Forget satellite images and aerial searches—the best way to find Malaysia Airlines Flight MH370 may be with mathematical techniques dating back to the 18th century, the BBC reports. That's how Air France flight 447 was found in 2009, using "Bayesian statistics" to measure the probability of the plane being in one place or another. Named after Presbyterian minister and mathematician Thomas Bayes, the technique allowed experts to apply several factors to each point on a map: For example, what was the chance it crashed from mechanical failure? How far do planes tend to crash from their last known location? What was the chance that search teams missed debris in various locations?
It's like picking a restaurant by balancing how full it is, what your favorite restaurant-review website says, and so on—except that experts hunting for Airbus A330 did that for each point where plane may have crashed in the Atlantic, Five Thirty-Eight notes. It was so hard that the US team of statisticians invited by France gave up, until they de-emphasized one statistic: that a plane's black box emits a signal after a crash 90% of the time. They changed their findings, and presto, the plane was found. Bayesian techniques have helped people find World War II U-boats, men overboard, and sunken treasure, but there's no evidence that Malaysia is employing them now. "I suspect that they just guess, like professional baseball managers used to do before Moneyball," says a biostatistician.