Lucky versus skill

I’ve added the result of a quick random generation of possible outcomes of two types of individuals. This is not Monte Carlo simulation as has been discussed below, but rather a very simple illustrative example. This will further illustrate the points about in my previous post (and do also read the comments for some more discussion, and add your own thoughts and questions).

For this example, we have two types of people

  1. Traders who pursue a risk strategy by not managing risk and operating in an environment where randomness affects the outcomes in a large way.
  2. Dentistswho operate in areas where luck has little to do with the outcome. Hard work and studying pays off, but randomness has less to do with ultimate success.

For trader (in this example) feel free to use “entrepreneur” or “writer” with similar effect.

The outcome is on the y-axis (the vertical axes) so that more successful outcomes are higher up on the graph.

Dentists and traders
Four things can be seen quite clearly:

  1. The dentists have on average better outcomes than the traders
  2. The range of outcomes is much greater for the traders
  3. Overall, you’re probably better off being a typical dentist than a typical trader, and certainly better off being a unsuccessful dentist compared with a successful trader.
  4. The successful traders are on average very much more successful than the successful dentists.

That last point is the important one. If were to imagine that traders with a “success score” of less than 0 were to leave, and then sampled the remaining populations to estimate the relative success or failure of the individuals, the diagram would look slightly different.

Diagram of success of dentists and traders with unsuccessful traders removed from the sample
Now if we consider the results, we observe the following:

  1. The traders now have on average better outcomes than the dentists. The mean, median and maximum outcome are all better for traders, and sometimes to a great extent.
  2. The range of outcomes is slightly greater for the traders
  3. Trading now seems clearly to be the better occupation.

So what is the conclusion? That we should all become dentists? No. That we should all give up our dreams and settle for tedium and mediocrity? Definitely not. The conclusion is that one needs to be exceptionally careful when considering the track record of a sample of individuals, when the sample suffers from some form on intrinsic selection. In this case, the population of traders that we see (in movies and in real life) are usually the successful traders. Furthermore, in this example, the successful traders owe their success to luck alone. In fact, their luck overcame their lack of skill. This holds to a certain extent in the real world as well.

Fooled by the Black Swan

Is your organisation one black swan away from disaster? Are you taking hidden risks in the quest for success, and using hope as your only risk management tool?

Nassim Taleb’s books should be required reading for life

Nassim Taleb is one of my new favourite authors. I’m actually a little slow on the uptake here since I am currently reading his 2005 book “Fooled by Randomness: The Hidden Role of Chance in Life and in the Markets.â€? meanwhile the New York Times has his current book “The Black Swan: The Impact of the Highly Improbableâ€? on their best-seller list. I wholeheartedly recommend “Fooled by Randomnessâ€?, and fully expect that once I have read his current book I will be able to do the same for that.

Dumb luck is a large contributor to success in an uncertain world

Do you want the success of your organisation to be at the mercy of dumb luck?

Nassim Taleb’s writings resonate with me, because I agree with them. He just has an infinitely more entertaining (let alone more convincing) way of explaining his viewpoints. One of his major themes is how poor our understanding is of randomness. And he’s not talking about the “average joeâ€? in the street. If anything, he is more scathing of those so-called experts of the financial markets who are made or broken largely by luck. The unlucky fall be the wayside not be heard from, whereas the lucky shout their own praises from the rooftops.

I am not going to go further into his arguments in this post – the book is worthwhile reading if you are interested. However, I do want to touch a theme introduced in Fooled, and further expounded upon (I assume) in “The Black Swan�.

At some stage, the sun is going to stop rising

In the not so distant past, it was assumed that all swans must be white. Every swan ever seen had been white. All the classical statistical inference would have attributed a 100% probability to all swans being white. Until the rather unfortunate discovery of a little place called Australia. Enter the Black Swan.

Don’t trust past experience blindly, and trust your intuition even less

In risk management terms (and when I talk about risk management I include managing an organisation in the face of uncertainty, which includes every organisation I have ever known), events that may seem extremely unlikely based on past information and experience may still happen. If the occurrence of a black swan for your organisation would be catastrophic, are you really prepared to just hope that the past experience to date accurately reflects the future?

