Algorithmic trading, black box trading, computerized trading, systematic trading. The words conjure different thoughts and feelings depending on who you are, how much you know, and where you sit as a participant in capital markets.
To the retail trader or the Average Joe, algorithmic trading might be mysterious, scary, or unfair –
“Aren’t algorithmic traders responsible for those flash crashes I am always hearing about”
“Algorithmic trading is something only reserved for the big banks and Wall Street insiders who have deep pockets – its just a way that the rich get richer by using an unfair advantage.”
“What the hell is an algorithm?"
To the fundamental trader, who bases his analysis and investment choices on forecasts associated with factors like global-macro trends such as interest rates and GDP or company-specific factors like cash flows, the state of a balance sheet, or management changes, it might seem outlandish or appear as a threat –
“Those guys base their investment decisions purely on statistics without regard to the fundamental values of a business – they actually risk money on that stuff?”
“Making investment decisions based on coded rules is too simple to work – it cannot work.”
But despite these common reactions, a great portion of the population is actually trading using systems and algorithms in one way or another.
What exactly is an algorithm and how can it be applied to trading and investing? Well an algorithm is simply “a process or set of rules to be followed in calculations or other problem-solving operations, especially by a computer” (as per Google).
So algorithmic investing is simply devising a set of rules for how you will buy and sell stocks, bonds, commodities, or any other financial asset and then executing on those rules.
Own a mutual fund or ETF that tracks the S&P 500, Dow, NASDAQ, or any other number of the myriad indices across the world? Congrats – you are an algorithmic investor (at least to the extent that you are using an algorithm to select part of your portfolio - algorithmic asset allocation across an entire portfolio is another topic altogether, which I will address extensively in future posts).
In order for a company’s stock to be in the S&P 500, it has to have a certain market cap, be headquartered in the US, it must meet certain turnover requirements (trading liquidity constraints), most of its shares must be publicly held, it must have been public for at least six months, and it must have four straight quarters of positive as-reported earnings. Source
Own a mutual fund or ETF that tracks the Barclay’s Aggregate Bond Index (the most widely used bond benchmark in the US)? Congrats – you are an algorithmic investor. (“Barclays Aggregate Bond Index is a market value-weighted index that tracks the daily price, coupon, pay-downs, and total return performance of fixed-rate, publicly placed, dollar-denominated, and non-convertible investment grade debt issues with at least $250 million par amount outstanding and with at least one year to final maturity” Source ).
There are lots of ways to design an algorithm – a set of rules for how to buy and sell stocks, bonds and commodities. The industry standards here in the US like the S&P 500, the NASDAQ, the Dow and the Russell 2000 are adequate tracking or measuring index algorithms – they do a great job of tracking and measuring what the US stock market is doing as a whole.
But for as good as these algorithms are at tracking the stock market, they are decidedly inferior at generating strong returns and minimizing risk relative to other, more sophisticated algorithms. Thus, on their own, they are generally not great investing algorithms.
They don’t have any meaningful risk-mitigation strategies, they hold on to stocks in long term downtrends, and they are over-diversified to the detriment of returns and risk.
As investors, our goals are to squeeze out the highest return for a given level of risk taken, and the industry standard tracking algorithms like the Dow and others are inferior relative to more sophisticated algorithms for actually investing and trying to manage risk.
Want an example? Here at The Intelligent Allocator, we have designed a simple algorithm designed to significantly outperform industry standard tracking indices on both an absolute and risk-adjusted basis (see the Alpha Momentum Strategy Overview for more details).
At The Intelligent Allocator, when we hear the words “Algorithmic Trading” or “Systematic Investing,” we think things like:
Offering the best chance of future success.
And if we can find algorithms that are simple, efficient, repeatable, testable, robust, and that have historically outperformed industry standard investing benchmarks substantially, then these are algorithms that we want to employ for our own trading accounts and share with others.
More to come.