Why it might be wise not to trade too much in the afternoon

A few years ago, I met with a new coaching client at a hedge fund. This individual defied every stereotype of how one might assume a hedge fund portfolio manager may be. Despite a 20-year history in the markets and a stellar track record, he was a very nervous and anxious individual. Unlike many of his colleagues at the fund, who mostly took medium- term relative-value type trades, this individual day-traded FX, something generally assumed the domain of the home retail trader. What was all the more remarkable was his incredible track record. In a typical year he made around USD 40-50 mio of profit, this year he had made around USD 100 mio, a large chunk of it in the space of a few moments when the Swiss pulled their peg against the Euro. What really stood out for me though, was that he never traded after 1 pm.

I asked him, “What if you have a strong signal in the afternoon which suggests taking action, wouldn’t it mean that you will be leaving a lot of money on the table by not doing anything?” He replied that that was often the case, however he was still better not to act, he said overall he believes that historically his afternoon trading provides him with a net loss, even if you included these trades.

This fascinated me because it was a live example of a theory that I had about afternoon trading. This theory was formed from research I had read about our how our ability to make effective decisions and to exert self-control in our actions diminished the longer the day went on. In a trading sense, this could mean that whilst a trader may have an edge in early in the day, as the day wore on, that edge was eroded to the point where it became a negative edge.

Decision fatigue, mental bandwidth and trading

The research which led me to this belief was produced by a group of psychologists who had analysed the decision-making performance of judges. The research analysed the decision-making capabilities of a group of parole board judges when it came to granting or denying parole of over 1000 cases over a period of nearly a year. Their decisions were aggregated and placed into a single time of day chart. This chart can be seen below:

A positive or good choice was given a score of 1, a sub-par choice ‘denying parole’ was given a score of 0. The quality of the choices can be seen to deteriorate as the day wears on, only improving for a short period after intraday breaks.

This chart and their research highlight the role of ‘Decision Fatigue’ in people’s decision-making. The mental work of ruling on case after case, whatever the individual merits, wore the judges down leading to lower quality decisions as the day moved on. This sort of decision fatigue makes people involved in repetitive high stakes decision-making prone to dubious choices as the day progresses.

This sort of fatigue differs from ordinary physical fatigue where you are conscious of your tiredness. With decision-fatigue, your mental bandwidth is stretched, but you are not conscious of it and do not feel it. Thus, you carry you carry on making decisions and choices regardless, though you do become more irritable, and do make progressively lower quality choices.

Decision fatigue usually manifests in one of two ways.

The first of these is poor quality or even reckless decisions where one acts impulsively instead of expending the energy to first think through the consequences. We see these in knee-jerk trading responses, and unplanned reactive trades. The other manifestation is ‘Decision Avoidance’, not taking an action but procrastinating or hesitating.

Time of day quality in FX trading markets

The following chart looks at the percentage of trades in five currency pairs that closed with a profit at broker FXCM from 10/01/2009 to 09/30/2010. I have assumed that most of these trades refer to day trading clients.

The most profitable period of the day was from midnight to around 5am NY Time. This is around 6am to 11am Central European Time. Since by far the highest proportion of Forex trading occurs in the European time-zone, this suggests that most Forex traders were profitable in the first few hours of their trading day. However, as the day goes progressed the trader’s profitability levels drop.

Levels of profitability appear to return to balance late in the evening, which one presumes coincides with much lower activity levels post-US markets and into the Asian/Pacific time-zone.

We cannot know the reason for this with any certainty, but this does seem to lend some credence to the possibility that ‘Decision-Fatigue’ could be at play.

Returning to the story of the Fund trader. Seen in this light, his practice of not trading into the afternoon seems a wise one. He is a seasoned trader; he knows when he is strong and when he is compromised. He knows he has no edge in the afternoon, why play when you have no edge, for him it is just playing to lose.

---