There are a lot of decisions that affect the risk level. This is the same with any activity. Cutting edge climbs generally involve enough "risky" decisions to

There are a lot of decisions that affect the risk level. This is the same with any activity. Cutting edge climbs generally involve enough "risky" decisions to be considered dangerous, I have lost too many friends to claim otherwise. But there is a lot of "serious looking" climbing, including quite a lot at high altitude as well as very technical routes that can be done at a risk level similar to non-"extreme" activities. Accidents on hard routes are either because of objective hazard (e.g. climbing underneath a serac, an obvious hazard that the climbers chose to accept because they thought the route was "worth it") or human error (e.g. rappelling accident). You choose where to climb and you assess conditions. Some hazards are hard to assess (avalanches from slopes that you cannot evaluate), but most accidents related to objective hazard involve a specific decision that was known by the participants to be risky. It is not the case that choosing to attempt a big objective involves an especially large risk. But if you never make risky decisions, then you are likely to fail at your objective (maybe climb something less impressive). The trick the prolific climbers who make it to old age have mastered is evaluating risk sufficiently accurately to get the big lines without dying. There are a couple who were just lucky, but by and large, luck runs out eventually so you have to be good at assessing risk.