034 - Intra-Day, High-Octane, Robust Futures Trading - Bob Pardo - Part 1 of 2
A Masterclass on Robustness Testing with Walk-Forward Analysis
Of the biggest problems quantitative traders probably face, the first is no-doubt over-optimization and the second is likely finding inspiration for new ideas. In-depth interviews with market wizards surely have to be one of the best ways to learn quickly, avoid common pitfalls and find untold amounts of inspiration hidden between the lines. Listening to experts that have been at it for decades, for me anyway, is an incredible education.
In this show I invite you to again spend over an hour with the legend Bob Pardo on the ins and outs of his trading, his philosophy and his edge. And the best bit is, this is just part 1 of 2. In the second part I'm going to deep-dive walk forward analysis with him and I'm sure I'll be walking away with some highly practical tips and tricks.
Bob’s career spans several decades of evolving market dynamics, groundbreaking system development, and a philosophy rooted in adaptability and robustness. His journey—from early days on the trading floor to pioneering walk forward analysis and working with the likes of Solomon Brothers, Dunn Capital, Daiwa Securities & Goldman Sachs—offers a compelling narrative for quantitative traders seeking both inspiration and technical insights.
Early Beginnings and Career Milestones
Bob’s introduction to trading started on the CME floor itself, working at institutions where he witnessed firsthand the raw energy of the market structure. His early exposure, including a stint working for John Meriwether (of LTCM fame), taught him the value of “coming up with numbers” for support and resistance levels. “I got my first taste of systematic trading when I realized I could offer traders clear, actionable figures,” Bob recalls. This approach not only helped him establish credibility early on but also paved the way for later innovations and a journey into algorithmic trading.
A pivotal moment came when Bob encountered Perry Kaufman’s work, rummaging through a sacked colleague's desk, which spurred him to test ideas and build custom trading software on an Apple II, no doubt putting all of its 48k of memory to good use! I'm not sure anyone was building back-testing software before Bob. He recalls coding strategies daily, iterating through ideas systematically from the beginning. This pioneering work led to what would eventually evolve into his product, “Swing Trader,” credited by some with inspiring the term itself.
Bob’s career included influential roles at prestigious firms. Notably, the XT99 fund he implemented at Dunn capital maintained a track record of close to 20%pa over 12 years. He went from strength to strength, continuously developing his ability to build robust, multi-strategy systems that delivered impressive long-term returns.
Trading Philosophy & Diversification
For Bob, the cornerstone of successful trading lies in diversification—across markets, timeframes, and strategy variations. He champions an approach of "exploration rather than optimization". So, rather than an approach apt to over-fitting, he likes to dissect the data & systematically explore it for alphas. Scientists can conduct many experiments before a breakthrough, and that's the mindset he brings to strategy development.
Robust systems, for Bob, are also built on flexibility and adaptation. Over the years he has obviously refined the approach (notably some of the biggest modern developments have been in risk management). Still, the enduring success of his walk-forward approach is testified to by the fact that is still foundational to his strategy development, and that the strategies he trades now are more or less only subtle variations on those that have been running extremely successfully for many, many years. Suffice to say he doesn't believe good strategies need to have a short shelf-life.
Boby now only trades intra-day futures markets. He advocates employing trend-following, counter-trend and pattern-based strategies based on finding strategies and markets that are as non-correlated as possible.
The Science of Building a Robust Strategy
Exploration vs. Optimization: Pardo invented the walk-forward-analysis approach to strategy development / optimization / maintenance. However, he emphasizes that he does not view walk forward testing as mere optimization but rather as an exploratory process—similar to how Edison iterated through thousands of experiments before discovering the light bulb.
Multiple Variations: Instead of optimizing a single strategy, Bob believes in testing multiple strategy variations to find the most robust configurations. He mentions that when building strategies with tools like Ranger, he often generates hundreds of variations, with many being viable for trading.
Timeframes & Adaptability: Over time, Bob has shortened the walk forward cycles to make them more adaptable to market conditions. He argues that faster, more frequent walk forward updates lead to better real-time performance because the market is constantly evolving.
Better Performance Metrics: Bob has found that one individual walk forward test is not a good predictor of future success. Instead, he relies on a series of tests over time, which provides stronger confirmation that a strategy remains viable. Hopefully we'll get into the weeds of that in the next show.
Risk Management & Selection: He also highlights that walk forward optimization is essential for risk management. Instead of keeping strategies fixed, he advocates for regularly selecting the best-performing strategies from a set of candidates, ensuring that underperforming variants are phased out in a systematic manner.
Strategy Variations and Robustness
Bob’s work has evolved with the changing market landscape. His early systems on daily data were eventually reimagined for intra-day trading, reflecting the shift toward higher-frequency opportunities. One of Bob’s signature approaches is his “trio” of uncorrelated strategies—a portfolio method that blends with-trend, counter-trend and range-bound market systems. This method not only minimizes drawdowns (with the trio’s drawdown roughly half that of any single strategy) but also adapts to the market’s natural bias shifts without overfitting. He stresses that a robust system need not work identically in every market; rather, it must be fine-tuned to capture the unique characteristics of each instrument.
Conclusions
Bob’s career is a testament to the power of systematic, adaptive trading. His journey—from early coding on an Apple II to developing cutting-edge intra-day futures systems—offers advanced quantitative traders valuable lessons in diversification, robust strategy design, and the relentless pursuit of system improvement. As markets continue to evolve with ever-faster data and more players, Bob’s insistence on speed, flexibility, and rigorous risk management remains as relevant as ever.
Trade well and prosper, and of course, reach out with questions, comments or suggestions for futures shows!
Get in Touch with Bob Pardo