By Mike Holden

The nature of expertise

Before we embark headlong on our journey into the realm of science looking for ways to become better gamblers, it is first necessary to establish an objective understanding of the marketplace in which we operate and the people with whom we share that marketplace.

In both cases, the original source of inspiration for the material you will be reading over the next two weeks comes from a book called The Wisdom of Crowds by James Surowiecki.

Surowiecki is a business and finance journalist from New York. He has written for The New York Times and Wall Street Journal and his book is an essential read for anyone who takes their punting seriously, especially those who frequent betting exchanges.

Next week we will be looking at the accuracy of exchange markets and ways in which we can benefit from respecting the views of the crowd in general. This week we will be concentrating specifically on the mindset of individuals, in particular the so-called experts.

Surowiecki's book works on the sub-title of 'Why the Many are Smarter than the Few' and is primarily targeted at the world of business and management. It challenges the conventional wisdom that says we should refer to leaders and experts when making decisions, instead suggesting we should learn to trust the collective intelligence of larger groups as they invariably come up with the best possible answers.

The theory is launched on a discovery made by British scientist Francis Galton in 1906.

Galton, notorious for his work on statistics, attended a livestock fair in Plymouth where he chanced upon a competition for guessing the weight of an ox. Basically, the organisers were running a book and nearly 800 people tried their luck.

Afterwards, once the weight had been declared and the prize winners had been settled, Galton took home the tickets and ran a series of statistical studies on the answers that people had given.

Despite the many expert farmers and butchers who participated, it was no great surprise to Galton that nobody came particularly close to guessing the exact weight of 1,197 pounds. To his astonishment, however, he found that the mean average weight given by all the participants was essentially perfect: 1,198 pounds.

Apparently, what Galton stumbled across that day was the simple but powerful truth at the heart of Surowiecki's book: under the right circumstances, groups are remarkably intelligent, and they are often smarter than the smartest individuals within that group.

Surowiecki highlights many other examples that further illustrate the point. One such example is the relative success rate of answers given on the 'Phone a Friend' and 'Ask the Audience' lifelines on Who Wants to Be a Millionaire?

Every week, the show pits group intelligence against the perceived expertise of individual intelligence and, every week, group intelligence wins.

Research tracing back to the show's beginning in America found that the "experts" did reasonably well, offering the correct answer 65 per cent of the time. But they paled in comparison to the audiences. Those random crowds of people with nothing better to do on a weekday afternoon than sit in a TV studio picked the right answer 91 per cent of the time.

Of course, these results wouldn't stand up to scientific scrutiny. I mean, we don't know how smart the experts were, so we don't know how impressive outperforming them was. And since the experts and the audiences weren't asked the same questions, it's possible that the audiences were asked easier questions.

Nonetheless, it's hard to resist the thought that the success of the Millionaire audience is a modern day example of the same phenomenon Galton caught a glimpse of a century ago and Surowiecki certainly does a pretty good job of convincing us about the existence of this dazzling concept.

Needless to say, if his theory is true, the implications for punters, especially since the advent of betting exchanges, are hugely significant.

When Surowiecki elaborates further, his case for undermining the value of expertise is equally compelling.

In his words, the fundamental truth about the nature of expertise is that it is "spectacularly narrow". Hence, becoming an expert - acquiring a vast amount of knowledge in a specialist field - is one thing, using that same knowledge in broader concepts such as decision making, policy and strategy is something different entirely.

Take someone like Jeff Stelling, for instance. He might know an awful lot about lower league footballers and the average bloke down the pub would be put to shame if he ever dared challenge him to a quiz on the subject.

But Stelling's greater knowledge would only carry him so far in predicting the outcome of future lower league matches, not nearly enough to justify the time and effort that had gone into acquiring that immense wealth of knowledge down the years.

Therefore, it's safe to say his margin of victory in the quiz would far exceed any such margin of victory in a contest for predicting the results of forthcoming games.

If anyone doubts the possibility that the average bloke in the street could actually humble Stelling in terms of forecasting results, I should point out at this stage that the great man did himself participate in such a challenge for The League Paper the season before last.

Every week, a panel of six guests, including Stelling, had to predict the correct score of eight Football League fixtures taking place over the following seven days. However, at no stage during the course of the competition did Stelling escape from the bottom half of the table. Indeed, the readers, a different one of whom was invited to take part each weekend, led the competition virtually from start to finish.

