JAMES HARRIS SIMONS

HOW HIS TRADING STRATEGY GENERATED 66% P.A.

Introduction

James H Simons or Jim Simons, the founder of American Hedge Fund, Renaissance Technologies is famously known for generating an average annual return of 66% by one of the funds of his firm and the same is responsible for catapulting him to become the wealthiest hedge fund manager in America. What is even more impressive is, he managed to generate this return over a period of thirty years from 1988 to 2018. Renaissance Technologies has amassed an Asset under Management of $75 Billion and is a trend setter when it comes to Quantitative Trading.

This post delves into Simons’ contribution to the world of Quantitative Trading, the trading strategies and associated reasons for Renaissance Technologies’ success and the consequential lessons that emerge for new age investors.

Quantitative Trading – LAYING THE FOUNDATION

From a time when one had to graph charts by hand and physically call the brokers to place a trade, equity trading has come a long way. Though the transformation mostly happened after the advent of computers, just the presence of technology does not guarantee any change in a field, one needs to be proactive and know how and when to use it. And that was exactly what Simons did with Equity Trading.

In today’s times, Quantitative Trading or using programs and machine learning algorithms to execute trades at lightning speeds is not that uncommon and hence this article because a vast crowd can relate to it. In fact, according to Mordor Intelligence, about 60-73% of the overall U.S. stock market trading in 2021 was algorithmic.

But when you dial back 3 decades, you can understand that this way of executing trades was naturally not that common, in fact it was unheard of and that is when Simons’ Renaissance Technologies, in the mid 70s, forayed into Quantitative Trading and that laid the foundation stone for what we are witnessing today.

Amongst the funds that Renaissance Technologies manages, the most popular one happens to be the Medallion Fund which was established in 1988 and has generated a gross average annual return of 66% till 2018. 

What’s the big deal in that you ask?

Well, to provide a context, Warren Buffet’s Berkshire Hathaway, generated average annual returns of around 16% over the same period and the S&P 500, one of the most commonly followed stock indices and which houses the top 500 companies in the USA, generated an average annual return of around 12% since 1957, when it was introduced.

Hence, you can see what Medallion Fund has managed to achieve and because of the same, it has gone on to become one of the most successful hedge funds in history. Its no wonder that the fund charges a very high management fee and the post fee annual return for the investors of this fund averages at around 40%, which is still excellent compared to the competition.

But all of this begs the question – How did Simons manage to eke out that kind of returns? A broad answer would be his foresight and how he decided to leverage computers and algorithms at a time no one was doing it. Because of Renaissance Technologies’ early adaptation of computers in its trading strategies, by the time Medallion Fund was launched, it was able to make innumerable split second trades in the equities and futures market and they kept improving and advancing their strategy as technology improved over the years.

How did he do it – A DEEP DIVE

Let’s delve a bit deeper into the reasons and investment strategies that propelled Medallion to the height it is sitting at today.

Not to undermine the importance of human emotions in Trading, but Simons’ achievements build a case in favor of keeping human emotions away when making trading decisions. Simons, when entering trades based on repeated market anomalies, did so purely on the basis of quantitative data.

However, due to obvious reasons, there is limited public data as to what Simons’ exact strategies were but that didn’t stop experts and mathematicians from making attempts in trying to decode his strategies.

1. Portfolio Level Statistical Arbitrage

A mathematician by the name of James Bakers propounded that Simons used something called as the Portfolio Level Statistical Arbitrage in order to achieve the remarkable results that he did. This particular strategy combined Statistical and Econometric techniques to reduce market risks and make consequential profits.

2. Deja Vu

Gregory Zuckerman, a special writer at the Wall Street Journal and a non-fiction author mentioned in his book “The Man Who Solved the Market” that Medallion’s profits can be attributed to “Trending and Reversion Predicting Signals”, particularly a one called Deja Vu and that the fund used complex signals and equity trades instead of simple pairs.

3. Ghosting Market Prices

Before starting a hedge fund, Simons worked as a NSA Code Breaker for the Pentagon and the core idea behind his team’s work was to look for patterns others failed to detect. This was replicated by a top researcher in Simons’ team in the mid 1980s wherein hidden or ghost patterns were identified in market prices and the same was most probably incorporated by Medallion as well.

