Fermi Questions (fermiquestions.org) is a Wordle-like game where you try to guess the answer to Fermi estimation questions with 6 or less tries. After each guess, you'll see if your answer was too high or too low. You win if your guess is within ±20% of the correct answer.
Example questions:
- How many new cars were sold in the US in 2024?
- How many humans have ever lived (including those currently alive)?
- How many chickens are slaughtered for meat every year?
The skill of Fermi estimation is also extremely useful if you want to perform better here on Manifold. As Philip Tetlock shows in his book Superforecasting, many of the best forecasters break down complex questions into smaller, more manageable components which is exactly what you can practice when playing the game.
In the first 24 hours after going live it had some 1,100 unique visitors, which primarily came from Hacker News. Resolution source is the unique visitors count on the publicly available analytics dashboard: https://plausible.io/fermiquestions.org
Update 2025-08-04 (PST) (AI summary of creator comment): In addition to the default bot-filtering by the Plausible analytics tool, the creator reserves the right to manually exclude any suspicious activity from the final unique visitor count.
Update 2025-08-04 (PST) (AI summary of creator comment): A unique visitor is defined as a distinct visitor to the website within a 24-hour period, based on their IP address. A user who visits on different days will be counted as a new unique visitor for each day.
1,000
3.00I've greatly appreciated people here sharing screenshots of their game statistics; please continue to do so! This is great fun!
Another clarification that might be important: For the purpose of this market a unique visitor is defined as a distinct visitor to the website within a 24-hour period, based on their IP address. So if the same user returns on a different day, they are counted by the analytics dashboard as a new unique visitor.
@Fion I like that! I'm currently working on improving the analytics for individual questions, and I find it fascinating to show what the median first guess of players was, along with some additional information. This is probably quite unique data we're gathering here because, to my knowledge, there is very little information on how well calibrated people are when answering estimation questions like the ones we have in our game.
After that is completed, I'll get to the revamp of the personal statistics and will certainly include the suggestion you've made about showing how often your initial guess was too high/low.
@whatiscrit This should be displayed in the statistics modal in the coming days, but 3.4 is certainly above average. When it comes to decreasing the max number of guesses, I'm somewhat unsure. I'm inspired here by Wordle which gives almost all players a feeling of success, with average players getting like 95% of all words right in 6 or less tries. The challenge really has to be to get the answer in as few tries as possible.
I love this game!
One thing that irks me: setting thresholds at +-X% (of the true answer) creates an assymetry between higher and lower guesses. This is especially grating with the +-50% threshold for orange: "orange up" means the answer is less than double your guess, while "orange down" means it's more than 2/3 of your guess - much more constraining!
It would be neater if the thresholds depended on whether the guess and and the true answer differ by X% of the larger one of the two.
@AhronMaline Yeah, ideally everything should be in log space IMO, but I guess that's not as friendly for the average user.
@A I'm sure there's a way to phrase "off by a ratio of 5:6 between the larger and the smaller number" without explicitly invoking logspace...
@Tarl Yes, I’ll start with some marketing in the coming weeks. Until now I’ve refrained from doing much in this direction because I wanted to improve the site first. Once I do some marketing, I will probably start with giving out prizes via a raffle to people who share on social media how many guesses they needed. I’ll likely also attempt some guerilla marketing such as donating money live to streamers (NorthernLion, etc.) who play similar games.
That said, I think it'll be pretty tough to get to the 100,000 unique visitors number even if I choose to spend $2-3k on marketing which is the plan as of now.
Love the game.
I discovered my model of the world concerning consumer products is too skewed (I didn't expect so many people would pay for Spotify or buy Apple goods, so my first estimates were way too low).
I read through the comments and like all of the changes, the tips actually helped a lot a couple of times. On the topic of sharing stats and seeing how you compare to others, I suggest using ⏫⏬🔼🔽 to differentiate far off and close guesses and using a similar style to chessle, quoting today's chessle:
Today, 46% of people got Expert mode correct.
Of the people who got it correct, it took 4.62 tries on average.
Also I really appreciate that you can keep playing through the archive of old questions, I like going at my own pace, somedays I won't feel like playing and other I'll play several in a row.
@TenShino Thanks for your kind words! I really like the idea to show how many tries it took the average player, and this will be implemented at some point in the near future.
@TenShino You can now see how many tries it took other players on average to answer any question. It doesn't look pretty and I think I'll change where it is displayed, but it is now live.
There is also this problem where it only counts from now on, and older questions have not much data since very few (if any) have played them since I implemented this feature a few hours ago.
Fermi Question of the Day: "How many Americans who served in World War II were still alive as of 2024?"
⬆️⬇️✅
@Robincvgr I need to memorize populations, that's gonna help me a lot here
I guessed the EU's population was like 350M (pretty off) and that the ratio of police officers is about 0.002 because that's about what it is in my town
@TheAllMemeingEye If you visit the site from the same browser, it should show your previous progress. Maybe you have initially played the game by opening the link here on Manifold, and now you’ve looked it up in a different browser? Anyways, at some point I’ll also allow the creation of accounts so that we don’t have to rely only on using local storage.