Cold comfort perhaps, but the hard statistics on air travel are reassuring
Trans World Airlines Flight 800 crashed off Long Island in 1996. Swissair Flight 111 crashed off Nova Scotia last year. John F Kennedy Jr's plane crashed off Martha's Vineyard last summer.

And now EgyptAir Flight 990 crashes off Nantucket. Four recent flights ending in tragedy off the US eastern seaboard.

Confronted with these plane crashes and the non-stop media coverage and speculation that accompany them, one might be excused for momentarily entertaining some strange ideas about a link between them. Indeed, some news reports have even implied that the Atlantic seaboard is a new Bermuda Triangle.

After a series of terrible accidents, it's human nature to try to discern a pattern among - and therefore an explanation for - disturbing events. Of course, we don't yet know what caused the crash of EgyptAir Flight 990. But it's important to understand that this recent spate of crashes is more than likely to be a series of completely random events.

Humans have an innate tendency to attribute significance to anomalies and coincidences. An event with a one-in-a-million chance of happening to any American on any given day will, in fact, occur 260 times each day in the US.

Thus in stories like this crash, where there are so many details regarding times, dates, locations, manufacturing specifications, grieving relatives, political intrigues and other items of possible relevance, there will undoubtedly be some seemingly unlikely links. When faced with such a torrent of information, our ability to separate the possible from the probable tends to deteriorate, and this leads us to speculate about cause and effect based on the flimsiest of connections.

A related foible is our often unconscious belief that if we can vividly imagine something, this makes it more likely.

The EgyptAir Boeing 767 rolled off the production line immediately before another Boeing 767 jetliner that crashed eight years ago in Thailand. It's easy for us to envisage tired workers suffering a momentary lapse of attention. But this does not amount to proof. Then there is the fact that the only person to get off the EgyptAir flight during its layover in New York was a "grief counsellor". Is this some grand paradox, or just one of countless possible oddities whose complete absence would be the oddest coincidence of all?

A useful exercise would be to examine - in as focused and relentless a way as the details of the EgyptAir flight are being examined - any ordinary flight from New York to, say, Madrid that did not end in a horrible crash. Undoubtedly, we'd find approximately as many disquieting stories about the plane's history, approximately as many last-minute changes in the passenger list and approximately as many of the million and seven other things that investigators and reporters are presently considering.

The statistics that really matter on the safety of air travel are in fact reassuring. As Arnold Barnett of MIT has noted, we have one chance in seven million of dying in any given domestic jet flight. That is, a passenger who daily and randomly takes a jet flight between American cities would, on average, go 19,000 years before dying in a crash. For international flights on American-owned airlines the chances are one in 1.5 million flights. EgyptAir and Swissair are also safe airlines.

But stories and scenarios are more seductive than statistics and induce in us a suspension of disbelief and critical thinking. What happened to EgyptAir Flight 990 is still a mystery, but it's more likely to be something ordinary than not. This doesn't lessen anybody's grief, but it should allay some fears.

John Allen Paulos, a mathematics professor at Temple University, is the author of `Once Upon a Number'. New York Times