I came across an interesting paper the other day that look at the changing nature of delusions within populations presenting at US psychiatric hospitals.
The researchers found something very interesting but fairly obvious. The nature of delusions is changing in that it is reflective of the culture at the time. The authors found that –
…more patients after 1950 believe they are being spied upon is consistent with the development of related technology and the advent of the Cold War.
Delusional content tended to reflect the culture at the time, with focus on syphilis in the early 1900s, on Germans during World War II, on Communists during the Cold War, and on technology in recent years.
Indeed, delusions now are being reported relating to computers, the internet and computer games.
Delusion content across the 20th century in an American psychiatric hospital. The International journal of social psychiatry. 2011. Cannon BJ, Kramer LM
This sent me off thinking about delusions in general since traders are a fairly delusional bunch of individuals and I wanted to know whether there had been any work done recently not necessarily on the cultural basis for delusions but rather in people estimating or giving a measure for an activity they were performing.
I have long known that people are very poor statisticians and monitors of their own performance but I was specifically interested in estimates. The reason I was interested in the estimation of given events is that new traders traditionally overestimate what they will achieve in the near term. A consequence of this is that they underestimate the capital they need to fulfil their goals. An extension of this is the usual shonk ads you see telling people they can quit their day job and intraday trade currencies with only $10,000.
Whilst there is no actual data that I could find quickly on traders I could find a lot from the field of exercise science, particularly on the nature of weight loss. In particular I came across a paper called Normal weight men and women overestimate exercise energy expenditure Willbond SM et. al.. J Sports Med Phys Fitness. (2010) 50(4):377-84.
This paper looked at something very interesting – it tried to quantify how much people overestimated the number of calories burned in a given fixed exercise. The study recruited 17 subjects of which 16 completed the study. Those that completed the study were half were men and half were women aged between 20-35 who were considered to be of a normal weight. The participants were defined as having been only moderately active in the past 6 months and had stable weight.
Participants were given two different exercise routines which had been calibrated to burn either 200 or 300 calories. At the end of the session, the participants were asked to estimate how many calories they had burned. This is where is gets a little odd and shows how delusional people are. The range for the 200 calorie exercise ranged from 120 to 4000 calories. The average estimate for the 200 calorie bout of exercise was 825 calories. The average for the group was out by a factor of four and the outlier of the group was out by a factor of 20. The results for the 300 calorie trial show similar poor estimations.
Take a deep breath and think about this – all these people had done was walk at a brisk pace for 28 minutes. To put this into context to burn 4000 calories you would probably need to run a marathon in under 2 hours and 30 minutes marathon carrying me on your back. This is a profound overestimation yet even the average guess was out by 400% – so much for the wisdom of the crowds that social networkers go on about. The crowd in this instance were pretty dim.
The question of estimating calorie expenditure is an interesting one because it is in many ways the flipside of asking people to estimate their calorie intake. Most people underestimate their intake by somewhere between 30% and 50% depending upon the study and the test group.
So, how does this relate to trading?
The answer can be found in reframing the question. Imagine you give a series of traders a demo account where they can trade any instrument they like. After the trial period you ask them how well they think they would do in trading for real and I can guarantee that the results will be probably as far from reality as when people are asked to estimate how many calories they burn in a given period. Overestimation is in my view a function of overconfidence and much work has been done on overconfidence in traders (see “Boys will be Boys: Gender, Overconfidence, and Common Stock Investment” with Brad Barber, Quarterly Journal of Economics, February 2001, Vol. 116, No. 1, 261-292.)
The notion of overestimation is an important one for traders because all too often we see traders quit their day jobs and head off into the field of full time trading with insufficient capital. Their decision having been based either on a demo account, a share trading competition or some dodgy course they have done. The amount of money needed to trade full time is sobering for many but for some a dose of reality proves to be no impediment at all. After all how hard could it be when you have managed to trade a demo account for four weeks without completely blowing yourself up.