Home > Economics, Finance, Investing, Science, Technology > High Frequency Trading vs Behavioural Finance

High Frequency Trading vs Behavioural Finance

*The following post has been edited for clarity from an essay I wrote.

I recently became a voracious reader of finance-related books. Being academically inclined, I would often refer to the original source, and many times contacted the authors directly when I needed further clarification. I found that most authors responded warmly. My relatively simple task of learning about ‘company valuation’ and ‘financial forecasting’ grew into a path of self-discovery. ‘Valuation’ led to ‘CAPM’, which led to ‘Modern Portfolio Theory’, which begat ‘Rational Expectations Theory’, which begat ‘Game Theory’ and ‘Efficient Markets Theory’, which led me to ‘Behavioural Finance’[1] and so forth.

Google Scholar and Amazon.co.uk became my best friends. Consumed by finance, I devoured every book and article I could find on the subject. Seeking the source, the man who started it all, I rummaged through the literature all the way back to Professor Edward Oakley Throp. Many regard Professor Throp as the ‘Godfather of Quants’ ̶ the first academic who used his applied mathematics skills in probability theory to beat the casino game of blackjack. Not content, he eventually took his skills to the biggest ‘casino’ of all: Wall Street, where his early work in warrant pricing paved the way for modern quants. I was mesmerised. Most people probably have not heard of Professor Throp, but they might have seen the Hollywood movie Bringing Down the House/21 or read the book. It is a true story based on the MIT blackjack team who took on Vegas and won. In the game of blackjack, cards dealt are not replaced. In mathematics, this is known as conditional probability. As the shoe (deck of cards) gets dealt the probability of the remaining cards can be established with greater confidence, tipping the usual 0.5% to 2% house edge into a positive edge for the players. Professor Throp was the first to exploit this and published his first ‘mathematical proof’ that the game blackjack was in fact beatable in the book Beat the Dealer. His work paved the way for subsequent MIT blackjack teams and also led to the growth of quant finance where academics took their research skills in applied maths to Wall Street. To this day, blackjack remains the only ‘beatable’ game in a casino, but casino operators have made it increasingly harder by using more decks and shuffling the decks (or shoe) more often. You may wonder why would casinos allow a game where it is beatable and still make money? Well, it is the allure of people who think they can beat blackjack by card counting (and don’t) that makes it worthwhile for casinos to keep that game. This is an interesting behavioural and psychological phenomenon. Even an experienced player playing perfect basic strategy gives up 0.5% on every hand. It takes a truly professional team to tilt the game in their favour. Even then, the player advantage is also minuscule and to make money, they have to play a long time (or leverage up) assuming they don’t get thrown out or chew on some knuckle-dusters first.

When I read Inventing Money by Nicolas Dunbar[2] I knew I had found my true calling.  Perhaps it was the allure of a career that combined elements of my love for physics and psychology with academia that drew me to the financial markets.

Today, as a ‘local’ (proprietary trader) on the NYSE Liffe and Eurex, I primarily trade as a speculative arbitrageur in Euribor futures and German two-year notes (Schatz). Trading as a liquidity provider, I employ a market neutral strategy in which my positions are usually hedged. However, such market-making strategies are the strength of black-box trading techniques such as High Frequency Trading (HFT) and algorithmic trading. Their stochastic models and execution are far superior to any human trader, and they can operate simultaneously in many different markets. I believe that proprietary traders like myself will therefore be redundant within five years.

Algorithmic trading now accounts for more than 60-70% of trading volume on the NYSE. This trend is set to continue as programme trading is also gaining a strong foothold in the futures and currency markets. The Bursa Malaysia (and other emerging markets) presents significant opportunities for such endeavours as these ‘virgin’ markets begin to embrace and incorporate more complex financial derivatives. Trading volume in the Bursa Malaysia is still primarily dominated by three (not profit-orientated) government-linked funds that operate through brokers known locally as Remisiers. Furthermore, CME Globex has recently offered to cash-settle Malaysian crude palm oil futures, paving the way for growth in this arena. (And the recent trend in major Financial Exchange consolidation represents another opportunity altogether.)

