A fatal incident involving Goliath at Six Flags Over Georgia on October 12 has raised an obvious concern for people who ride large roller coasters: does the risk rise with a ride’s height and speed? The evidence supports one comparison—roller coasters as a group produce more reported injuries per ride than other fixed-site attractions—but it does not establish whether 200-foot hypercoasters are more dangerous than smaller coasters. Nor does the public national data identify injury or fatality rates for rides built by individual manufacturers.
That information gap is especially uncomfortable after a death. The Cobb County Police Department confirmed that one person died and others were injured following an incident at the park; WSB-TV reported four additional injuries. Witnesses told reporters a person fell from Goliath, but as of late October 12 authorities had not publicly determined how the incident happened. It would be premature to claim that a restraint failed, a rider was improperly secured, or the coaster’s design caused the death.
What Goliath is—and what is being investigated
Six Flags describes Goliath as a 200-foot-tall hypercoaster that reaches 70 mph, with substantial drops, airtime and strong forces. It opened in 2006 and was built by Bolliger & Mabillard (B&M). Riders are secured with individual lap-bar restraints, according to the park’s published accessibility and rider guidelines.
The central unanswered issue is whether the occupied seat and its passenger-containment system functioned as designed. That requires examination of the train, restraint position and locking mechanism, any relevant sensors and warning indicators, operator procedures, maintenance records and available recordings. The fact that Goliath was stopped and passengers evacuated, or that witnesses describe a fall, does not by itself identify the failure mode.
Six Flags also suspended operation of Raging Bull and similar rides at other parks as a precaution. Raging Bull is another B&M hypercoaster, making those checks an appropriate precaution while the facts are uncertain. A precautionary closure should not be reported as a finding that the two rides share a defective part.
What the injury rates actually show
The most useful broad comparison comes from the National Safety Council’s 2024 North American fixed-site amusement ride survey, prepared for the industry association IAAPA. It estimates injuries requiring treatment beyond ordinary first aid, including the smaller subset classified as serious. The 2024 results show a higher overall reported injury rate on roller coasters than on other ride categories.
| Ride category | Estimated injuries per million rides, 2024 | Estimated serious injuries, 2024 |
|---|---|---|
| Roller coasters | 1.27 | 49 |
| Family/adult rides other than coasters | 0.59 | 62 |
| Children’s rides | 0.46 | 7 |
The roller-coaster rate was approximately 2.2 times the family/adult non-coaster rate. The survey’s serious classification covers deaths and injuries resulting in immediate hospitalization for more than 24 hours, excluding admission for observation alone. Dividing its estimated serious-injury counts by the corresponding estimated ridership gives roughly 0.10 serious injuries per million coaster rides and 0.05 per million other family/adult rides. Those last two rates are calculations from rounded published estimates, not separate official survey statistics or a demonstrated statistically significant difference.
The scale matters: even at 1.27 reported injuries per million rides, most roller-coaster trips do not lead to an injury requiring medical treatment. But that aggregate rate cannot answer the question now confronting Goliath riders. It mixes everything from relatively modest coasters to very tall, fast and forceful ones, and combines different causes and severities. A rate averaged across that population can conceal genuine variation between ride designs.
Fatalities are an even harder comparison. A published study of U.S. roller-coaster deaths from 1994 to 2004 identified 40 deaths associated with coasters, including 29 patrons. Of those patron deaths, 11 involved external injuries such as falls or collisions and 18 involved medical conditions that might have been triggered or worsened by riding. The study was not a current fatality-rate survey and did not provide ride counts for hypercoasters or manufacturers, so it cannot answer whether Goliath-type rides carry elevated fatality risk today.
Why nobody can give a credible B&M-versus-Intamin ranking from these figures
The survey has only three ride categories: children’s rides, family/adult rides and roller coasters. It publishes no breakdown by height, maximum speed, acceleration, restraint design, age of equipment, manufacturer or individual attraction. Its responses come from a sample of parks and are extrapolated to estimate North American totals. The 2024 edition received replies from 227 facilities, with 164 supplying ridership data; the reported sample accounted for approximately 65% of estimated ridership.
A manufacturer comparison would need at least two kinds of records for each model or ride: verified injuries and deaths classified by cause, and a denominator showing how many rides or passengers were exposed. Counting newspaper reports about B&M, Intamin or Gerstlauer coasters cannot substitute for the denominator. Manufacturers have different numbers of rides, designs, operating years and attendance levels, and news coverage is uneven.
With the public evidence available here, it would therefore be unsound to claim that B&M hypercoasters are unusually dangerous, unusually safe, or statistically safer than a competitor’s products. The lack of a usable comparison is itself a problem for consumers who are expected to assess the safety of rides marketed largely by height, speed and intensity. A regulator, operator or manufacturer may possess more detailed records, but the annual public survey does not equip families to make that comparison.
