What the current sources actually show
What the current sources actually show gives the /guides/leveling reader a concrete decision without hiding the evidence boundary.
One reviewed speedrun video records an 18 minute 10 second run to Trial 10 after repeated attempts, strong item luck and a plan centered on speed effects and instant infection destruction. A recent Reddit discussion argues that ending around Trial 20 can produce better experience per hour than spending several hours on much deeper runs; commenters share examples, but no official formula or controlled dataset is provided. These sources support two ideas: minimize non-objective delay for short targets, and question whether deeper always means more efficient. They do not support a universal XP-per-minute calculator. Run time, team quality, maps, disasters, item drops and deaths change the outcome.