About Typical
Most word games ask you to find the one right answer. Typical asks the opposite: not what you would say, but what everyone would say. It is a game about the shared obvious, and about how far from it you happen to stand.
Where the idea comes from
Games like Contexto hide one word and rank every guess by how close it comes. Typical turns that around. There is no hidden word. Every answer on the list pays something, and the closer to the centre of what people think, the more it pays. The question is never "can you find it" but "how typical is what you found". We built it because the second question turned out to be the more interesting one at the dinner table.
How the lists are made
Every question has a list of between about 160 and 260 answers, ranked from the most typical to the rarest. The rankings are authored: an editor orders the answers by how typical they are, question by question, and the lists are checked against each other so that a Something made of metal and A kitchen tool do not argue about where a spoon goes. They are a judgement about typicality, not survey data, and we say so here so that nobody mistakes rank 1 for a measurement.
The player numbers, on the other hand, are real. "More typical than 72% of 1,240 players" and "1,141 had it" count the people who actually finished that question. Nothing about any one player is stored to make those counts; the privacy policy explains how.
Who makes it
Typical is made by Finais Consulting UG (haftungsbeschränkt) in Frankfurt am Main, Germany, a small company that also makes Qanat, a daily puzzle about water. The full company details are in the legal notice. The quickest way to reach the people behind the game is the feedback form; for anything else there is the contact page.
Questions people ask
Is it free?
Yes. Today's two questions and the weekly challenge are free, and there is no account to make. The site is paid for by advertising, which runs only if you agree to it in the dialogue shown on your first visit.
When do new questions arrive?
At midnight UTC, every day. That is 1 am in London in winter, 2 am in Berlin, 7 pm the evening before in New York. The weekly challenge changes on Monday and stays open until Sunday.
Why did my answer say "not found"?
Because it is not one of the answers on that question's list. That costs you nothing: only answers count as guesses. Lists are long, but they are not endless, and a very specific word (a Burmese cat for An animal) is usually covered by the general one (Cat). If you think an answer is genuinely missing, tell us through the feedback form; lists are revised.
Why did it change what I typed?
Spelling is forgiving. If a typo, a plural or a close synonym can mean exactly one answer, the game puts that answer into the field for you to confirm. Press Enter to play it, or change it. It never scores a correction silently.
What is a good score?
Seven guesses is the best anyone can do; nine or ten is very good; the median is usually around eleven or twelve. More telling than the number is the verdict at the end: how many players you were more typical than.
Can I play yesterday's questions?
The archive shows the last six weeks: your own results, and the questions you missed. A question you did not play on its day can still be played from there.
Is there a German version?
Not yet. The lists are written for English and are not translated, because what counts as typical shifts with the language: mac and cheese ranks very differently in Ohio and in Bavaria. A German edition would need German lists, written from scratch.
Where do my results go?
They stay in your browser, on the device you played on. Clearing the site's data deletes them. We do not see them, and there is no account to attach them to.