In this respect, it's necessary for football punters to know the difference between being an expert on football and being an expert on football betting.

An expert on football doesn't really count for much in the world of football betting, but an expert on football betting does at least stand half a chance of finding an edge on the crowd. Although it is also important to realise that, even then, it might only ever be an edge, at best.

In order to better understand this distinction, it's perhaps necessary to call upon the findings of other academic sources in this specialist field.

J. Scott Armstrong is a Professor of Marketing at Wharton in Pennsylvania. He is widely regarded as the world's leading academic authority on forecasting and his work covers anything from the forecasts of government experts about the 'war on terror' to bond-fund managers on the Wilshire 5000 Index. He is equally sceptical of the level of value generally attached to expertise.

As far as Armstrong is concerned, experts are only slightly better at forecasting future events than laypeople. Indeed, he insists that, above a surprisingly low threshold, expertise and accuracy in most fields are entirely unrelated.

Then there is James Shanteau, one of the world's leading thinkers on the nature of expertise. He has spent a great deal of time coming up with a method for estimating just how expert someone is. Yet even he suggests that "experts' decisions are seriously flawed."

Shanteau recounts a series of studies that have found experts' judgements to be neither consistent with the judgements of other experts in the same field, nor consistent with previous judgements they have made themselves. For example, the level of agreement between experts in a host of fields he has studied is below 50 per cent, meaning that most experts are just as likely to disagree as agree.

Experts are also surprisingly bad at what scientists call "calibrating" their judgements. If your judgements are well calibrated, then you have a sense of how likely it is that your judgement is correct. But experts are much like normal people: they routinely overestimate the likelihood that they are right.

A survey on the question of over-confidence by economist Terrence Odean found that physicians, nurses, lawyers, engineers, entrepreneurs and investment bankers all believed they knew more than they did. Similarly, a recent study of foreign-exchange traders found that more than 70 per cent of the time, traders over-estimated the accuracy of their exchange-rate predictions.

In other words, it wasn't just that they were wrong; they also didn't have any idea of just how wrong they were.

And while in betting circles we can only kid ourselves so far in terms of how often we get things right before the bank balance offers a non-too-subtle reminder of the truth, we should recognise that this sense of self-misrepresentation seems to be the rule amongst nearly all experts in almost any field.

One general conclusion that these academics agree upon is that the best experts are those who know the full value of their expertise by calibrating their judgements, which effectively means keeping records as a source of constant reference for analysing self-performance and self-credibility.

Despite the old clichés about not being able to trust weathermen, a weatherman can be trusted more than experts in almost any other field because it rains on 30 per cent of the days when a weatherman has predicted a 30 per cent chance of rain.

So what are the lessons to be learned from these theories, if we choose to believe them?

Well, if you consider yourself to be an expert on the markets you are betting on but you don't regularly check your profit/loss figures, the chances are you're not nearly as shrewd as you think you are.

In such a scenario, it would probably pay to develop a greater understanding of just how often you are correct by keeping records of your all your betting activity.

The results might come as a shock to the system but they will, in the long run, provide you with a better understanding of your place within the overall scheme of things and help you to negotiate your way through future sticky patches.

Besides, a bit of humility never did anybody any harm. I mean, we've all got our own stories to tell about times when over-confidence has cost us an awful lot of money, haven't we?

If you know the average price at which you back or lay over the course of an entire football season and you know the strike rate you have achieved over the same period, it's much easier to make sense of fluctuating fortunes in the future.

I conducted a statistical review of my own Net Gains service over the summer and was quite astonished to find that my strike rate was only 32 per cent. Having achieved a profit margin in excess of 10 per cent, I thought it was reasonable of me to assume that my strike rate was much nearer 50 per cent.

However, it has now become a source of even greater comfort to know that I need only hit one winner in every three tips to stay on course for long-term success.

The greatest comfort in all of this, however, perhaps lies with those who like a bet butdon't consider themselves to be an expert.

It's all too easy to become intimidated in a world overcrowded with information and just as easy to grow disillusioned when things aren't going too well but the theories presented by Surowiecki suggest that the playing field is more level than many punters perhaps realise.