top of page

CBFORE  -  Our Platform

The CBFORE Revolution: Demystifying the Financial Markets


So far, the world of financial data analytics platforms applied to the financial markets has been mostly commanded by two main approaches: Quantitative and Fundamental.

Usually both of them more or less failed in providing their adopters with a clear and objective picture of what’s really happening on the financial markets.

Even worse, these two approaches spectacularly crumbled down every time they had to face a financial crisis or a sudden shock on the markets.



The Quantitative approach failed because it frames the market behaviors into rigid mathematical rules and models, while they actually are a social phenomenon driven by human greediness and emotions.

The Fundamental approach failed because it relies on data that are already obsolete once they are issued, while the markets are always projected into the future.


Even the most recent frontiers in quantitative research, AI and unstructured data (sentiment analysis from social networks, analysis of economic activity based upon satellite photos, and so on), have proven to be unsuccessful, notwithstanding an initial great hype.

Unstructured data are very complex to be processed, much resource-demanding (not to mention its huge environmental impact), and their usefulness could never be actually proved.

AI (meaning with it, its current improper synonym: deep learning) has been massively applied to analyzing and trading the financial markets, but it eventually proved to be unable to efficiently manage the many outlier events that affect the data produced by the markets.

On top of this, models based on deep learning networks and algorithms are dangerously too close to a black-box system, hence it’s almost impossible to understand and correct their behavior when  they are proven wrong.


The CBFORE project is born from the willingness by some expert portfolio managers to find a third, effective way in analytics applied to financial market data, starting from the unique point of view that only its prospective users could offer.



The foundations of the CBFORE project are extraordinarily simple and straightforward, actually.

They take inspiration from the concept of “The Wisdom of Crowds”, well explained by Jim Surowiecki in his homonym book from the first years of the 20th century.

Surowiecki describes many studies elaborated by Sir Francis Galton, in which it’s clearly proved how the average guesses by a crowd of people are always almost perfect in estimating unknown values (when called to guess the weight of cattle during a fair, for instance), while the single individuals are usually spectacularly wrong when facing the same task.

Hence, when a crowd is sufficiently large, in some mysterious way it is able (in average) to guess and predict things that a single person usually mistakes by a lot.


In a few words, that’s the hidden, immense power of statistics, and it can be applied to the main financial markets as well, where the number of players is so huge to create perfect samples of the “Wisdom of Crowds”.

This is why prices, volumes, volatility, and their shifts and changes are all the data we need to apply this wisdom to the financial markets, and no exotic data or opaque black box algorithm are actually needed to do this.


The exponential growth recently reached by data analysis and machine learning tools is the last thing we were missing to get a definitive grip on the precious wisdom coming from the masses of historical and live data we have. Now we are ready to finally demystify the financial markets.


Starting from these foundations, the CBFORE project declined the Wisdom of Crowds to the financial markets through the following principles.

First: traditional, structured data have still much to tell us, and they can be granted with a new life thanks to the machine learning and increased computing power that are available now.

Second: the adoption of a fully “empirical” approach. By framing them within the right model and rules, the historical and live market data can tell us how human beings act and react to the financial markets, and what are the usual habits that repeatedly drive them.

Third: the model of the market has to be built in a hybrid, more efficient way. The skeleton of the main market model is moulded by the multi-annual experience of expert human beings, and the rules, sub-models, and parameters that make up the empirical part of the model are designed by machine learning and AI algorithms, a complex computing task that human beings can’t execute.

In this way, the best of both worlds (humans and machines) is exploited at the maximum possible level, and everyone does what knows at best.


Too many Startup companies in the Fintech sector totally overlooked their impact on the environment, by using energy wasting techniques like massively huge neural networks (deep learning, Generative-AI), or in some way contributing to the increase in cryptocurrencies mining.

We modelled our business from the start on a strictly limited environmental footprint, by applying deep learning algorithms only to selected parts of our analytic engine, and by adopting a massive use of lighter, synthetic metadata at production stage.

What we finally got to, it's the first working instance of an hybrid, neurosymbolic approach to financial risk analysis, the new frontier in the long road to effective and reliable AI. 


Once the analytic engine at the core of the CBFORE project had been finally deliberated to the test phase, the first results of this radically innovative approach were surprisingly quick to come.

