Optimize to find a real algo trading edge, not a stochastic illusion

In this series, Martyn Tinsley embarks on a journey to challenge the backtesting and optimization status quo that prevails among many algorithmic traders. In this first tutorial, he demonstrates how parameter values that provide no edge at all often appear to perform in an optimization, as if they do.

He also demonstrates how parameter values with little or no edge can often produce 'far better' results in an optimization than the parameters that offer the best edge, and the best chance of success in the long term. This is due to the stochasic (random) effect prevelent in all backtesting processes.

Finally, a solution is proposed involving increasing the sample size (number of trades) that the system generates.

This video is a "must-watch", and every algorithmic trader needs to be aware of this phenomenon, otherwise they could be choosing ineffective parameter values from optimizations, based on stochastic effects, leading to sub-standard performance when traded in a live account.