Actuaries and risk management

“Actuaries only look at the past so they are Fooled by Randomness.” This is a superficial description of actuarial work. Without a doubt, actuaries look to the past to infer certain parameters about the future. I’m not convinced this is necessarily bad as long as one realises that the past is not all there is to the future. The impact of HIV/AIDS and annuitant mortality improvements are typical of areas where actuaries have recognised that the past does not reflect the future and attempt to adjust for this in their calculations. Actuaries have the unfortunate job of trying to accurately estimate what this unknown future scenario will look like, rather than recognising that the risks exists and managing it.

When it comes to managing potentially catastrophic risks, Mr Taleb’s preferred practice is to limit all risks no matter how unlikely they may seem. The good news here is that if the consensus view is that the risks are extremely unlikely, the costs of mitigating those risks (transferring, hedging, reinsuring, selling etc.) should be relatively low. Mr Taleb prefers to find ways to use past patterns to make a profit, but use a sophisticated paranoia when managing the risks. He goes further and aims to benefit from the occasional black swan that flies his way. Again, more of this in his very worthwhile reading books.

Understanding all these potential risks, and understanding the potential for financial or operational impact on your organisation is not easy. Some of the results can be counter-intuitive, and simply drilling through the analytical steps to get to practical, useful steps requires a combination of common sense, uncommon insight into risk, and a tool-set capable of meeting the problems head-on.

In general, the human mind is a pretty poor tool for understanding a probabilistic world and making good decisions in the face of uncertainty.

What makes a good decision?

As an aside, another element of Mr Taleb’s thinking that I read with a fervently nodding head is that in an uncertain world, decisions should not be evaluated based on the outcome, but rather on whether it was the right decision given the information available at the time the decision was made. This is also not to say that the outcome never provides any information about the quality of the decision, just that it usually doesn’t. For example, take the decision to call “headsâ€? on the toss of a “fairâ€? coin. If the coin dutifully lands heads up, does that make the decision a good decision? I would argue very strongly that it does not. A more difficult example to agree with is that of a fund manager selecting a particular stock. If it the share appreciates in value over some period, is that sufficient evidence to show that the “buyâ€? call was a good one? Again, I would argue that it doesn’t. Especially when there is a large selection of fund managers making calls on all manner of stocks on a regular basis. Some of them have to be right some of the time. And a few of them will be right a great deal of the time just through luck.

Do you have a Black Swan?

Now if your organisation has a currency exposure, perhaps you are importing a component of your production process, or maybe your sales are partly to a foreign country, should you be bullish on the exchange rate? Should you be tring to time the market? Or should you be managing your risk by removing the areas of uncertainty over which you have no control, and where you are likely to be less informed than most professional currency traders, and where those self-same professional currency traders are playing in a massively uncertain world, where those with good “track recordsâ€? are more likely to be lucky than skillful? What happens if the Black Swan of a sharp exchange rate depreciation (or appreciation) is enough to wipe our your year’s operational earnings?

“We can’t afford risk management”

A common response to the argument for risk management is that hedging (or reinsurance, or put options or credit guarantees or business interruption insurance) is expensive. As I alluded to before, if the risk really is that unlikely, the cost should be relatively low. If the cost is high, it may reflect that others have a more prudent view of the possibilities of those risks than you do, which should start the alarm bells ringing immediately. The other side is that do you really believe than an appropriate way to build and manage your organisation is to continually take a small but very real risk of a catastrophic risk in order to make additional profit? If that is the primary source of profit for your organisation, then the fundamentals of that business may need to be revisited. Selling far out of the money naked call options as an income source may never get you into trouble and yield a modest revenue stream. Very few would agree that this is a good long-term strategy for success. A good number of the few that do have already been burnt in the process.

So what now for understanding and managing risk?

So there are four major points I would like to conclude with:

  1. If you operate an organisation, you operate in an uncertain world and are exposed to risks

  2. Just because you have never seen a Black Swan, doesn’t mean you will never see one.

  3. If there is a risk that could severely damage your business (a Black Swan), you had better have a better risk management strategy than closing your eyes and hoping

  4. Identifying these risks, measuring them and understanding their impact on your business, and then understanding the options available to you in managing those risks is an important and non-trivial exercise.