4. Trend and Quant Fund – Mixing it Up 

The Medallion fund used a combination of Trend Fund and Quant Fund strategies. The Medallion Fund features a collection of a minimum of 8,000 signals based on the short-term patterns of the entire batch of scientists who work for Renaissance Technologis. The Fund then leverage these signals to trade in the markets innumerable times per day, with a trading reach that covers exchanges around the globe.

To re-focus on the Quant side, Simons re-launched the fund in mid 1990s with a renewed focus on Quants and  generated 56% return in its first year.

Renaissance – STANDING OUT FROM OTHERS

In 1990, Renaissance clocked in a gain of $1 million in a single day for the first time and later on, it became a common phenomenon in the fund house. 

But was this common at that point for most funds?

Surely not and in order to get such out of the ordinary results, one needs to do things the same way as well. And Renaissance’s approach was no different.

A popular method of trading at that time was “Convergence Trading” and this was popularized by John Meriweather’s Long Term Capital Management Fund. This approach involved determining the price of financial assets based on complex mathematical models. Two relatively opposite priced assets would be chosen – one expensive and one cheap and respective positions would be taken in both. Then, with the expectation that their price would converge to their real level, the relevant asset would be sold and bought.

Renaissance however did not go this route at all and they focused on trades paying off within a limited time frame. And based on this model, their traders conducted rapid fire trading on numerous US and International Future Contracts and the asset class included physical commodities, financial instruments, essential currencies and equity and mortgage derivatives.

Renaissance’s Thing – SETTING TRENDS

This is what the New York Times had to say about Renaissance after its subsidiary installed a direct trading link with the German Futures Exchange. 

As mentioned earlier, thanks to Simons’ foresight, Renaissance was clearly ahead of its competitors in Quantitative Trading. And as an extension of that foresight, the Fund never shied away from investing in infrastructure that would perform high precision trades simultaneously across the world.

A testimony to that fact was when in 2016, Simons decided to use Atomic Clocks, which are world’s most high-precision time instruments which would allow him to execute trades at multiple exchanges with synchronicity to the billionth of a second. And since this was a first, he had to file a 16-page technical document for the same with the U.S. Patent and Trademark Office.

One Model – TO RULE ALL

Instead of developing different mathematical models for different asset classes, a single model, sort of like a Master Model was developed that could be used to  trade different asset classes. This was developed by Henry Laufer, the Vice President of Research at Renaissance.

But how is this important? Because having a single model ensures a central repository of a large volume of data that can be leveraged for trades across all asset classes and running model correlations across them also becomes a possibility. Furthermore, due to the flexible nature of the model, any new ideas, improvements or fixes can be continuously incorporated and it keeps getting bigger and better over a period of time.

More than One – THE TEAM

To achieve anything significant, you always need hands because there is only so much one can do single handedly. And a big credit for the feats Renaissance Technologies and Medallion have achieved goes to its team. 

Given his mathematical background, Simons’ approach towards Medallion was more from a Mathematical point of view rather than Finance and that mindset showed in his team selection as well. Instead of hiring people having Wall Street experience, Simons always hired people having a background in programming, cryptography and science. He even went on to hire Computational Linguists with experience in building speech-recognition computers.

Some of the key members in his team included talented scientists and mathematicians like Lenny Baum, James Ax, Elwyn Berlekamp, Henry Laufer, Peter Brown, and Mercer.

About Jim Simons

James H Simons studied mathematics at MIT and got his PhD in the same field from University of California, Berkeley at the age of 23.  After his PhD, he worked with the U.S. National Security Agency (NSA) to break codes during the cold war. However, he was evicted from the Soviet code-cracking team at the Institute for Defense Analyses because he was against the Vietnam war.

From 1968 to 1978, Jim Simons was a mathematics professor and chair of the mathematics department at Stony Brook University. In 1976, Simons was honored with the American Mathematical Society’s Veblen Prize, the world’s highest honor in Geometry, awarded every five years. He won accolades for his work in the fascinating field of differential geometry. His flagship work — the Chern-Simons theory, a theorem that he crafted with renowned geometrician Shiing-Shen Chern has received critical acclaim from the scientific community.

