There’s no doubt that the majority of general aviation accidents can be attributed to some human element and are not simply an unexpected failure of a system component.
But there’s several reasons I would suggest that there’s still a greater risk.
Redundancy of systems and robustness of systems in an airliner more than offsets the risk of the systems complexity the airliner brings.
Turbine engines are more reliable than piston engines. A large jet engine can take $10 billion to develop over a 10-15 year period, and easily cost millions or tens of millions each, so jet engines aren’t exactly fighting fair with pistons, LOL.
Speaking of engines, there’s always more than one of them on an airliner.
Systems like anti-icing are quite common on an airliner and rare on 172-class airplanes.
An airliner is crewed, and I don’t just mean in the air. The dispatchers are part of the crew. The operations center is part of the crew. Systems specialists, weather specialists, etc. are all part of the system. At an airline, How many people are involved in making the wether go/no-go decision? It’s not one. How many people are involved in getting and verifying the right amount and quality of fuel got loaded onto the airplane? It’s not one.
Engine analysis in a modern jet is absurd. The airline, and the engine manufacturer, receive telemetric data in real time about the performance of the engine, so any temperature anomaly or vibration is reported.
An airliner is well equipped from an excess power and ceiling perspective to move quickly through or out of adverse conditions like icing or avoid weather. A 172 flies through an atmosphere that airliners only pass through momentarily.
Speaking of the airspace used, many airliners will spend almost their entire life in radar coverage flying on an IFR flight plan.
The systems automation in a typical airliner of today means that a flight control computer, an engine control computer, and augmented stability systems sit between the pilot’s input’s and the airplane itself. For instance, an airliner may have engineered into its flight controls the ability to manage a critical-engine-out throttle and rudder inputs, where a light twin pilot has to manage all that themselves.
We could keep going. With a whiteboard and a few minutes we could brainstorm our way to another 5, 10, or 20 factors that tend to make airliners safer than GA aircraft.
But to me, it really comes down to how every decision is managed by multiple people (some not even on the plane) through the lens of a rigorous, disciplined process, supported by robust systems.