S&P 500 Bear with Full Tactical Tilt


Category : Smart Beta


Download Brochure



S&P 500 Bear with
Full Tactical Tilt

S&P 500

Cumulative Performance



Annualized Performance



Annualized Volatility



Max Drawdown (Day)



Average Daily Exposure




1) Strategy Specific information

S&P 500 Bear with Full Tactical tilt aims at profiting from a decline in the S&P 500, its benchmark, over the recommended holding horizon (1 year). This strategy alleviates the timing aspect of shorting the market.

S&P 500 Bear with Full Tactical tilt is an improvement on low-cost beta investing, thanks to a tactical allocation overlay. The strategy implements temporary deviations from the benchmark by over/under-weighting the exposure to the benchmark itself. This ensures that the strategy represents a purely tactical and timing-based process, without any factor (value, growth), market-cap (small cap) or risk-weighted (equal-weight, minimum variance, etc.) approach.

The strategy is always short the benchmark, and enters into tactical tilts (+/- 100%) of limited duration based on the detection of probable bullish or bearish movements. A behavioural analysis determines the excesses of the benchmark, and contrarian tilts are therefore implemented to trade against those excesses (overweight after a bullish excess, underweight after a bearish excess).

The tactical tilts are intended to reduce volatility relative to the benchmark.

2) The “Complexity” family of strategies

All the strategies in the “Complexity” range are grounded in the research in complex systems, as those systems offer a better analogy with financial markets than Modern Financial Theory (MFT). The frequent bouts of exuberance, intense risk-aversion or herd behaviour observed on financial markets invalidate the main assumptions and conclusions of MFT and question the validity of long-term forecasting. In fact, prediction in financial markets more closely resembles weather forecasting than odds calculation in a lottery: in meteorology as in financial markets, the accuracy of predictions decreases rapidly as the length of the prediction increases. “Better predict 7 times the weather for tomorrow than once for the whole week!” Long-term statistics in weather forecasting actually back this saying, with around 85% success rate for overnight prediction versus slightly more than 50% over a 10-day horizon.

The endogenous instability and non-linearity observed on financial markets render inefficient the traditional equilibrium approaches and normal distribution assumptions: prediction has to accommodate the existence of unpredictable chaotic phases, and the appropriate tools should therefore be used. This toolbox is found in statistical physics, at its root the study of large numbers of particles and their interactions. This strategy makes large use of such tools. Moreover, the strategy uses concepts first introduced by B. Mandelbrot, the 20th-century mathematician, who coined concepts such as self-similarity, Noah & Joseph effects, etc.

This strategy makes full use of those concepts to achieve a robust, dynamic and auto-adaptive pattern-recognition process. This performance engine is built on sound statistical physics methods, such as numerical analysis, denoising, filtering, etc.

High-frequency and high-quality data is needed for this strategy as only price data is used as raw material in the process. The quality of data is monitored continuously and 2 separate sources are used to ensure continuity and reliability. The numerical analysis relies on the use of 1-, 5- and 15-minute data points.