Heuristics vs expertise

Here's a nice piece by Jarndyce pointing out that, in football betting, heuristics beat the experts.
Not just football, of course. Gerd Gigerenzer points out in this book and this paper (pdf) that simple rules of thumb do well in lots of other contexts.
So why do people ignore this, and continue to rely on pundits? My day job has thrown up one answer.
A few days ago, I wrote this, pointing out that big discounts on investment trusts can lead to good returns. This is, of course, not original. It was pointed out by Richard Thaler and colleagues 16 years ago (pdf), and a big literature has since emerged on the subject, of which this is the latest addition.
So, we have a simple heuristic: big investment trust discounts often (not always) lead to good returns.
And I've gotten letters of complaint. Things are more complicated than this, they say. Trusts are risky. You need experts to help you understand these risks.
This is guff. But it's powerful guff. There are three reasons why people believe it:
1. An inferential error. Of course, things are more complicated than a simple rule suggests. But it doesn't follow that experts can help us make better predictions. Quite the opposite. They may be prone to the illusion of knowledge; they become more confident in their judgments than the data warrants.
2. The representativeness fallacy. People think a good picture of a complicated thing must be highly detailed. Not necessarily. If you want to find your way out of a forest, the rule "go straight ahead" can often be better than a detailed account of every tree.
3. Wishful  thinking. People want to believe that their expertise matters,not least because their livelihood depends on it. Did I mention that all the complaints came from financial advisers?

Comments

[dearieme]

I don't want "expert advice" on the unknowable - will HSBC do better than Barclays. I want it on technicalities: the investors in "split" investment trusts might feel the same, or all the Life Insurance and Pension investors who'd never heard of Market Value Reductions.

[CJC]

Is what you describe in #2 actually the representativeness fallacy?

[Gael Ovide-Etienne]

James Surowiecki, with his book "The wisdom of the crowds" talks about experts versus the crowds and its wisdom. It basically says that experts are usually not as accurate/right as the average answer made from a crowd. the examples in the books are stunning!

I strongly recommend this book!

[rjw]

interesting. seems to me though that the line between a useful heuristic and a cognitive bias might sometimes be a very fine one. all in that magic phrase - 'what the data warrants' ?

[Jason]

I think you need to be careful. Intuition is not the be all and end all solution to everything, there are a bunch of things that need to be in place for it to work well. Primarily, any feedback needs to be rapid, correct and correlate to some recognizable pattern (we often "recognize" patterns in what is, in fact, utter randomness).

My general experience is the other way around. I find people far too often prefer simple explanations to complex ideas (the market is down b/c of oil prices, the S&P 500 is a representative index, etc.) and stick to those.

[skin care]

nice
http://www.skincareinfo.us/

[freeastrology]

nice
http://www.freeastrology.us/

[exporters]

good
http://www.exportersinfo.com/