> We’re going to go out on a limb here and guesstimate that the remaining “maybe” answers can, given enough time and effort, be turned into “yes” or “no” answer

> We’re going to go out on a limb here and guesstimate that the remaining “maybe” answers can, given enough time and effort, be turned into “yes” or “no” answers, and that these will be distributed similarly to the 20:17 ratio of the fully answered headlines. The OP does admit to going out on a limb, but I don't think this is a justified assumption. Without having seen the headlines in question, I would guess they are in the maybe category because they are dubious at best.