Benford's law has been used to identify fake accounts on twitter by collecting the number of followers of suspicious accounts and then collecting the number of followers each of those followers has and running that as a data-set against Benford's law.
I think something like that would be very useful for HIVE.
And as far as timing votes for profit, I believe all votes should be equally weighted no matter who votes or what time they vote.
The "band-wagon" votes are toxic (gamified for profit only).
AND, when I first joined, there was a view counter. The consensus on why it was taken away was due to complaints POB was dead and posts with hundred(s) of votes only had like 3 or 4 views. From bidbots to autovoters it was obvious most of the highest paid posts received very few actual eyes on them. LINK
Fake accounts are not as big an issue for Hive as they need some HP behind them to make any difference. Strategic voting is toxic if it does not take account of actual content.
The views number was not accurate enough to be useful. I tend to judge posts by the real interaction they get. Anyone can get hundreds of tiny votes, including me.
Perhaps,
I don't know if we can reliably measure views across the various Hive dapps. In the old days there was just Steemit.
I'll judge the interest in a post by the comments it gets rather than the votes.
Even so, some view-time statistics are better than no data.
I'd still really like to see page views over 20 seconds.