Behavioural finance relaxes the rigidly rationalist assumptions of financial economics by incorporating principles from psychology, anthropology and sociology into standard models of financial markets. Herein lies a conundrum. Implicit in the advantages of algo/programme trading system is the belief that human emotion is thoroughly eliminated. But imagine a universe where programme trading has come to completely dominate all aspects of our financial markets. Asset pricing is determined ‘rationally’ by computer models. Algo traders no longer design systems to capture minute mispricing anomalies arising from the fragility of humans  ̶  after all, in this universe there is no such need as humans are no longer active participants in the markets. It is a case of machines against machines. It necessarily follows that ‘behavioural finance theory’ may be ineffective in such a setting.

Both as a student of behavioural finance theory and as a trader watching and participating in the same markets increasingly being dominated by algo trading, I sometimes feel like I have a front-row seat at a ‘train wreck’ waiting to happen.  Hurtling down the fibre optics tracks is the train of data emanating from the co-located exchange servers of algo trading groups. Standing in its path since the 1990s is the research in behavioural finance by Professor Robert Schiller and Professor Richard Thaler, among others.

What would happen if, or when, our financial markets were thoroughly dominated by algo trading? Would behavioural finance be rendered obsolete? Would it spawn a new field of study known as ‘algo finance theory’ that perhaps incorporates AI and neural networks? Would the new ‘algo finance theory’ do unto ‘behavioural finance’ what ‘behavioural finance’ did to the ‘efficient markets’? Will efficient market theory rise like a phoenix from the academic ashes again once the machines take over completely (since algos are ‘rational’ and ‘more efficient’)?

The May 2010 flash crash put the media spotlight on the esoteric field of HFT.  Having spoken to HFT academics and practitioners, I’ve learned that many believe behavioural finance is a moving target  ̶  algo finance will never kill herd behaviour completely. Due to the time scale of HFT, the total dollar amounts are relatively small. HFT scalpers are already all over the Forex market, yet when the US employment figures or interest rate figures are released, the market goes ‘bonkers’, just like it always did. Instead, lower frequency computer trading (where the money stays on a trade for a long time), whether it’s human or computer, can drive the real changes in long-term price. For example, the Black Monday crash in 1987 (when HFT did not exist) was exacerbated by portfolio insurance strategies that used programme trading. Programme trading triggered stop-loss orders, which led to a downward spiral.

I want to bring all these strands together in research that will use the recent financial crisis, the flash crash of May 2010, the explosion of HFT, and the rush for Ultra-low latency Direct Market Access, among other factors, to explore the potential for growth of algorithm trading in emerging markets and the academic as well as investing opportunities that might result. No other subject piques my interest more.

[1] I was already familiar with works by Daniel Kahneman, Amos Tversky and Daniel Ariely, having studied psychology at college. However, revisiting their work from a completely new behavioural finance perspective was like rekindling an ‘old flame’.

[2] Inventing Money is the lesser known but more technical book covering LTCM’s demise (the other being When Genius Failed).



  1. May 14, 2011 at 10:13 am

    Wow,, That is awesome!!

  2. May 14, 2011 at 9:11 pm

    I never thought of it that way, well put!

  3. October 8, 2011 at 5:28 pm

    Recently I have seen a Movie named “21”..Its all about blackjack ..so those who wanna know about it more clearly can watch it..

    • October 9, 2011 at 12:37 am

      The movie is good, but I would highly recommend the book Bringing Down the House, which the movie is based on, and Busting Vegas. Both by the same author.

  4. G. A. Rutland
    September 27, 2014 at 6:33 pm

    From the perspective of one with similar interest, but reluctant to write about it, this is an excellent article even better with Prof. Thorp’s name properly spelled.
    Best Regards,

  1. No trackbacks yet.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: