On this web page you will find our weekly probability of random trading index. This benchmark indicates the probability on an aggregated basis that our trading systems are generating random trading results (like trading by random coin flipping).

The purpose of this index is to identify how much of the performance of our trading systems is due to their system logic and how much of it is due to randomness. Therefore we create lots of random trading sequences using random number generators. Then we compare the out of sample-performance (since 03.04.2001) of our market timing models to the random trading performance sequences. It is very important to use out of sample-data in order to not be fooled by curve fitting and over optimization.

This way, we can measure the true probability that our market timing models are trading at random. We use this technique to search for market timing models and trading systems that are producing trading results that are located on the tail of the return distribution. If our stock and bond market timing systems could not be distinguished significantly from the random trades, we do not trade these systems.

The "Average Probability of Randomness" indicates on a percentage basis how often the random trades outperform our SNIPER Systems Composite Portfolio and a Buy and Hold Composite Portfolio.

Example of our Weekly Probability of Random Trading Index (Out of Sample)

Probability of random trading of our market timing system

IMPORTANT: This randomness index is not intended to provide personal investment advice. This randomness index has been prepared solely for informational purposes, and is not an offer to buy or sell or a solicitation of an offer to buy or sell any security or instrument or to participate in any particular trading or market timing strategy.
Please also read our detailed risk disclaimer and disclosure statement.