The diagram below shows the major areas of research used by traders and investors to identify and exploit trading edges. The approach is often dictated by the chosen investment objective (performance vs. out-performance) and timeframe (short-term vs. long-term). For example, long-term investors looking to consistently outperform the market are likely to focus their research on fundamental and/or sector analysis. Intermediate-term traders seeking consistent performance may take a quantitative or technical approach to their research work, while very short-term traders looking for regular profits will likely focus on developing algo/HFT systems.
Most mutual funds use fundamental analysis to select stocks for their portfolios. Fund managers and “value investors” pour over financial statements and attempt to find what other’s have missed. However, considering that only a small minority of funds manage to consistently outperform the market, one must wonder whether trading profitably using fundamental analysis is a realistic goal for most investors, private or institutional. If the brightest minds in the largest financial institutions can’t consistently beat the S&P500, who are we to think that we can do better?
It would be naïve to think that the average independent trader can use fundamental research to consistently outperform the market, let alone to consistently make money. On the other hand, an investor wishing to consistently outperform the vast majority of mutual funds should consider investing in the market itself by buying into a mirror fund or a low-fee S&P500 ETF (eg. the SPY).
Traders with a strong background in economics may be tempted to seek outperformance by looking at the big picture – analyzing the relative strengths of different market sectors, economies or even continents (comparing emerging markets vs. Europe vs. North America, for example). Domestically, the smart money has been known to find outperformance by predicting macro-economic changes (e.g. the US housing crash) or to profit from the market bias that can result from government intervention (e.g. Quantitative Easing).
Marco-economic analysis is often used by hedge-funds, who have the ability to tap into a multitude of investment vehicles both at home and abroad, and who can profit from both long and short positions. A number of these funds do manage to outperform the market consistently. However, trades based on macroeconomics are few and far between, plus the investment timeframes tend to be long.
Traders here seek to exploit major shifts in market segments, often driven by technological innovation and a resulting change in consumer behaviour. Investors bet on companies, or groups of companies, that show promise and bet against those that don’t. These decisions are not based on fundamental analysis, although this may also be used, but rather on the expectation that a major market shift is taking place.
An example of a pair-trade based on sector analysis – in the publishing industry – is to have gone long Amazon and have gone short Borders. Another example – in the entertainment industry – is to have gone long Netflix and gone short Blockbuster. Betting on the (now obvious) shift from brick & mortar trade towards online sales would have made the insightful speculator very rich.
Trading profitably using sector analysis is within reach of most people. What is needed is the ability to see the big picture, plus the conviction and self-confidence required to put these trades into play. Unfortunately, similarly to bets based on macro-economic research, sector analysis trade setups tend to be infrequent, so it’s not realistic to expect predictable and recurring returns from this style of trading.
Technical analysis offers a very useful way to graphically represent market events: recent highs, bullish periods, areas of resistance & support, changing trends, etc. Moreover, it can provide a way to categorize otherwise subjective market conditions. For example a trader can chose to define a bull market as a S&P500 close above the 200 day moving average, or a 100 day MA above the 200day MA. A “breakout” can be defined as a close above a Bollinger band, or maybe a close above a 50 period high. This ability to clearly define specific market conditions allows the trader to use technical analysis systematically – that is, as a trigger or filter in mechanical systems.
One of the major strengths of technical analysis is that many market participants either actively use it or at least track it, making it something of a self-fulfilling prophecy. For example, traders in a long position may decide to take profits at a key resistance level tracked by most market participants (yesterday’s high, pivot point, 10 day moving average, etc) because they expect the price will hit a ceiling and possibly revert downwards. These traders will therefore sell their positions, thereby applying a downward pressure on price. So the mere expectation of resistance has helped create the resistance itself.
At its worst, technical analysis is the astrology of finance, a world full of gurus with a colorful lexicon (shooting star, inverted hammer, doji), all experts at explaining market movements after the fact. Indeed the problem with technical analysis is that only rarely can it be used to generate actionable trading signals. Moreover, people can see what they want in a chart. So it is important not to get caught up in the mumbo-jumbo, but rather focus on the basic technical aspects that can add value to a trading plan, such as determining trend and recognizing some key resistance levels.
Quantitative / Algo (execution edge)
In tandem with the rise of the machines, the past two decades has witnessed the rise of the “quants”. Quantitative analysts, who used to fill back-office or risk-management support roles for trading desks, have seen their influence grow substantially. So much so that there are now firms whose trading activities are exclusively based on algorithmic trading.
Algo-trading firms make their money with “execution edges”. The market inefficiencies behind these edges are often extremely short lived so they can only be exploited through high-speed trading. The competition for these small but recurring profits is fierce and the resources employed by the large market participants are huge. So this arms race is simply beyond the reach of the average trader, however clever and well-equipped he may be. Even some of the most potent names in the industry have at some point been humbled by the very technology they employ, either because they got overrun by the complexity of the system (eg. Knight Capital in 2012) or because their trading algorithms fell apart during times of extreme volatility (eg. Citadel Tactical Trading and many others in 2008).
It has also been suggested that algorithmic edges are getting increasingly difficult to exploit as more and more market participants have got in the game. Moreover, some of the more controversial techniques used by algo-trading firms (eg. spoofing) came under scrutiny and were deemed illegal after the Dodd-Frank financial regulation law of 2010.
In short, statistical arbitrage, high-speed trading, market-making and other execution-edge trading techniques are best left to those with the resources needed to execute them.
Quantitative (statistical edge)
Another major field of quantitative analysis focuses on finding statistically significant patterns in historical data. The objective here is to identify a pattern (either by testing a hypothesis or empirically), quantify its statistical relevance, and then determine whether the edge is actually “tradable” – that is, whether it can be used to influence or trigger a trading decision. These patterns can be seasonal, trending, breakout or mean-reverting in nature. The metrics typically used in pattern analysis are price-action, volume, breadth and relative-strength.
Statistical edges are surprisingly prevalent in stock market time series, irrespective of what time increments are being studied. Seasonal bullishness or bearishness, for instance, can be found in monthly charts (some months are more bullish than others), daily charts (some days have a tendency towards “continuation”, others towards “reversal”), and even 30 minute or 5 minute charts (periods of the day have their own trading personalities). Recognizing these bearish or bullish tendencies allows traders to know in which general direction the wind is blowing, and adjust their trading plan accordingly. Moreover, compounding these small seasonal edges – e.g. going short when the market is in a downtrend, during a generally bearish month, during a generally bearish week day during a generally bearish time of day – can result in an attractive risk/reward proposition. Statistical edges can also be found when studying recurring events, such as option expiration weeks, FOMC days, macro-economic announcements, etc.
The three major concepts quantitative analysts use to find statistical edges are trend-following, momentum and mean-reversion. All three are powerful forces. Some work best with equities, others with forex or commodities. Mean-reversion strategies are particularly suited to “swing-trading” timeframes on stocks, while trend-following strategies tend to work better on longer timeframes. But some traders are able to adapt these techniques to several different timeframes and financial instruments.