Eight studies every mechanical trader should know

A lot of what is said about trading is opinion. Some of it has been measured. These are eight studies I think every mechanical trader should know, grouped by the question they answer. For each one: what the researchers did, what they found, and what it means in practice. Figures are quoted only where the paper or its published abstract states them.

Can individuals day trade for a living?

1. “Day Trading for a Living?” Fernando Chague, Rodrigo De-Losso and Bruno Giovannetti (SSRN 3423101). The authors followed individuals who day traded Brazilian equity index futures from 2013 to 2015, using data from the country’s securities regulator. Among those who persisted for at least 300 days, 97% lost money. Only 0.4% earned more than a bank teller’s starting salary, and the authors found no evidence that traders learned and improved over time.

2. “The Cross-Section of Speculator Skill: Evidence from Day Trading.” Brad Barber, Yi-Tsung Lee, Yu-Jane Liu and Terrance Odean (Journal of Financial Markets, 2014). Using the complete day trading records of Taiwan from 1992 to 2006, they found that less than 1% of day traders were able to predictably and reliably earn positive returns net of fees.

The default outcome of day trading by feel is losing money. Not a slower profit. A loss.

What it means: the starting assumption should be that an untested approach has no edge. That is not pessimism, it is the base rate. A method you can test is how you find out whether you are in the small group that does have one.

Can you trust a backtest?

3. “The Probability of Backtest Overfitting.” David Bailey, Jonathan Borwein, Marcos López de Prado and Qiji Jim Zhu (SSRN 2326253; Journal of Computational Finance, 2017). The paper argues that the usual protections against overfitting, such as keeping one period aside for testing, are unreliable for strategy backtests. It proposes a method (combinatorially symmetric cross-validation) to estimate the probability that the best of several backtests is overfit. The key idea: the more variations you try, the more likely your best-looking result owes its shine to chance.

4. “Backtesting.” Campbell Harvey and Yan Liu (Journal of Portfolio Management, 2015). When many strategies have been tested, a reported Sharpe ratio must be cut to account for the testing itself. The cut is not a flat 50%: strong results are only moderately penalised, while marginal results are cut heavily. The authors call the common 50% rule of thumb a serious mistake.

What it means: count your tries. Every parameter you tweak and every filter you add is another test. Keep the rules simple, decide what you will test before you look, and keep data the system has never seen.

Do simple rules work over the long run?

5. “A Century of Evidence on Trend-Following Investing.” Brian Hurst, Yao Hua Ooi and Lasse Heje Pedersen (AQR; Journal of Portfolio Management, 2017). The authors built a simple time-series momentum strategy across equity indices, bonds, currencies and commodities and tested it back to 1880. Trend following delivered positive returns in every decade they studied, with low correlation to traditional assets, and it did well in most of the worst periods for a traditional stock and bond portfolio.

What it means: an edge does not need to be clever. A rule anyone can write down (“own what has been going up, short what has been going down”) survived more than a century of wars, crises and regime changes. Robust usually looks simple.

Why discretion is hard: the human side

6. “Are Investors Reluctant to Realize Their Losses?” Terrance Odean (Journal of Finance, 1998). Using the records of 10,000 accounts at a discount broker, Odean found that investors sold winners far more readily than losers. The habit was not justified by what happened next: the winners they sold went on to do better than the losers they kept. This is the disposition effect, and it is exactly what a fixed exit rule removes.

7. “The Psychophysiology of Real-Time Financial Risk Processing.” Andrew Lo and Dmitry Repin (Journal of Cognitive Neuroscience, 2002). They measured skin conductance, heart rate and other signals of ten professional traders during live trading. The body reacted to market events and to volatility, and the patterns differed between more and less experienced traders. Emotion is part of how traders process risk, even at a professional desk.

8. “Endogenous Steroids and Financial Risk Taking on a London Trading Floor.” John Coates and Joe Herbert (PNAS, 2008). Following 17 traders over eight days, they found that a trader’s morning testosterone predicted that day’s profitability, and that cortisol, the stress hormone, rose with the variance of the trader’s results and with market volatility.

Your body is part of every discretionary decision you make. A rule is the only part of the trade it can’t touch.

What it means: the urge to cut a winner, hold a loser or press after a good day is not a character flaw. It is biology, and it is measurable. Small samples, yes, but they point the same way. Training it away is slow and uncertain. Moving the decision to a rule written in advance takes it out of reach.

Putting it together

• Most people who trade by feel lose (studies 1 and 2). Assume no edge until you can show one.
• Showing one is harder than a pretty backtest (3 and 4). Count your tries and test on unseen data.
• Simple, robust rules can work for a very long time (5).
• The hardest part of trading is the human doing it (6, 7 and 8). Rules decided in advance take that part out of the moment.

Sources

• Chague, F., De-Losso, R. and Giovannetti, B. “Day Trading for a Living?” SSRN 3423101; FGV EESP working paper.
• Barber, B., Lee, Y.-T., Liu, Y.-J. and Odean, T. “The Cross-Section of Speculator Skill: Evidence from Day Trading.” Journal of Financial Markets 18 (2014).
• Bailey, D., Borwein, J., López de Prado, M. and Zhu, Q. J. “The Probability of Backtest Overfitting.” SSRN 2326253; Journal of Computational Finance 20(4) (2017).
• Harvey, C. and Liu, Y. “Backtesting.” Journal of Portfolio Management (2015).
• Hurst, B., Ooi, Y. H. and Pedersen, L. H. “A Century of Evidence on Trend-Following Investing.” Journal of Portfolio Management (2017).
• Odean, T. “Are Investors Reluctant to Realize Their Losses?” Journal of Finance 53(5) (1998).
• Lo, A. and Repin, D. “The Psychophysiology of Real-Time Financial Risk Processing.” Journal of Cognitive Neuroscience 14 (2002).
• Coates, J. and Herbert, J. “Endogenous Steroids and Financial Risk Taking on a London Trading Floor.” PNAS 105(16) (2008).
• These are summaries for education, not investment advice.