With the explosion of computing power, numbers and statistics dominate our lives like never before. Today, we can come up with a statistic to support just about any argument with a click of a mouse. The question is whether we make the mistake of using statistics like a drunkard uses a lamp-post -- for mere support, rather than for illumination.

Consider the case of today's financial markets. With the economy perched tenuously on the cusp of "green shoots" economic recovery and financial Armageddon, both bulls and bears can marshal endless statistics to support their positions. And not a day goes by when you don't read about a new stock market indicator confirming that "every time the market has done X, it has then done Y." Some old hands have a more low-tech approach. Former Fed Chairman Alan Greenspan looks at sales of men's underpants, reckoning that men are less likely to throw out old underwear when the economy is bad.

Despite the obvious shortcomings of statistics, arguing against an impressive array of complex data is somewhat akin to blasphemy. You just don't do it in polite company. Just as your lawyer has little interest in revealing that his legal gobbledygook masks a stunningly boring profession, economists, bureaucrats and financial analysts have little incentive to wean you from your belief in the "Holy Grail" of statistics. As George Bernard Shaw wisely observed, "all professions are conspiracies against the laity."

"There are three kinds of lies: lies, damned lies, and statistics."

-- Mark Twain

"Lies..."

Cite an economic statistic from an authoritative source, and it becomes as credible as "1+1= 2." Ironically, the only thing you can say for sure about that statistic is that it is wrong -- and that it will be revised within a month. Here in the United Kingdom, economists had believed the country's growth in 2005 had slumped to 1.5% or less. Today, that estimate has been revised to 2.2%. That 0.7% difference may not sound like much. In fact, it means that the original estimate was off an eye-popping 46%. Yet, government policy decisions are made on the basis of similarly calculated statistics today. Now, think about how similar (mis)measurements relating to today's U.S. economy are (mis)shaping economic policy responses at both the Fed and the Obama White House.

Financial analysts like to think that they are above the fray. After all, they ignore economists' statistics and prefer to "kick the tires" of a company before they invest. Like the iconic Benjamin Graham, they focus "on the numbers" with a relentless rigor. That sounds impressive -- that is, until you are an investment banker, and you actually see how those numbers are put together just for the benefit of these financial Jesuits. The numbers aren't "lies" per se. But as in any good marketing campaign, the numbers are presented in the best possible light. Moses didn't descend from Mount Sinai with company balance sheets and income statements taken directly from the hand of God. Sadly, many financial analysts act as if they did.

...Damned Lies...

Author Rex Stout noted that "There are two kinds of statistics: the kind you look up and the kind you make up." What happens when "1+1= 3" and things just don't add up? China's reported first-quarter 6.1% expansion in GDP doesn't quite tally with collapsing demand for electricity and a 3.5% drop in oil demand. Communist countries like China, the Soviet Union and East Germany developed a strong tradition of fudging the numbers. The only ones they seem to have fooled were U.S. academics who long argued that the economic achievements of East and West Germany were comparable. Those arguments, of course, collapsed with the Berlin Wall in 1989. But let's not pick on Communist apparatchiks and American academic apologists. Today, there is a whole cottage industry of "shadow statistic" analysts who recalculate U.S. government statistics to their "true value." Needless to say, they argue that we are a lot worse off than official statistics reveal.

...and Statistics...

Lost in an array of complex looking equations, it's easy to forget just how flimsy expertise based on statistics is. When I was in third grade, I collected baseball cards. I knew all the relevant statistics of every baseball player in the major leagues. But think of how a major leaguer would have reacted to my "expertise." Though we were both "experts," my expertise was a far cry from going out there and swinging a bat. I can't help thinking senior executives running companies may feel the same way about 26-year-old Wall Street analysts who write reports on their companies.

And statistical conclusions are only as strong as the models from which they are derived. Statistics are terrific for actuarial tables for insurance companies. Take 300 million people in the United States and you have a pretty good idea how many will be born and how many will die in any one year. Although the Grim Reaper may surprise you when he arrives, on the whole he is the most predictable of characters. The same cannot be said for far more complex financial models that claimed to measure the financial risks of everything from Fannie Mae to CDOs. Garbage in; Garbage out.

"Statistics are no substitute for judgment"

-- Henry Clay

Statistics is a story of a discipline misapplied. The economist Frank Knight made the distinction between "risk" and "uncertainty." Statistics are useful for calculating "risks" -- say, a game with unknown outcomes but known ex-ante probability distributions. It's the reason the House always wins in the end in Vegas. But statistics is not useful for "uncertainty" -- calculating outcomes in situations with an infinite number of unpredictable outcomes. In other words, don't use statistics for any part of life outside the limited realm of Vegas gambling halls. And that includes financial markets and the economy. In Knight's terminology, Fannie Mae, Freddie Mac, Lehman Brothers, AIG all thought they were operating in the world of "risk." Instead, they were operating in a world of "uncertainty."

The lesson? Take all statistics you read with a grain of salt.

That's why statistics is like a bikini: "What they reveal is suggestive, but what they conceal is vital."

Sign up now to receive The Global Guru e-Newsletter FREE! Email Address: Yes! Please send me Nicholas Vardy's FREE weekly newsletter, The Global Guru.