Stralliance


Stralliance's methodologies

  • Stralliance's impressive 9-year (theoretical 2015-2021, live 2022~) performance statistics:

    Avg +62% pa net returns, with only 9 losing months over a 110 month period;

    Our worst historical daily cumulative DrawDown over 9 years:  5.4% DD  (S&P 500: 33.9% DD);

    Profit to risk ratio:  10.4  (S&P 500: 0.2);

    Yearly Sharpe ratio:  2.0  (S&P 500: 0.5);

    70.8% trade win rate;

    Avg 30 daily trades per month.

  • We are Long/Short market-neutral - our returns are not dependent on market trends.

  • We trade three of the world's most liquid futures contracts:
    10 Year Note (ZN),  eMini S&P 500 (ES),  eMini NASDAQ (NQ).

    Liquidity is never an issue, so capital gates (redemptions halt) are not required.

  • Our trades are open from a few minutes up to a max 22 hours for each market session.  We are not exposed to any long-term or weekend geopolitical risk.

  • Safety target:  maximum 10% risk (i.e., cumulative drawdowns), currently less than half our target.  Capital safety is our prime priority.

  • Performance target:  30% pa + net returns, currently exceeding this by x2.  Consistent returns is our major priority.

  • Stralliance trades a basket of 5 (2 volatility-capture, 3 directional) NeuralNet-based, 100% systematic methodologies.  All strategies, as well as our 10-stage risk management, are fully systematic.  All five systems adapt to market conditions in (almost) real-time.

  • Our five core systems compliment each other well.  There is a low correlation between systems performance, hence the low volatility and smooth returns curve.

  • Stralliance's proprietary multi-layered Recurrent Neural Networks (RNNs, 42,000 man-hours' development) have been specifically designed to avoid over-fitting.

    Over-fitting (or curve-fitting) is the very common process of developing systems and optimizing backtests for best-fit past performance, resulting in unsustainable performance metrics which are unlikely to continue into the future.

  • Stralliance's risk-adjusted performance far exceeds that of any top-performing funds.


We live in exponential times.   Markets are evolving at an accelerating pace.   Standing still is not an option.

  • Constantly emerging disruptive technologies such as sub-millisecond HFT algorithms, instant news-scraping bots, increasingly smarter neural networks, and the general trend towards shorter-term trading - these and other factors result in rapidly changing market characteristics over shorter spans of time.

    Static trading methodologies which may have been profitable just a few years (or even months) ago, rapidly become unprofitable and obsolete, as their edge disappears under today's shifting market conditions.   In order to maintain an edge over these rapidly evolving markets, it is essential for modern trading methodologies to be both fully systematic and self-adapting to rapid changes in market conditions.

  • Stralliance's AI-based methodologies are 100% systematic, and are continually evolving.

    Our trading methodologies are based on proprietary Recurrent Neural Networks (RNNs) finding hidden profitable patterns in these ever-shifting markets.   Our edge comes from the correct application of RNNs, a balanced portfolio of complementary strategies, and the understanding and mitigation of the Quant's system-development curse: over-fitting.

    We aim to maintain our profitable edge over ever-changing market patterns, and take advantage of new opportunities as they emerge in the markets.   As emerging market patterns overtake and outperform older ones, new methodologies are being continually researched, developed, monitored, reviewed and implemented through our R&D pipeline process.   Our Research & Development process never stops.

  • Capital risk exposure is actively managed through systematic position sizing - our systematic 10-layer Risk Management procedures are an integral part of our smooth returns.   Contract lot sizing is adjusted in real-time through accurate systematic position management for each individual trade, which results in relatively predictable smoother returns.

  • Stralliance is Long/Short market-neutral.  We are agnostic as to future predictions, and do not make financial market forecasts.   We determine profit probabilities through a confluence of complex interlocking factors - there is nothing “predictive” in our methodology.

    We apply systematic situational analysis, combined with skillful integration of probabilities, to make intelligent choices which tend to pay off on a probabilistic basis over the long term.   In a world overflowing with competing hedge funds, Stralliance is uniquely an “edge” fund.

  • Stralliance's cutting-edge methodologies and neural networks are the result of Jose Silva's high motivation and passion for financial engineering:
    - Approx 42,000 man-hours ($12 million just in programming fees) full-time R&D and programming since early 2010;
    - More than 22 years' experience in the markets, and over 80,000 man-hours designing & building unique R&D tools since 1999.




© 2015-2024   Stralliance  Capital  Management

Past performance is not necessarily indicative of future results.  The risk of loss in all types of trading can be substantial.
The presentation of information and data provided do not constitute a solicitation for investment funds, nor an offer to transact in any commodity interest or any other related financial product(s).