(a summary of play-tester feedback on the rules here)

Although this had way less feedback than other games (I can imagine several reasons why), those who played it mostly dug it.

humility feedback

“It’s trivia for people who understand error,” one play-tester sagely noted. To me, either that comment or this one summarized it best:

I loved how this game normalizes (and encourages!) being open and precise about how little you know about something.

Or in a snarkier vein:

Finally the confident, smooth-talking factoid dispensers get their comeuppance.

Or perhaps most accurate of all:

I learnt the hard way that the world’s largest pizza was VERY heavy.



Can’t you win by just putting a range of “zero to infinity” for every question?

Nope. This ruse, though admirably sneaky, won’t work.

For example, say we’re guessing the height of the tallest living tree. Our players give the following guesses:

  • A: 100 to 10,000 feet
  • B: 50 feet to 50,000,000 feet
  • C: 0 feet to 1,000,000,000 feet

Each is very wide. But notice that A’s is semi-reasonable (“no more than two miles”), B’s shows at least a bit of thought, and C’s shows no thought whatsoever. As a result, even though they all got it right (the answer is 383 feet), A scores 2, B scores 1, and C scores 0.

Thus, if your opponents are going super wide, you can win by being just a little more reasonable. Which makes going super wide every time a losing tactic.

Why is the scoring system so confusing?

Mostly to create the right incentives, as in the example above. My explanation has gone through lots of iterations; this is the best I’ve got so far, but I’ll keep working on it.

What happens if there’s a tie?

I favor splitting the points in half (e.g., 2.5 for each). But other solutions work too.

What if you know the exact answer?

Feel free to put a range of width zero. For example, ask me the single-season NHL record for goals scored, and I’ll say “92 to 92,” because I know my Wayne Gretzky facts.

Shouldn’t you score this with a ratio, rather than a difference? For example, isn’t “1 to 1000” in some sense a much wider range than “300 to 1500”?

You’re right. For simplicity’s sake, I’ve described the rules using a difference. And for small enough numbers, or numbers where the order of magnitude is obvious (such as a person’s age), it’s probably more sensible to use the difference anyway.

But for big numbers, where the guesses will span several orders of magnitudes, you should probably divide, rather than subtract, to calculate the “width” of your range.

(Shout-out to play-tester Kim for an excellent discussion of these issues. Thank you!)


Coming Up with Questions

Some players (especially kids) found this a frustrating and difficult part of the game. Others relished it.

First, it’s a good idea to be as specific as possible:

We found it helpful to specify units and dates (population of Cambodia as of 2018 and average distance from Saturn to the Sun in miles were some of our examples)

Here are great suggestions from play-tester Benjamin (who has actually played this same game in his social circle for years; mine is an independent “discovery” of the same rules):

1. world records – most hot-dogs eaten in 1 minute, fastest rubik’s cube solve using feet…

2. year when stuff came out – movies, books, songs, earliest recording of a phrase…

3. age – of celebrities and of fictional characters. we liked to ask for combined or average ages, like the average age of all the actors in the Avengers.

4. distances and travel time – it is very easy to google aerial distance between 2 place (kremlin and pentagon), and it is also easy to get a travel time estimate (using google maps or even a traffic informed app). bonus – travel time using public transport only.

5. mathematical estimation of huge number – for example, the square root of X!, X being the number of players. work well with college students.

I also loved Joe Kisenwether’s idea here:

In one game, we tried to come up with questions which were resolved by experiment How many inches high will the geyser go when I drop these Mentos into this coke?  How many nuts will there be in this unopened bag?

Here were some of my favorite specific ideas:

  • Hottest land temperature ever recorded
  • Current day of the month (without looking)
  • Average number of petals on a cultivated rose
  • Number of episodes of The Late Show that David Letterman hosted
  • Age of Britney Spears
  • Number of books Agatha Christie wrote
  • Number of in-ground pools in Texas
  • Average weight of a humpback whale
  • 1 knot, converted to mph
  • Number of selfie-related deaths worldwide from 10/2011 to 11/2017
  • Number of neurons in the human brain
  • Atomic number of Xenon
  • Percentage of U.S. adults who believe chocolate milk comes from brown cows
  • Number of Billboard #1 songs by Katy Perry minus number of Billboard #1 songs by Johnny Cash (answer may be negative!)
  • Number of species of birds that can fly backwards
  • Number of words (noun or verb) that the Scots have for “snow”
  • The 1000th prime number
  • Number of words in Hamlet’s soliloquy (“to be or not to be…”)
  • Total length of every episode of the Simpsons
  • Number of paperclips manufactured each year
  • Height of tallest-ever ice cream cone
  • Number of known beetle species
  • Population of Atlantic Puffins
  • Number of books in the New Testament

And finally, some sage advice:

The players should negotiate any restrictions (“no questions about a porn star’s number of partners”) before playing.


What Should I Call the Game?

I originally called the game “Humility.” Then, my friend Adam Bildersee, who is much cleverer than I am, suggested “Out-range-ous.”

The folks I met at a Protospiel play-testing conference found this a little too punny, and preferred “Humility.” By contrast, you all showed a small but meaningful preference for “Out-range-ous,” though I enjoyed the suggestion “Fun Within Limits.” Final name TBD.


Other Thoughts

The rules say that everyone should write down their answer before hearing anyone else’s. But tweaking this rule can add some drama:

Towards the end we let the teams with low scores answer only after they heard the other answers – makes it more interesting. [Also,] time limits make it more fun and exciting, we usually use 2 minutes (for teams to discuss).

Dylan Kane noted the similarity to a classic event at Mathfest and elsewhere: the Estimathon.

The difference is that it’s run like trivia with teams of 4-8 people. Teams receive 10 questions to answer in 30 minutes.

Two big differences.

First, scoring is logarithmic and lowest-score-wins. You divide the larger guess by the smaller guess, and that is your score. I.e. guessing 30 to 60 would be worth 2 points, 20 to 120 worth 6 points, etc. Always round up so scores are a whole number. Lowest score wins, and wrong answers double your total score (with the trick of adding 10 points at the start so that answering all questions wrong doesn’t result in a score of 0).

Second, each team gets 20 guesses, so they can improve on their initial guesses – once for every question, ten times for one question, or anything in between.

The gamerunner projects every team’s score and which questions they have right and wrong. Makes for a really exciting last few minutes.

The largest challenge I’ve found running it with students is that they often lack humility and really struggle to guess with a sufficient range, and I have trouble calibrating the difficulty (which is tricky because in an estimathon all the questions are handed out at once).

Here’s a brilliant idea for how to penalize overconfident guesses:

In lieu of any tangible penalty other than failure to score points, it has been suggested that the worst guess (narrowest of the losers, or widest of the winners if none) has to wear a silly hat until the next round.

Given the present (and past) quality and quantity of our presidential primary debates, I would suggest that 1 debate be devoted to the presidential candidates playing this game. Each candidate produces 5 questions, 1 for each of the following categories: foreign policy, economy, health care, education and *1 wild card*.

I remember Amy Klobuchar being ridiculed for not knowing who the president of Mexico was and Andrew Yang being asked how many aircraft carriers the US Navy had (this would be a great question for the game, actually).

I have no strong feeling about which candidates from this current/past cycle would be best at this game.