We should follow whatever seems probable, and avoid whatever seems improbable.
The most exciting discoveries are those that challenge our assumptions.