I do concur that any fine grain tuning of the option to represent a hint as neither an attributable certain fact nor a non-attributable fact should not be 'clucky'. Quick, easy, defaulted, and simple.
The solution to cover this middle ground needs to support directing targeted research now and the future in to the right hints to maximise research time to achieve best value.
A simplistic option adds little value unless it is coupled with something else that can be used to filter and sort on. For example (1) a guess weighting/percentage as to how close the hint is, or (2) coupled to the need for a specific corroborating fact that would allow the hint to be used or not, or (3) both.
With 90k people in one tree and 13k hints to be reviewed still on it and now 20 years down the line, with Ancestry still adding huge data sets... I have a feeling that I could wading through tens of thousands of hints for years to come. And over another 20 years and time spent considering repeated hints will mount up!
Guess well just have to see if Ancestry get the hint...
....pardon the pun... I couldn't resist. ;o)
Many thanks.
James