On the Strange and Tragic Accident at Six Flags Over Georgia

By Charles C. W. Cooke

Tuesday, October 13, 2026

 

NBC reports:

 

A person died, and several others were injured at Six Flags Over Georgia on Monday, officials said, after witnesses reported seeing a man fall from a roller coaster.

 

The Cobb County Department of Emergency Communications confirmed the death but did not provide additional details about the incident at about 4 p.m. at the amusement park just outside Atlanta.

 

Naturally, I don’t know what happened here. But I do know a lot about roller coasters and how they work, so I thought it might be useful to put this incident into context for those who don’t.

 

The context? That this is absolutely unheard of. One is nine times more likely to be killed on a commercial airplane than on a roller coaster. Especially that roller coaster. The ride in question, Goliath, was designed by a Swiss company called Bolliger and Mabillard (B&M), which, since 1990, has been the gold standard within the amusement industry. In its history, B&M has constructed 136 roller coasters, 133 of which are still operating, and 18 of which are of the same type as Goliath. Despite those roller coasters having given hundreds of millions — perhaps billions — of rides, nobody has ever died on one of them as the result of a malfunction.

 

The lap bar system on B&M’s hyper-coasters (that’s Goliath’s type) is generally considered to be the best in the business. The design is simple. The molded restraint is shaped to fit over the thighs and against the abdomen — which means that, in order to escape, a normally sized rider would have to squeeze both his hips and his thighs past the clamshell. Ceteris paribus, that is not possible, as this image from the patent filing shows:

 

 

The restraint’s locking system uses a mechanical ratchet at the pivot point. This allows the clamshell to move inward (often, during the ride, the restraint will click down further) but prevents it from opening again. When the bar is pressed down toward the rider’s thighs, its multiple redundant locking catches (“pawls”) engage a toothed wheel that must be released before the bar can rise. Under normal operating procedure, the restraint must be down at least three clicks before the ride’s computer system will allow the train to leave the station. After the train is outside of the station, there is no way to release a restraint without specialized equipment that is available only during an evacuation or scheduled maintenance. The best available estimate is that each seat is able to restrain 35 kilonewtons — about 7,900 pounds-force applied at the rider-contact point — before breaking. That is roughly equivalent to the combined weight of two Ford Escapes hanging from each restraint.

 

Today, Twitter is full of people insisting that the same thing as happened on Goliath “nearly” happened to them on a roller coaster in recent years. It almost certainly did not. I have no doubt that most of those people truly believe that they were in danger, but, in reality, they weren’t. Modern roller coasters are designed to make the riders feel as if they are going to fall out, and most are highly effective in achieving this sensation. Still, it remains just that: a sensation. Statistically, one has a far greater chance of being killed at an amusement park by a vending machine than by a roller coaster. The tolerances involved here are just extraordinary.

 

That said . . . this incident does look different from any other I’ve encountered. As far as I can see, the restraint that supposedly failed did seem to be up at the time of the accident — or at the very least, it seemed to be less far down than it should have been — and, unusually, it seemed to be the only restraint in the row that was in that condition. When one adds in that a) the deceased apparently fell out toward the end of the ride, rather than the beginning (this matters, because, if the restraint had been up from the start, the rider would almost certainly have been ejected earlier) and b) that one of the other passengers suggested that, before he fell from the train, he’d lunged forward to try to catch his phone, I’ll confess to being baffled.

 

It seems to me that there are four options here:

 

1.      That this was a catastrophic failure of B&M’s design, akin in nature to a plane crash or a person’s being struck by lightning. The chance of this happening is at least one in two billion, but, somehow, this was the one. This failure could have been in the computer system, the sensors, or the restraint mechanism itself. Knowing what I know, I find it extremely hard to believe that this is what happened, but it has to be considered given the available details.

 

2.      That Six Flags Over Georgia is responsible for a catastrophic maintenance problem that left one or more parts of the safety systems fatally unreliable.

 

3.      That there was no technical problem, and that Six Flags Over Georgia is responsible for misusing a working ride. This misuse could have involved the train being deliberately dispatched while in an unsafe condition (the safety rules can in theory be overridden, but that process is too complicated to be enabled by mistake), or it could have involved a rider who was too large to ride being allowed to ride nevertheless (conceivably, a rider with an unusually large abdomen could stop the bar before it was lowered securely over the thighs, leading to a situation in which the ratchet was technically locked — and remained locked — but there was sufficient clearance for the rider’s hips and thighs to move past it during negative-g portions of the ride).

 

4.      That this was a perfect storm, involving negligent maintenance, plus indifferent ride operations, plus unusual behavior from the rider, and that, together, this led to an impossibly unlikely accident.

 

We will see. This is all very strange. Until now, it has always been the case that people who died on roller coasters fell into one of a handful of camps. Those are 1) that they should not have been riding in the first instance (because, for example, they had no legs, or had a terrible spinal injury); 2) that they walked onto the tracks and were hit by a train; 3) that they had a preexisting condition (e.g. an aneurysm or a weak heart) that they did not know about; 4) that they just happened to die on a roller coaster, but were not killed by it; or 5) that they stood up or jumped out or convinced the operator not to secure them. This one is different. I’d be shocked if, after 36 years of perfection, B&M’s systems were suddenly called into question. But you never know. A full inquiry will be forthcoming.