Deconstructing “the gambler’s fallacy”: discover how to create winning strategies in sports betting

Many bettors and self-proclaimed professionals take the gambler’s fallacy as an absolute and indisputable truth. I believe that it is wise to consider everything and to take each one’s own conclusions based on proven facts. Taking advantage of the existence of tools such as Betamin Builder, we now have the possibility to analyze specific historical data within the reach of everyone. With the help of this tool, I will dare to raise the doubt about the well-known gambler’s fallacy, and I will reveal how it is possible to beat the bookmakers. So, we will get the answer for this question: is it really possible to create a winning sports betting strategy?

 

We will help, of course, Wikipedia for a definition for the gambler’s fallacy:

The gambler’s fallacy, also known as the Monte Carlo fallacy or the fallacy of the maturity of chances, is the belief that, if something happens more frequently than normal during some period, it will happen less frequently in the future, or that, if something happens less frequently than normal during some period, it will happen more frequently in the future (presumably as a means of balancing nature). In situations where what is being observed is truly random (i.e., independent trials of a random process), this belief, though appealing to the human mind, is false. This fallacy can arise in many practical situations but is most strongly associated with gambling, where such mistakes are common among players.

 

Well, up to here everything very logical, very clear and indisputable. A coin, or a dice, have no memory, and each release is an independent event, it is completely true and absurd to rethink something like that. If when I flip a coin I get several faces in a row, in the next shot the probability of getting a face will be 50%, no higher or lower, and there is no more to say.

But when we talk about sports betting things change. On a coin, it is very clear what is the probability of getting heads or tails, 50% without any doubt, but at a football game or other sporting event, how do we know what the exact probability is? Is each game an independent event even if it is within the same league or competition? Does the player of a tennis match, or the players of a football game, have no memory like coins?

Faced with these questions, let’s go one step at a time. The players are not like coins and do have memory is something that is quite obvious, and therefore their behavior may vary according to the recent results of previous matches, what we do not know is how.

That each match within the same competition is an independent event is also quite obvious that is not true. In fact, in the final stretch of a competition will surely not behave the same as a team trying to avoid relegation that another with better position, and this difference in behavior is caused by previous events.

On what is the exact probability of occurrence of a sporting event, many factors come into play here. On the one hand, the oddsmakers propose opening odds more or less fairly adjusted to the probability of the event, but these odds change in time until their closure. Other factors come into play, such as balances for each outcome in the bookies, and this balance is modified by what is known as “the wisdom of the crowd“, that is, there is a theory that affirms that the general public is capable of correcting for itself the errors of the oddsmakers themselves. In fact, there are several fairly conclusive experiments on the performance and results of “the wisdom of the crowd” in any field, not just in sports betting.

Without going into greater complexities, it can be concluded that the starting odds are proposed by professional oddsmakers and then the possible errors are corrected by the general public. And both in the case of the oddsmakers and in the case of the general public, the human factor comes in. Both are, and humans do have memory, unlike coins.

As a conclusion of all this, I think we have more than enough to raise doubts about whether the gambler’s fallacy applies to sports betting or not.

Watch out! I’m not raising doubts about the gambler’s fallacy itself, which is completely true. If we knew exactly the exact probability of any event, there would be no discussion. What I am proposing is that there may be miscalculations at odds due to pre-event streaks.

In other words, because the factors that define betting odds at sporting events are human and have a memory, is it possible that they make mistakes when assigning odds to events affected by a streak, such as matches played without winning or losing, matches followed without scoring a goal, etc.? Is the gambler’s fallacy, as we know it, really applicable to sports betting?

Let’s put on the white coat of scientists, let’s go into the experimentation laboratory and see it with concrete data from Betamin Builder. A great advantage of using this tool is that anyone can check the results by yourself, there is no reason to believe just what I expose here, which could be manipulated. Try it for yourself.

Here we go, raising the following test cases:

1.- What if we bet ALWAYS on team playing at home, without any conditions?

Obviously, we lose money. You can see that an odds filter is used between 1 – 1000, which actually does not filter anything. The reason is to be able to obtain the results since Betamin Builder requires the use of a filter to show results.

2.- Now the same case, but when the home team comes from a streak of 5 or more games without winning, what happens?

Impressive and more than interesting result that encourages further investigation. There is no effective odds filter, there is no leagues filter, in other words, there is no possible data-mining, only concrete facts, and a more than representative sample, there are more than 10000 bets included in the result!