Among the many examples:

  • The timely prediction of the bleak outlook for the British Pound well before the Brexit Referendum (inherently predicting the outcome of the Referendum itself, against all polls).


  • The exact detection of the maximum and minimum prices of US Treasury Notes through the 2018-2019 turmoil, and of the S&P500 Index through the Covid19 pandemic crisis.


  • The steady profitability in the long-only-full-invested stock selection, by piling up over +6% average net annual extra performance over the S&P 500 and EuroStoxx 50 benchmarks on stress-tested backtests covering more than 15 years.


  • The creation of a proprietary Option price model able to track with extreme precision any option chain on the market, and to fully maintain its reliability even through extraordinary events like the Covid19 crisis, where it managed implied volatility levels well over 500%.


The engine is already working in all its features, and a fully functioning Analytic Platform is working at a production stage.

The current Platform release produces its reports in pdf format.

The reports can be requested “in bulk” on a periodic schedule customised upon the client’s needs, or on demand through a direct request service by email that can be used by the client to receive single customised reports in real time.

The launch in production of a fully remote working web architecture, and of a “light” mobile version of the analytic platform is already at an advanced stage of development.


The analytic engine, already working in its definitive form by processing trillions of financial transactions every day and 20-year-plus of historical data, is the “substratum” upon which it is possible to build vertical applications dedicated to managing at best the risk and the opportunities on the main financial markets.

Three are so far the vertical applications built upon this engine, that are already available in their definitive production release:


  • MONEY TRACKER: Daily monitoring of the most relevant positions held on the main financial markets by potential sellers and buyers, and their changes and movements. For sure, it represents the most effective application yet of the famous saying “Follow the Money”.


  • ALPHA BUILDER: Selection support system for asset picking investments, in order to optimize equity and currency portfolios, and to effectively design Structured Products. Applied to the S&P 500 and EuroStoxx 50 stock components through the last 20 and 15 years respectively (long-only, fully invested, no leverage, annual rebalancing), it generated an extra profit versus the benchmark index of more than +6% net on average per year, excluding dividends and with no profits reinvestment.


  • OPTION MANAGER: Investment risk hedging support system through strategies based on option derivatives. Any possible strategy is evaluated through its whole life-cycle, not just at open and expiration times (as it's usually done). Thanks to its radically innovative empirical model of the implied volatility surface, this system provides reliable option price estimates and evaluations in any possible market scenario, including the most extreme ones, where the mathematical/stochastic models proved to be dangerously unreliable.


The demand for analytic platforms like this one is strong like never before, due to the increased frequency of financial crisis and of deep shocks on the Markets, but the current offer in this specific financial data sector is very limited both in term of quantity and quality.


Particularly for the Option Manager app, the timing of its release couldn’t be better.

The Option trading market has been skyrocketing for many years now: daily volumes are up +74% from 2016 to 2020 or +790% from 2000 to 2020, with many newbie investors approaching this market for the very first time and managing these powerful (and potentially much dangerous) instruments with little or no proper education at all.


The CBFORE platform displays in a very clear way the exact potential risks and opportunities arising from any conceivable strategy in options in any market scenario.

The need for an Analytic/Educational tool like this is really huge and even urgent right now.



Through these three market-ready vertical applications, the analytic engine is already fully able to promptly produce real revenues and profits for its users, by detecting risks and opportunities in a proper and detailed way.


The final target of the CBFORE project is the creation of the first objective, fully-automated, machine-learning-based and data-based Research Department, dedicated to anyone who has to deal with financial markets for any reason and with any kind of exposure.

The information provided by this integrated platform can be used by CFOs dealing with currency exposures, Asset and Fund Managers to optimize their trading operations, Banks and Brokers that are keen to provide their customers with actionable info and insights, proactive Risk Managers who want to go the extra mile to assess financial risk, and Corporations and financial institutions that have to comply with complex financial reporting to Financial Regulators.


It’s not just a new analytic platform among the many ones available on the market.

It’s the definitive disruption of the current financial data market, the sector within the Fintech market that is expected to grow the most, and exponentially, through the next few years.

A revolution that will finally open the way to a brand new generation of objective analytic platforms applied to the financial markets, with no competitor in sight for a long time to come.

bottom of page