Hotel rankings are ads. "Top 10 hotels in Barcelona" is usually a list of whoever paid the most, and from the outside you cannot tell which is which. So I built one where the rule is printed on the box. One formula, applied identically to all 245,901 hotels, published in full: a guest rating gets pulled toward its country's mean in proportion to how little evidence stands behind it, then re-based 70/30 against that country. A 10/10 from twelve guests does not outrank a 9.1 from four thousand, and an 8.6 in Japan is not the same sentence as an 8.6 in Egypt. Nothing is submitted by hotels and nothing is for sale. The commission is identical whichever hotel you pick, so there is deliberately no mechanism by which a property could pay its way up. Two things I would genuinely like feedback on: 1. If you know a city well, open its list and tell me what it got backwards. That is the only test that matters to me. 2. It runs entirely on Cloudflare's free tier, no server, and the read path never touches a database. Happy to go into how if that is interesting. Thanks for looking.

Forum Threads

p/hotelroost

•

2mo ago

Should a 10/10 from 12 reviews ever beat a 9.1 from 4,000?

We launch tomorrow, and this is the decision I keep second-guessing.

HotelRoost scores 245,901 hotels with a single formula. The hard part was never the maths it was deciding how much to trust a small sample. A brand-new guesthouse with twelve perfect reviews genuinely might be excellent. Or it might be twelve friends.

View all

HotelRoost

Hotel rankings with the formula printed on the box

Hotel rankings with the formula printed on the box

Forum Threads

Should a 10/10 from 12 reviews ever beat a 9.1 from 4,000?

Forum Threads

Should a 10/10 from 12 reviews ever beat a 9.1 from 4,000?