Wayne Himelsein

Over my 25yrs in the game of quantitative investing, and drenched in equations and models, I've honed in on the only math that really matters: Minimize drawdowns during unfavorable times. The multiplicative nature of positive returns during good times does all the rest.

Someone recently asked me to “teach them quant”, which got me thinking - how do you really teach that? Sure, one can be taught math, modeling, testing protocols, etc. but as to being a quant, that’s just in you, or not. What makes me a quant, above all, is my drive to quantify.

A fun conversation…

A great irony of quant research is the idea that the data is full of possibilities. Yes, it is - the next innovation may be in there. At the same time, its full of constraints and obstacles; from bias to randomness to overfishing decay - an enemy that grows stronger with time.

Claude Shannon described information as “the resolution of uncertainty”- the view ahead may be fuzzy, but as more information rolls in, the resolution increases toward clarity. In financial markets (and life) the question for each of us is what resolution we need to take action.

The essence of the Black Scholes equation - that one can fully eliminate risk by buying/selling units of the underlying as it moves up/down - rests on the core assumption of a frictionless market. That’s like saying that a Porsche can do 0-60 in 1 second; friction is everything.

Quant finance tries to build reliable forecast models - a search for answers. More focus needs to be on the hunt for what can't be forecasted - a search for questions. “In Mathematics the art of proposing a question must be held of higher value than solving it.” -George Cantor

I've been thinking about thinking, e.g. the tendency to overthink in trying to optimize decisions. Resolved that it either improves outcomes or else adds resilience if the outcome turns out to not be ideal. "No problem can withstand the assault of sustained thinking." -Voltaire

When markets don’t behave as we want, or in a way that is not ideal for our sound process, the next best thing to do is to build tools that illuminate the details and disparities. “A good tool improves the way you work. A great tool improves the way you think.” -Jeff Duntemann

Trading recent markets, I’ve found benefit in being more nimble; that is, being more ready and willing to wholly flip an assessment. A good long can be a good short, or vice versa. The question is at what moment, and for how long? “Forever is composed of nows.” -Emily Dickinson

Trading this market can be quite frustrating, with historically high likelihood setups repeatedly failing. I keep myself sane by always remembering, great trading/PM skills are not built on the things that go right, but on managing the the things that go wrong.

Immersed in quant research and testing for decades, I’ve been bombarded by - and fended off - so many of those sneaky human biases. Amongst many, Causation is one of the worst offenders. Beware the tendency to assume how events are caused by other events - mostly, they’re not!

I can't even count the number of times that an idea of mine has been rejected. Ironically, a valuable lesson I've learned is that when critiques are the loudest, its more likely I'm onto something! “An idea thats not dangerous is unworthy of being called an idea.” -Oscar Wilde

Shorting volatility is sneaky and misleading; and not so different to a Ponzi scheme: Instead of paying early investors with new investor's money, you’re paying for today’s winnings with an accruing probabilistic loss — one that the Law of Large Numbers guarantees will come.

Unwinding Quant myth: Our best efforts are not necessarily in the hunt for new or better signals, but in digging deeper inside familiar ones. And our greatest effort may be in the extraction of flaws; by focusing on what hurts performance, one can find "errors" and reduce them.

The greatest misconception in Quant Finance is the power ascribed to the models. People seek the highest math, the best inputs and the choicest data, where all along, it's sound theory and assumptions that make it work. Algorithms don't make successful strategies, people do.

I come across many speculators who search far and wide for more information, more data, more opinions, to analyze a decision. But its not about knowing all there is to know, it's simply about knowing what's relevant: Signal quality over quantity, curation over accumulation.

Systematic models are highly subject to alpha decay, as widespread & hefty processing power relentlessly chase anomalies. Discretionary models less so, but highly prone to human error & discipline variance. My ultimate combo: constant discretionary innovation to quant systems.

A trader buddy of mine recently asked me: Do you think we have a natural bias when we trade? Yes! We trade our personality; optimists seek good longs, pessimists good shorts. Our style stems from who we are. But from there, we add necessary rules and grow, else lose and quit.

As I trade this wild market, vigorously aware of the raging war and potential consequences across the globe, I see, once again, that nothing much changes for my rules or process. The beauty of first principles is that no matter what happens "out there", they prevail "in here".

So much financial media (Twitter included) focuses on prediction. As if public content can offer valuable forecasts. Ironically, given market dynamics, any forecast is just a starting point. Profitability results mostly from how we react to what comes. Reaction over prediction.