Mercurius is an Italian fintech startup that aims at assetizing sports betting markets through the usage of artificial intelligence and machine learning technologies. Founded in 2018 it released Tradr in 2019 delivering positive results to its users since then.

What Prediction Markets are and how they work?

Prediction markets are systems where you can place bets on a particular outcome to happen or not, like whether Israel will go to war with Iran, or how much global temperatures will rise because of climate change.

These systems ensure that we have a financial stake in being accurate when we make forecasts, rather than just trying to look good to our peers.

So a comprehensive definition of prediction market could be- A speculative market designed so prices can be interpreted as probabilities and used to make predictions.

Let’s make a practical example to understand this link between price and probabilities

In a fictional prediction marketplace called "Futurne" a trader decides to sell a contract that agrees to pay one dollar if a particular state occurs at a particular time in the future and pays zero in all other states. See (Arrow-Debreu security)

An example: a contract that agrees to pay 1 dollar if Tulsi Gabbard becomes the 46th president of the United States in 2020.

After some trading, these shares are all bought from others at an equilibrium price of 0.60$ each.

This activity between traders can be considered as a zero sum game (if Tulsi Gabbard wins the seller loses 0.40$ per share sold and the buyers profit 0.40$ per share bought, vice versa if Tulsi Gabbard wins the seller gains 0.60$ per share and the buyers lose 0.60$).

If the equilibrium price represents the true probabilities of the event occurring or not the expected profits both for the seller and the buyers are zero.

Hence the probability of Tulsi Gabbard to become the next U.S. president derived from the transaction happened in the "Futurne" marketplace is: p(-0.40$)+(1-p)0.60$=0

Where p is the probability of Tulsi Gabbard to win and (1-p) is the probability of Tulsi Gabbard losing the election.

After some easy calculations, we find that p=0.60 -> 60% and (1-p)=0.40 -> 40%

As you can see this holds only if the equilibrium price represents the true probabilities of the event which is quite a strong assumption. To solve this problem and to let the assumption holds is the existence itself of prediction markets.

Profits, information advantage, and skin in the game

The existence of prediction markets permits to the individuals who have better information and/or better forecasting skills to profit from less-skilled ones thus there is an incentive for these individuals to use their information and skills, and by observing the equilibrium prices the public can derive the probabilities generated from the interaction by these skilled individuals.

Also by having a broad set of participants in the market where each participant owns a piece of valuable insight, the resulting equilibrium price is the money-weighted average of this knowledge. Hence the more information is crucial, the more the owner wants to risk money on it; see (Wisdom of the Crowd)

Since there is a skin in the game principle built-in these markets, in the long term the less fit individuals go broke forcing the fittest to compete against each other, which makes earning profits more difficult, which, in turn, means the market is more efficient and requires superior forecasting abilities.

Sports betting market and financial market

The sports betting market and the financial market are both comparable to prediction markets as long as some important unique features are taken into account.

Financial markets

The first feature that is unique to financial markets and not prediction markets is that an important percentage of their participants are hedgers.

Hedgers are participants with the sole purpose to use financial markets to reduce a particular risk they face, so their direct information contribution to improve the forecasting power of prices is of minimal impact.

Example of a hedger:

A gold mining company knows that it will gain $10,000 for each 1 cent increase in the price of gold over the next 3 months and to lose $10,000 for each 1 cent decrease in the price during the same period. To hedge, the company’s treasurer should take a short position in the gold futures market that is designed to offset this risk. The futures position should lead to a loss of $10,000 for each 1 cent increase in the price of gold over the 3 months and a gain of $10,000 for each 1 cent decrease in the price during this period. As you can see the profits & losses of the company’s business offset the profits & losses from the futures position and vice versa.

The second important feature to consider is that insider trading is illegal in financial markets while it is welcome in prediction markets. Allowing insider trading in prediction markets permits that the most important information about what is going to happen is reflected in the price.

Sports betting market

If sports betting is done via a bookmaker the resulting price/odds may not have forecasting value; this happens if the bookmaker doesn’t properly adjust the odds on the information brought by its customers. Bear in mind this topic will be covered in another article.

If sports betting is done via a betting exchange then the resulting odds contain forecasting value, because the odds generating process is the same as the one that happens in prediction markets.

In betting exchanges, sharp players usually don’t suffer from bans and limitations and this leads to having their forecasting skill discounted in the odds.

The kind of hedging which we have seen in financial markets - whereby an actor protects itself against a commercial risk - is absent, so all the action tends to be speculative, which is one of the principles on which prediction markets are based on.

Example of absent hedging in a betting exchange:

You are the owner of Liverpool FC and if you win the next league match against Manchester City you are going to win the English Premier League earning £1,9million in prize money. To offset this risk you could lay Liverpool FC to win in a betting exchange and earn a percentage of the prize pool no matter what the outcome of the league match is.


Prediction markets help us as a society in having a better glimpse into the future while allowing its actors to make business on top of it. The whole point is to have information and a way to correctly evaluate it. And, after all, as Marco Blume says "betting is having an opinion with skin in the game".