About Us

The case for rankings

Have you ever though about how seldom we know which thing is the best? Outside the world of sports best is a fuzzy concept, but we should't we have some idea? Let's take restaurants as an example.

Current ways of evaluating

We have some great ways of evaluating restaurants, e.g reviews, ratings, recommendations and expert opinion pieces. But they all have their limitations.

Reviews and Ratings: The problem is that only the most motivated - typically pleased or displeased - tend to review. E.g. Google maps star ratings are actually measuring the ratio between guest who are extremely satisfied and those who are dissatisfied, good restaurants will all more than 4 stars but the difference between a 4.1 and a 4.5 is not meaningful.

Recommendations: From a close friend they can be spot on, but the small sample size (sometimes 1!) is problematic. If your friends loves a restaurant is not a very strong argument for the restaurant being the best in town.

Expert opinions: Are often insightful for a small selection of restaurants, but they are reliant on a single human and hard to scale. Also, their tastes might not align with the general public.

Above all, humans are not good at doing absolute evaluations. Rating a restaurant from 1 to 10 is hard, but comparing two options directly? Much easier.

A new way of evaluating

At Rank Tank we have a new way of evaluating things. We ask the user about restaurants they know and present two options and simply ask which one they prefer. The voter does not have to struggle to come up with a score, importantly, they also evaluate the two options at the same time and using the same mental yardstick.

In our rankings all restaurants compete and are compared relatively, this means that we can ask users to quickly provide a little bit of information and use the information from many to piece together an accurate and comprehensive ranking.

To create the comprehensive rankings we once again look to sports and the world of chess. In chess, the best player is not the one who wins the most games, but the one who beats the best players. We use the same principle and you can read more about it here: Elo rating system