How Our Rankings Work

Ryan Maddock

Updated June 4, 2019150.2k views

Why We're Here

Having rankings with integrity is critical. Rankings are what we're all about. They're our reason for being, our raison d'être —and hence, they're muy importante! We're well aware that opinions are subjective, and that when it comes to questions about which movies, TV shows, fast food chains, or sneaker lines are the best, there's no definitive right or wrong answer. But to us, that shouldn't have to mean that you should have to rely to one person's subjective opinion or wade through a sea of "hot takes" for recommendations. That's why we're here: the single biggest reason Clark Benson started Ranker was to give people an alternative to all the ranked lists on the internet that are based on the thoughts of one opinionated expert or random blogger.

How It Works: The Basics

Ranker lists are built by lots of people who have experience with what they’re ranking — in other words, they're based on the “wisdom of the crowd.” Anyone can vote either up or down on any and all items on the list. Many of our lists are also re-rankable, which means that people can make their own ranking of all the items on the list, a ranking that can't be affected or changed by other users. Since the people who are doing their own re-rankings are taking the time and effort to login and make their own version of the list, the Ranker Algorithm weights the results of re-rankings higher than it does other votes.

There are four primary factors that affect an item's ranking on a list: the number of up-votes, the ratio of up-votes to down-votes, how often it's ranked, and where it's ranked. When you look at a list and feel that an item’s ranking doesn’t seem to make sense based on the vote counts, it could be because of the impact that re-ranks have on rankings. Yeah, it’s complicated, but trust us: we’ve done a lot of testing and data-science wonkery over the years to tweak how we weight results, and we're confident that a list on Ranker is about as accurate a picture of voter behavior as anyone can get. What's more, because the majority of people who vote on Ranker lists come from search engines, our rankings have “representative sample” integrity.

How We Keep Our Rankings Fair

Since Ranker is a pretty large site now, (thanks, Ranker visitors - rest assured that we are always trying to make Ranker better), occasionally there are problems with people trying to “game the rankings.” For example, the fanbase of an up-and-coming singer may organize a push to make sure that singer becomes

1

. Trust us, we understand fandom - everyone at Ranker is a fan at heart, each with their own interests - but if we allowed our rankings to be gamed, it would defeat the entire purpose of Ranker.

If either a human or machine at Ranker detects “biased voting,” sooner or later, visibly or invisibly, the impact of those votes will be corrected. All this being said, while we have plenty of processes in place ready to address this issue, any crowd-voted system at our scale will never be airtight, and bias will occasionally slip in. That's why we're always working to develop new controls on biased voting.

While we recognize that every individual who votes has their own bias, our goal is to fairly sample a wide range of different individuals with different biases to create a true portrait of the "crowd's wisdom" on any given subject, from film, to television, to sports, to practically everything else. But it takes more than just our data scientists and engineers to keep voting fair: it takes millions of voters like you sharing your subjective opinions to create as big and as representative a sample as possible of the public at large. So keep voting! Between your rankings and our lists, we're creating an incredible resource for people across the world who want rankings that come from hundreds of thousands of real people like them — not from one opinionated blogger.