Did Goliath have earlier warning signs?
Goliath was associated with another fatality shortly after it opened. In July 2006, 45-year-old Michael Corry lost consciousness on the ride and died after being transported to a hospital. Contemporaneous Associated Press reporting said an autopsy identified a congenital heart problem and that the coaster was reopened after inspection found it operating properly.
Get Giving Up The Ship by email Disney and theme-park news, changes and practical advice without the fan-site spin.
That event deserves acknowledgment, but it does not establish a history of people falling from Goliath or of its lap bars failing. It was reported as an apparent medical emergency. Our review has not located a substantiated public pattern of earlier Goliath rider ejections or comparable restraint failures, although we have not obtained the park’s repair logs, employee reports or the state’s inspection files. An absence of searchable news reports is a narrow finding, not a clean bill of health.
A much more direct historical example of possible warnings before an ejection comes from a different coaster. After Rosy Esparza fell to her death from New Texas Giant at Six Flags Over Texas in 2013, police-file reporting described an employee’s concern that her restraint sat unusually high. Employees also recounted earlier problems with the train’s restraint-indicator system. The train was permitted to depart, and the ride later reopened with changes to its restraint arrangements.
New Texas Giant did not use Goliath’s B&M trains and restraints. Its history establishes no mechanical explanation for what happened in Georgia. It does show why investigators should compare the physical condition of a restraint with what a control panel or station check indicated, and should ask whether earlier reports pointed to the same concern.
A manufacturer’s history must be examined by failure type
B&M also built Fury 325 at Carowinds, which closed in 2023 after a visitor spotted a substantial crack in a support column. A North Carolina Department of Labor investigation attributed the failure to bending fatigue developing near a weld, according to WRAL’s reporting on the investigation. The damaged support was replaced and the coaster underwent extensive testing before reopening.
That episode demonstrates that a major engineering defect can escape scheduled checks on a large coaster. It does not demonstrate a shared restraint problem across B&M trains: a fractured support weld and a possible passenger-containment failure involve different components and different mechanisms. Grouping every serious incident involving one manufacturer into a single accident total would obscure those distinctions.
Six Flags Magic Mountain’s now-retired X2 raises yet another question. Our investigation of X2’s injury history found repeated allegations of traumatic head injuries across several years, as well as the company’s position that the attraction had passed its safety tests. X2 was an Arrow Dynamics fourth-dimension coaster with independently rotating seats, making it a poor design match for Goliath. The reason to mention X2 is its documentation of repeated allegations of injury, as the two rides do not share hardware.
How should a rider evaluate an extreme coaster?
Height, speed and maximum advertised G-force are incomplete measures of safety. What a passenger experiences depends on the shape of the track, the direction and duration of acceleration, changes in body position, restraint fit, operating conditions and medical susceptibility. A lap bar can be an appropriate restraint on a properly engineered and operated ride; an over-the-shoulder harness is not automatically safer simply because it looks more restrictive.
From a rider’s perspective, the most meaningful publicly accessible warning signs are a pattern of serious injuries of the same type, a documented equipment defect, a regulatory stop order, credible concerns raised before an incident, and evidence about what was repaired or redesigned afterward. A medical death on a ride and an ejection from a ride must be distinguished. So must an engineering failure detected and corrected before injury and repeated unresolved reports of a similar injury mechanism.
For guests who do not fit a restraint securely or cannot meet a ride’s stated physical requirements, the practical answer is to skip the ride, whatever its safety record. But the public should not be asked to make up for defects in engineering, inspections or operating procedures through personal caution. That burden belongs with the manufacturer, operator and regulators.
The disclosure question after Goliath
Georgia’s amusement-ride rules require an operator to take a ride out of service after a fatality or serious accident, notify regulators immediately and follow with a detailed written report. State regulation can establish whether a ride may resume operation; the public still needs sufficiently detailed findings to understand the issue and any corrective action. Because permanent amusement rides are excluded from the relevant federal consumer-product definition, there is no single nationwide Consumer Product Safety Commission investigation database for these coasters.
A useful investigation report would identify the cause of the incident, the occupied seat’s condition, whether alarms or previous maintenance records raised a concern, and the results of checks on similarly designed trains. It would also distinguish any deficiencies found on Goliath from those actually found elsewhere. Until then, the precautionary shutdown of comparable rides is reasonable, and declaring the entire category inherently unsafe—or completely safe—is unsupported.
There is a broader accountability question that this fatality should force into view: why can the industry publish a national injury rate down to hundredths per million rides, yet provide the public no comparable way to assess the risk of a 200-foot hypercoaster or the safety record of its specific model? A low average injury rate is reassuring about the industry as a whole. It cannot replace the ride-level evidence families need after something goes wrong.