This is just an idea, not a set plan...
Rank is calculated based on number of reviews (which IS already a factor) and quality ratings only. There is a 'panel of experts' who all vote for quality of each review- say 5 people. My vote is one, and four other people volunteer to do this. That is to ensure that each review gets an equal number of votes for calculation.
Other 'usefulness' voting may count, to a smaller degree, or maybe we leave it out entirely to prevent the highly variable attention different toys and/or reviews get from affecting rank.
In addition, rank is calculated independently for each contributor, on the same scale as now (0-10), not on a curve. Therefore, your rank doesn't change except when you submit new/follow up reviews and they are voted on.
Thoughts? Clearly, this is not a perfect system (there's no such thing), but does it sound like one that will do a better job of representing the quality and experience of each reviewer without placing too much importance on extraneous factors, such as popularity?
Rank is calculated based on number of reviews (which IS already a factor) and quality ratings only. There is a 'panel of experts' who all vote for quality of each review- say 5 people. My vote is one, and four other people volunteer to do this. That is to ensure that each review gets an equal number of votes for calculation.
Other 'usefulness' voting may count, to a smaller degree, or maybe we leave it out entirely to prevent the highly variable attention different toys and/or reviews get from affecting rank.
In addition, rank is calculated independently for each contributor, on the same scale as now (0-10), not on a curve. Therefore, your rank doesn't change except when you submit new/follow up reviews and they are voted on.
Thoughts? Clearly, this is not a perfect system (there's no such thing), but does it sound like one that will do a better job of representing the quality and experience of each reviewer without placing too much importance on extraneous factors, such as popularity?