Blind ranking game
Ten slots, one card at a time, no take-backs. Pick a list and tap a slot.
Pizza toppings
1 of 10
This list was written by a visitor, not by us.
1 is the topping you always order
Red onion
Your friend's board
How this version plays
Every list here holds 20 things and the board has 10 slots. A round deals you 10 of the 20 in a random order, so you cannot count on your favourite turning up, and you cannot hold slot 1 for it. With 10 things for 10 slots you would always know what is still to come. With 20, you do not.
- Slot 1 is the best and slot 10 is the worst. The card on top says what "best" means for that list.
- Tap a slot to lock the card in. On a keyboard, press 1 to 9, or 0 for slot 10.
- You cannot move a card until the board is full.
Your blind score
When the board is full you can see all ten at once, and the arrows let you put them in the order you would choose now. The blind score is how much of that order your blind board already had.
45pairs in a board of ten. The blind score is the share of them your blind board had the right way round.
It counts pairs. Ten things make 45 pairs, and for each pair the question is whether your blind board had the two the same way round as your final order. Change nothing and the score is 100%. Swap one neighbouring pair and one pair of 45 has flipped, so the score is 97.8%. A board in exactly the reverse order scores 0%. Statisticians know this count as Kendall's rank distance, after Maurice Kendall's 1938 paper in Biometrika; the score here is the share of pairs that agree.
Want to beat it? There is a worked-out rule for where each card should go, and a calculator that shows it.
The lists
There are 20 lists and 400 things in all, written for this site: food, chores, weather, hobbies, smells, sounds and other things everybody has an opinion on. They are everyday things on purpose. For anything else, from your friends' cooking to the films you watched this year, make your own list and send the link.
Source for the score: M. G. Kendall, "A New Measure of Rank Correlation", Biometrika 30 (1 to 2), 1938, pages 81 to 93.