Not a bug  Poll calculations off


Well-known member
I started a poll with 9 options, you can select more than one answer. 6 answers have been picked one time and the percentage next to each one says 25%.
How many people have voted on the poll? It is theoretically possible to have 25% of users in the poll vote for 6 out of 9...

Lets say I have 9 options:
  1. A - 1 vote (user #1)
  2. B
  3. C - 1 vote (user #2)
  4. D
  5. E - 1 vote (user #3)
  6. F - 1 vote (user #3)
  7. G
  8. H - 1 vote (user #4)
  9. I - 1 vote (user #4)
As you can see, there are 6 votes... between four people. Now lets look at option #9... it has 1 out of 6 votes, which makes you think it should have 17%; correct? Actually, thats only partially true...

While #9 has 17% of the votes... it has also been voted in by 25% of users who took the poll... which if you ask me, is a FAR more important metric. It essentially gives more weight in the long run, based on the number of users who share the same sentiment. If you ran the percentages strictly based on vote counts, then it would give people who simply selected more options, more weight in the poll.

So instead of calculating the ratings as...

Number of votes for option / Number of votes in the poll.

It calculates as...

Number of votes for option / Number of participants in the poll.