The great “Man vs. Machine” debate in the financial markets – The verdict is in!
In our view, The Man vs Machine Debate in the financial markets, has always been a heated one. But we feel it’s changing and that is not really even a debate anymore! We are going to explain why.
But First… We view Machines as the clear winner and believe that this trend will only strengthen going forward.
Don’t get us wrong, being human, we are “cheering for people”. But the evidence is getting impossible to ignore; each year machines (algos) are continuing to out-perform manual traders, exponentially.
If you asked us this even 2 years ago we would have said it’s a “toss up” but the data is now overwhelmingly one-sided; the amount of volume switching to algo sources also strongly supports this.
Traders are simply going with what works. According to both EBS and Morgan Stanley via the financial times, the ratio of algo to manual traders in 2010 was approx. 70:30, in 2012, it is 84:16, in 2014 it was 89:11. This creates a lot of distortion in the market, and changes how it behaves – magnifying an already arduous problem and increasing the difficulty for manual traders to navigate.
Fewer manual traders are able to out-perform algos and we expect this trend not only to continue but to exponentially intensify. This doesn’t mean that the “human element” is entirely removed; algo trading isn’t some sci-fi movie robot sitting at a desk and trading the markets, but rather proud/brilliant/creative teams of coders and mathematicians essentially looking to “solve the problem” of how to profit from the markets using a unique toolset.
These people are writing computer programs to do the heavy lifting for them; they are still deeply engaged in their work but don’t require the same amount of “screen time” and most importantly they don’t need to depend on emotion (trading is one of the few tasks in the world where emotion is proven to be a detriment).
There are things that computers will likely never do better than humans – i.e., creating art, making music, writing etc… However, trading is not one of them. Moores Law of exponentials clearly proves that computing power globally is growing at an unprecedented pace, and not surprisingly, so have powerful quantitative programs trading in the markets across every single asset class – from equities to currencies to sports betting.
Hold on tight, as the ride may get even wilder. Any good fundamentalist will argue that computers cannot cope with or understand complex global geopolitical and financial situations. However the ability for quants to crunch, model, and forecast this data has greatly improved in just the last year alone, and continues to advance at an astonishing rate.
During some of our research on this subject, we have been blown away by the recent advances taking place which will be impacting us all in one way or another. For example; in 2014 bioengineers at Stanford University developed a circuit modeled on the human brain. Sixteen “Neurocore” chips simulate one million neurons and billions of synaptic connections, claimed to be 9,000 times faster as well as more energy efficient than a typical PC. This year, Intel and Micron announced 3D XPoint, a non-volatile memory claimed to be up to 1,000 times faster, up to 1,000 times higher endurance and similar in density compared to NAND. Production is scheduled in 2016.
We also now have freely available, open source AIs (artificial intelligence) – which any kid with an internet connection can access. Genetic algo building and neural networks are becoming common tools for clever people and organizations to use to solve a vast array of problems. Lastly we have D-wave quantum computers currently being sold out of Canada which are not even fully understood by their developers in terms of how they function – however the first commercial quantum computer is going to be available before this year is out.
Naturally, some of the most brilliant minds will be (and already are) applying these advances to the financial markets. We understand that one does not require this degree of sophistication in order to create a successful algo (in fact some of the best algos employ the simplest of strategies), but there is no denying that these tools are becoming more easily accessible than ever before in our history, making them readily to the average person, and also the average trader.
In light of our research and careful analysis we strive to keep our entire portfolio to be upwards of 90% algo based. While it will be mostly systematic, each of our algo teams we work with include a significant and continual human adaptive overlay which we highly value.