poker closer to the luck end, supported by the distribution of bettors' performances.
The environment of skill
Expertise acquisition usually follows 'Naturalistic Decision Making,' where prolonged practice and feedback help memorize and recall data patterns. This requires a stable and regular environment for predictable cause and effect. Such an environment exists for skills like playing chess or tennis, but can the same be said for sports betting?
In a luck-dominated environment, focusing on decision-making processes over outcomes is crucial.
The challenge in forecasting for betting is that skill doesn't merely involve choosing winners; it's about outperforming others' choices.
As the paradox of skill shows, betting is a relative-skills competition. Bettors compete in a market, with success depending on evaluating whether odds reflect all available information accurately.
A zero-validity environment?
Validity measures whether perceived causes are true causes, and whether our assessments consistently confirm that. The analyzed sports bettors generally failed to demonstrate consistency or validity, indicating that betting markets, dominated by luck, are neither regular nor predictable.
Where luck disrupts the link between skills (causes) and profits (effects), a skilled player can lose money, and an unskilled one can profit.
The betting market's complexity and randomness, driven by unforeseeable news, severs the cause-effect link between a bettor's actions and their bankroll's growth.
The limited scope for feedback complicates skill acquisition in betting, akin to practicing roulette.
Daniel Kahneman describes betting as a zero-validity environment, highlighting the illusion of skill in perceived betting successes.
When does betting become a game of skill?
Perhaps the question isn't if betting is a game of skill but when it becomes one. With repetition, the influence of skill grows in activities involving luck and skill.
For instance, tennis and poker demonstrate how skill emerges over numerous repetitions, distinguishing skillful players through their strategies and decisions.
Over time, even with a slight advantage, the skilled bettor's success becomes evident, though it may take numerous wagers to manifest.
Overcoming chance and the Paradox of Skill
In nearly zero-validity betting markets with minimal skill opportunities, several strategies can enhance success chances:
Bet in markets with a higher variance of skill
In layman’s terms that means betting markets that are less familiar to the majority of players and the bookmaker too, with less readily available information and news to process for forecasting. The bigger the difference between the least and most skilled forecasters, the smaller the role of luck.
Think relative, not absolute
Where there is competition and luck, it’s only relative performance that matters. I might be able to predict 9 out of 10 Premiership results at a weekend but if the person I’m betting against can do all 10, I’ll still lose. As Nate Silver says of poker, “you can make 95% of your decisions correctly and still lose your shirt at a table full of players who are making the right move 99% of the time.” In this context, it’s also important to remember that it’s not really the bookmaker you’re trying to beat, but everyone else.
Focus on the process, not the outcome
When competing in an environment where luck dominates you should focus on the process of how you make your decisions, rather than the outcomes themselves. In betting, analyze the way you forecast, not the size of your bank balance. Where a lot of luck breaks the link between skills (causes) and profits (effects), a skilful player can lose money whilst an unskilled player (relatively speaking) can profit.
Study the closing line
It is now well established that our closing line provides one of the best measures of ‘true’ result probability. The amount you consistently beat it by will provide a very good indication of the profit you can expect to make. Consistently beating the closing line by an amount greater than our margin implies that you are sufficiently more skilled than many of your competitors.
Now you know how to become a profitable bettor. Sign Up Now or click HERE to play at 7x7Bets, the most reliable and trustworthy online casino in India. Don't forget to claim your withdrawable real money welcome deposit bonus, weekly cashback bonus and referral bonus!