fbpx

What makes the BIF Strategy unique and
why do these results say something?

BIF: A combination of statistical significance and economic arguments

The BIF strategy is a strategy that buys good companies at a cheap price, focusing on the company behind the stock. The demonstration that the strategy works and will work in the future is based on a combination of statistical and economic arguments. In addition to the statistical significance shown in our backtest (a test on past data), there are also arguments (philosophy and principles) that the opportunities the strategy takes advantage of will continue to occur in the future, which we call economic arguments. Our strategy is based on the principles and philosophy of investors such as Peter Lynch and Warren Buffett. An essential aspect of this philosophy is that the market is not always rational. This causes inefficiencies in the market, in which case certain companies are mispriced. Given that most people’s investing behaviour does not change, there will continue to be mispriced and undervalued companies that the BIF strategy is trying to find. Because we expect that the irrationality of people’s behaviour will not go away, we expect our strategy to continue to work. This is also something that has been visible in the market for over 100 years, a lot longer and further than our backtest.

BIF performance vs SP500 2024 results 2024

Why do we not provide an average performance metric?

We have conducted many backtests over different periods. These periods differ in length, and therefore, it is also very hard to give an “average” performance metric. Besides that, most algorithms show one particular chart about a certain period, indicating overperformance versus a benchmark. It says much more to give an overview of outperformance over multiple periods than just an average performance for just one period. Unfortunately, we can’t see the future either; we don’t want to mislead investors. You must take it with a grain of salt everywhere you find an average performance measure.

Why should you be very careful when performing a backtest?

When conducting a backtest, a whole set of biases comes into play. One of the most essential biases in the case of equity research is the “survivorship bias.” Survivorship bias refers to the biased view that can arise when only successful entities are included in the analysis while unsuccessful or non-existent entities are excluded. In equities, this can arise because data from companies that are no longer publicly traded is harder to come by, unavailable or otherwise. For instance, in our backtests, that is based on creating a selection of the SP500. It is very important to have good data on which companies were in the SP500 at the time and to ensure that the data quality is the same as for the companies still listed. We have taken good care of this in our backtests.

Methodology of our backtest(s)

To know if a backtest is of high quality, it is very important to understand what the methodology of the backtest is. What is unique about the BIF strategy is that it is developed through a bottom-up approach. This means that the principles around the above philosophy were implemented before looking at stock performance. However, if this happens the other way around, chances are high that there is model overfitting. If overfitting occurs, it is much more likely that the model (our algorithm) will only work in the past and not in the future. So with the BIF-strategy, we have taken this into account.

 What is the length of our backtest(s)

The periods range from a minimum of 3 to a maximum of 6 years. The backtest uses the BIF strategy and reweights the BIF portfolio quarterly. We generated the outcome from every possible combination (between possible start and end points) between a minimum of 3 and a maximum of 6 years, testing whether the BIF strategy outperformed the S&P 500 over the same period. Going forward, we are working hard to extend this backtest’s length to ensure the results are even more robust.