He quit the world of Academia in 1978 and founded a Hedge Fund called Monometrics. Monmetrics employed both fundamental and technical analysis to execute trades and though he was successful to some extent here, he decided to bring a change in his approach.

In 1982, Simons changed Monometrics to Renaissance Technologies wherein he went full in with Quantitative models, at a time Quant Trading as a concept was unheard of.

Renaissance Technologies handles four funds:

• Renaissance Institutional Equities Fund
• Renaissance Institutional Diversified Alpha
• Renaissance Institutional Diversified Global Equity Fund
• Medallion Fund

Amongst the four funds, Medallion was the one which caught everyone’s eyes and catapulted Simons to become one the greatest Hedge Fund Managers. In fact, such was the popularity and demand of getting into Medallion that Renaissance had to stop accepting money from external investors as the capital under management inflated to a massive scale due to compounding. Due to its size, the fund had trouble scaling the market signals, and it could not handle the volume without any volatility in its price.

Simons has had an immense impact in the world of Mathematics, Finance and Research and to acknowledge the same, in 2014 he was elected at the National Academy of Sciences (NSA), which is an elite body founded by Congress under President Abraham Lincoln to advise the federal government.

Simons also co-founded the Simons Foundation with his wife, Marilyn Simons, in 1994. Scientific research, education and health are the prime areas of focus for the foundation.

A part and parcel of Life – FAILURES

It wasn’t always a comfortable ride for Renaissance over the years. The global recession of 2007-08 led Renaissance Technologies to sell assets at rock-bottom prices in order to reduce losses. The quant models went haywire and did not function as they should have as they were unable to factor in widening credit spreads due to a plunge in real estate prices.

The situation was not rosy in the next recession as well which happened in 2020 due to COVID. The results produced by the Quant Models of Renaissance were not what normally people were accustomed to.

In a March 30 2020 filing, Renaissance said, “The beta models, which help determine portfolio exposure at funds for outside investors, in recent volatile markets have not performed as expected.” However, Simons believes that “The things we are doing will not go away. We may have bad years; we may have terrible years sometimes. But the principles we’ve discovered are valid.”  

Recently in 2021, according to Bloomberg, clients redeemed more than a quarter of Renaissance’s externally-managed funds. They pulled or asked to remove about $5 Billion from December 2020 to February 2021. Then they pulled out $6 billion more through June 2021. The redemptions were the highest in April but slowed in May and June.

What’s in it for Us – THE LEARNINGS

One of the biggest lessons that we have from Jim Simons and his work is the predictability of human behavior. When considered on a mass scale, humans are predictable beings and their decisions in business, politics, economics and all areas that directly or indirectly affect the financial market have a repetitive pattern. There will definitely be outliers as was with Simons as well – during the Global Recession and COVID, but when reckoned over a long duration, an element of predictability can be established.

Now, that is not to say that it is easy to understand and predict those patterns but it can be done and Simons’ quant models are an evidence of that. Backtesting assumes a huge role here as the past data is all we have and all the answers or patterns related to human behavior are hidden there. The more we observe, identify and understand these patterns, the more we bring in the aspect of predictability in our models.

Simons also showed that we need to rid ourselves of human biases by cutting off human emotions as far as possible when it comes to trade. In fact his scientific approach and pure quant models were meant to counter biases, both cognitive and emotional. They propose hypotheses, then test and use or review them to achieve a predetermined output.

The strategies also showed us that it is prudent to avoid illiquid stocks, options, futures and cryptocurrencies as money can be lost in the bid-ask spread. There could be issues associated with volume and you may not be able to exit a position if and when you want. It would be best if you also aimed to trade various signals on multiple asset classes.

And lastly, Simons’ journey shows us that no matter how intelligent or educated or experienced one is, he/she always needs the help of a group of people to make a significant impact in this world. He didn’t just restrict himself to the quest for spectacular returns, but he framed his success by building a highly-efficient system and an immensely collaborative team of top-notch researchers and mathematicians.

want to learn how to algo trade so you can remove all emotions from trading and automate it 100%? click below to join the free discord and then join the bootcamp to get started today.