Bogus top 50 November 21, 2006
This is of historical interest only
With the abolition of the second ranking system that was the only ranking system in use at the time, this page is now merely of curiosity value.
A glitch or an abandoned experiment?
There were problems with vanishing votes in November 2006, during which time a dramatically different ranking table sometimes appeared. I never saw these bogus rankings for myself but I'm grateful to Prisrob for posting the top 50 on the customer reviews discussion board. Most people think that the disappearing votes were the cause of the strange rankings, but during that time my votes total was greater than those for Marc Ruby and Barron Laycock added together because they each lost many thousands of votes while I only lost about 500. In these bogus rankings, both of them improved while I slipped. I therefore cannot believe the vanishing votes caused the problem. I think that Amazon were experimenting with the ranking algorithm although I find it hard to believe that Cindy Penn could have been ahead of me on any calculation at that time. However, Sherlock Holmes once said Eliminate the impossible and whatever remains, however improbable, must be the truth;. It appears to me that Amazon were experimenting with extra higher thresholds but they might have been trying other things too. Of course, there may be another explanation altogether.
As a basis of comparison, I've included the rankings for the start of 2007. Ranking movements throughout 2006 showed relatively little change so the fact that there is a six week gap between the two dates used here is irrelevant. I have adjusted the rankings to remove two people who were removed from the rankings (one was eventually reinstated but I left the table as it was during their absence), so there are only 48 people listed.
0 comments:
Post a Comment