It is probably mostly a reporting bias. Almost all new mathematical discoveries don't have simple pictures to draw and are about highly abstract concepts which

It is probably mostly a reporting bias. Almost all new mathematical discoveries don't have simple pictures to draw and are about highly abstract concepts which are difficult to write about for a general audience. Analysis is an extremely active field, PDEs have almost endless amounts of open research questions. Usually they are not very flashy and can be very non-geometrical. Algebra is extremely hard to write about for a general audience. Trying to communicate any result which does not have a simple visual interpretation seems like a nightmare. Numerics are often things where advances are hard to relay to an audience, as it usually is about incremental improvements instead of breakthroughs.