WASHINGTON ― Several years ago, the Intelligence Advanced Research Projects Activity, an in-house think tank for the intelligence community, launched what is known as the “Good Judgment Project.” The idea is to test through forecasting competitions the factors that lead analysts to make good predictions.

One of the most interesting findings, according to a participant in the project, is that forecasting accuracy doesn’t necessarily improve when analysts have access to highly classified signals intelligence of the sort the National Security Agency has been collecting secretly from the phones and messages of world leaders and, it sometimes seems, nearly everyone else. In fact, the top forecasters, drawn from universities and elsewhere, performed about 30 percent better than the average for intelligence community analysts who could read intercepts and other secret data.

Good predictions don’t necessarily correlate with access to secret data, in other words. Indeed, according to Philip Tetlock, a University of Pennsylvania professor who heads the project, too much information can sometimes overwhelm analysts and decrease their forecasting accuracy. This problem of separating the essential “signal” from the chatter of “noise” has been well known to scholars for decades. But it seems to have been forgotten by modern-day intelligence agencies in their push to collect.

The NSA obviously operates on the theory that more data is better. And it has been ingenious at expanding the information available for analysts. If there’s a thrice-encrypted channel such as “The Onion Router,” NSA will crack it. If there’s a privacy scheme such as “secure sockets layer,” NSA will decrypt it. If there’s a world leader with an accessible cellphone, the NSA will tap it. But this mad dash for signals lacks the essential quality of sound judgment.

Reading the daily revelations of NSA electronic surveillance, some people have likened the situation to George Orwell’s all-seeing “Big Brother” in his novel “1984.” That image doesn’t match my own impression of the egghead military leaders of the NSA. Rather, I think of the “sorcerer’s apprentice” in Walt Disney’s movie “Fantasia,” who created a disaster by using tools and spells whose consequences he didn’t understand.

The NSA documents that have surfaced reveal an exuberant, almost adolescent quality among the tech wizards as they blow through privacy barriers. They give their top-secret projects colorful code names such as “Boundless Informant” or “Egotistical Giraffe.” They create compartments with mottos that sound like playground boasts: “Nothing but net” or “The mission never sleeps.” Hannah Arendt spoke famously of the banality of evil. This group makes one realize that childishness can be a characteristic, too. Like many hackers, NSA operatives seem to have done things sometimes for the thrill of it, because they could.

NSA leaders defend themselves by arguing that they need the largest possible sample of “metadata,” registering calls and messages, so they can later sift this haystack looking for the dangerous needles. The problem is that the NSA seems unwilling to recognize any limits on how big the haystack can be. The Foreign Intelligence Surveillance Court flatly warned in 2011 that some of the agency’s activities were illegally broad; the NSA supposedly changed its domestic procedures. But a report in Thursday’s Washington Post suggests that the agency used its foreign intelligence authority to collect similar information overseas. This should worry even people who support the NSA’s broad intelligence and counterterrorism missions.

The hemorrhage of secrets has come from leaks by former NSA contractor Edward Snowden. It’s hard to see him, even now, as any kind of hero: The way he chose to reveal programs that were legally authorized (albeit in some cases unwisely) has created severe problems for the United States and will cost tens of billions of dollars for U.S. companies that cooperated with court orders and NSA requests.

But it’s a fact that we are living now in a post-Snowden world, and that U.S. policies for intelligence collection will need to be different.

It’s hard to imagine global agreement on a framework for spying, which by definition involves breaking other countries’ laws. But America and the many, many other countries that conduct surveillance do need new rules of the road. Conventions against torture, chemical weapons and abuse of prisoners don’t prevent wars. But they do limit extreme activities of combatants. Something similar is needed here.

The worst outcome for the global economy, alas, now seems most likely ― which is that nations will try to ring-fence their data within national borders. That anti-globalization move won’t stop the spies but it will slow commerce and innovation, and make digital life harder for everyone.

By David Ignatius

David Ignatius’ email address is davidignatius@washpost.com. ― Ed.

(Washington Post Writers Group)