3.- Let’s see what happens if we raise the streak to 8 or more matches without winning:

Again, very interesting results, with a still representative sample of more than 3000 bets analyzed, we have almost tripled the yield obtained!

Well, although it seems very unlikely, there may still be the doubt that these results are caused by luck. If we observe the same trend in a completely different market, the evidence will be much greater.

Let’s analyze the market of Over/Under 2.5 Goals:

1.- What happens if we ALWAYS bet on under 2.5 Goals without further conditions?

Again, we obviously lose money. You can see again that an odds filter is used between 1 – 1000, which actually does not filter anything. The reason is to be able to obtain the results since Betamin Builder requires the use of a filter to show results.

2.- Now the same case, but when the away team comes from a scoring streak of 3 or more games, what happens?

Again interesting result. There is no effective odds filter, there is no leagues filter, again no data-mining is possible, only concrete facts, and a sample of more than 5000 matches and the results have improved considerably.

3.- Let’s see what happens if we go a little further, a scoring streak of 4 or more games:

Impressive, we have improved the results much more, turning from negative results to positives, with a sample that can still be considered representative of more than 2000 matches, and without any type of data-mining possible. This could be a good start for a new bet strategy!!!

Conclusions

As I said at the beginning, I invite everyone to draw their own conclusions. The tool to do it, Betamin Builder, is available to the public for free, there is no excuse.

I will expose my own conclusions, which of course could be debatable or wrong:

  • The gambler’s fallacy applies without any doubt to independent events that accurately known probability of occurrence, such as the toss of a coin, a dice, etc. Therefore, in casino games, it is completely true that if we have several reds in the roulette the probability that the next one is red again is the same as in the first throw and is not modified by the previous results.
  • In sports betting the situation is different and it is not possible to apply the theory of the gambler’s fallacy in the same way; the results of the experiments carried out make it evident. The reasons are:
    • Sports events are usually framed within the same league or competition, with continuity, which means that they are not completely independent events.
    • Athletes have memory, unlike coins or dice.
    • The exact probability of a sporting event is unknown, we only have one approach. It is determined based on the human factor (oddsmaker and wisdom of the crowd), which does have memory.
  • The experiments carried out with Betamin Builder make it abundantly clear that the odds assigned to sporting events are being deformed with the recent and more or less elongated in time streaks. Therefore, it is possible to take advantage of the streaks and beat the bookies taking advantage of them.

 

How to take advantage of this and beat the bookies?

Very simple, you can follow with very few clicks one or several of our public betting strategies based on streaks, such as:

  • Backing the Draw / Apostando al Empate
  • Over 2.5 against the trend / Más 2.5 contra tendencia

 

You can also be more creative and go further, you can create your own football betting systems in 3 simple steps:

  1. Think about what kind of streak you want to base your strategy on. A streak of wins, losses, goals, etc.
  2. Analyze with the help of Betamin Builder and your own creativity what are the ideal filters to apply to your own strategy based on streaks, from among the wide range of available filters.
  3. Shape your strategy based on odds filters. You may not want to bet on very high odds, or your strategy may not be profitable for a certain range of odds, etc.

 

And what else is possible to do to beat the bookies with the help of Betamin Builder?

Another similar approach, which could be the basis for another article, and which is used in other public strategies that we invite you to follow, is to be based on the recent form instead of relying on some kind of streak. We invite you to exploit your creativity with these clues.

Comments 0

Betaminic Admin

March 26, 2018

Hi johnberg,

Thanks for your questions and for the interest shown in our tool. Regarding your query about the sample of the proposed strategy (857 bets), in principle, we think it is a sufficient sample to be able to evaluate the strategy. The more historical a strategy analyzed, the more reliable it will be in the future, but a sample close to 1000 picks, as is your case, we think it’s enough.

You also ask us about league filters and filters in general. We advise that behind a strategy there is a logic behind, it’s not about looking only that the historical data have good results. If a strategy is created and it is forced to select only leagues that are positive, without reasoning, it tends to optimize a strategy excessively, and it is not good. If you are analyzing, for example, a strategy for leagues that have many goals, or if you are studying only for certain countries because they meet similar conditions, you should not exclude any league, just because it shows negative results. It also depends on the sample or the strategy in question, for example, if a league shows bad results year after year with a large sample for a certain strategy, it can be excluded, because it could be that the league does not adapt to the specific filters that are being used.
As a summary, we tell you that the filters that are used in a strategy, if they have causality, a logic behind, will be more likely to remain profitable in the future.

Kind regards,
Betaminic