PHOENIX, Ariz. (TheStreet) -- Michael Lewis may have written "the" book on the power of numbers. But he's got no problem telling anyone who bothers to ask that, duh that power is not perfect.

"It's right there in there in the book: Billy Beane fired Jeremy Giambi because he pissed him off, not because of what the numbers told him," said the author of Moneyball: The Art of Winning an Unfair Game, about the manager and former outfielder of the Oakland Athletics.

I caught up with Lewis (more on that in a sec) as he rolls out what will probably be his next masterful deep dive, Flash Boys: A Wall Street Revolt, which shines a light into the dark-matter universe of high-frequency trading.

Lewis was kind enough take my questions on his 2003 classic account of how Beane, the quirky, numbers-obsessed Oakland A's manager, assembled a near-championship team using data and not just baseball common sense.

And I had to hand it to the man: Here was a legit literary heavyweight, with everything to lose and nothing to gain, having a highly technical conversation about the core thesis of Moneyball in an executive conference room full of top-level editors and writers as part the Society of American Business Editors and Writers annual conference. Lewis, it turned, out had no problem tiptoeing extemporaneously through the minefield of the theoretical limits of numbers, data and information.

"Anybody who wants to use data to not make a decision points to Moneyball," he said with some sadness. "And that is just not what the book does."

Moneyball never the answer?
I had first gotten wind of the limited case for information-based analysis in baseball -- and life itself -- about a year ago when meeting Randy Levine, president of the New York Yankees. With almost no prompting, Levine stated emphatically that even though Moneyball had become the brand for teams that use pure data to win, he knew for a fact that Billy Beane had not relied strictly on information when making baseball decisions for the near-championship run of the Oakland A's.

"Moneyball was a great movie," was the quote I jotted down. "But there was a lot of Hollywood in it. Baseball doesn't work that way." Levine declined interviews when I followed up to confirm. Since then I have been chasing after Lewis to comment on Levine's sense that numerical analysis in baseball is not the all-knowing, all-seeing force many hold it up to be.

Lewis not only confirmed Levine's skepticism about the be-all, end-all power of numbers, but in many ways was as articulate and measured a spokesman for the limits of data and information as I've ever spoken with.

"The problem is, people use numbers to model something they don't have," he said. "It's not runners in scoring position that matters. It's what's a kid going to do when you give him a million dollars."

Playing some "Skeptic-ball"
After my direct interview, Lewis sat for the full, on-stage discussion in front of a standing room-only business journalist audience. And here again, Lewis was open about how Beane's approach to using numbers to sift out value in baseball was a transformational idea -- 10 years ago. Since then, the use and misuse of database analysis has dramatically changed the effectiveness of applying data to solve problems.

"Mindlessly sifting through numbers, hoping to find some sort of truth, is never how it worked," he explained. "Billy Beane knew exactly what he had to be looking for to find untapped value in players. In fact, I was the only person at the time who was willing to have that conversation with him. That's why he spoke with me. I asked him those questions."

The real danger that lurks, Lewis said, is the deafening conversation about numerical analysis in sports and business. "The noise level about the data in decision-making in baseball is now louder outside the front office than it is inside it," he said. "The numbers geek has a voice, but everybody knows it's not a black-and-white thing."

And worse, Lewis sees a trend that more and more data are being used to protect those who wield it, not actually solve problems. "People think that just because there are numbers, there is this aura that all things can be quantified," he said.

"It's just not so."

Hiding from the universe
The morning after my Moneyball meet-up, I happened to bump into Diana B. Henriques, investigative reporter and author of Wizard of Lies: Bernie Madoff and the Death of Trust. It was she who interviewed Lewis on stage the day before and led him through many questions about the limits of information in business and sports.

I'll leave it to her to sum up what we've all got wrong.

"It's not that people are lying with numbers," she said. "It's that just they can't face the randomness of life. So they use the data to try to protect themselves from the universe, not embrace it.

"It is so utterly and terribly human."

This commentary comes from an independent investor or market observer as part of TheStreet guest contributor program. The views expressed are those of the author and do not necessarily represent the views of TheStreet or its management.