Box Relatives

Thoughts about puzzles, math, coding, and miscellaneous

Puzzle Party Idea


Here in the Los Angeles area, puzzle-minded people like myself gather every two months for what we like to call a “puzzle party.” Several regulars will bring puzzles or games and we’ll solve them as a group. In the most recent one I presented a pairs solving puzzle (but one that’s doable solo) — you can download the PDF by clicking here.

Still, I got to thinking about what I might present for a future installment, and I came up with a game I’d like to call “Crosswordash.” I need some help perfecting the idea, though. If you’re interested in adding your thoughts, I’d love to hear them. Details after the jump …

If you’ve been following me on Twitter, you may have read a tweet or two about a certain terrible crossword that for some reason is still popular among local crosswords. It is *so* bad that it inspired an idea for a new game. As you may have guessed from the name, “Crosswordash” is a crossword-type version of Balderdash, but instead of words for which you make fake definitions, in this game I give you a crossword entry and you write a particularly bad clue. Specifically, the game goes as follows in each round:

  1. The emcee gives a crossword entry from one of these puzzles (and, preferably, a date on which the entry appeared).
  2. Each team then writes down a possible crossword clue for the entry. Teams will be alerted ahead of time not to worry if they’ve never written a crossword clue before, as it doesn’t appear the people who make these puzzles have either.
  3. The emcee then reads off each team’s clue plus the real clue in random order.
  4. A round of voting happens in which each team tries to determine which is the true answer. Additionally, each team has the option to vote for their “favorite” answer — usually a made-up one that is clearly wrong but funny.
  5. Scoring proceeds as in regular Balderdash — one point for every team that guessed your answer, two points if your team guessed the correct answer, and three points if your submitted definition was close enough to the actual answer. There’s also one “like” point for each favorite you got; these points are kept separate so that at the end, there are two different winners.

I think this could work because in some ways it’s very similar to Balderdash, but with an extra twist of awfulness. For instance, there’s CYME = [Flat topped cluster of flowers] and HENO = [Tewa village], but also things like SEBD = [Switchboard (abbr.)] and SUNN = [Crotolaria juncea] (?!?!). To top it off, the game should include plenty of common words with horrible clues, like HER = [Of she], MA = [1/1000 of an ampere], and METE = [A boundary line].

So now I need your help. Do you think this could be a fun game? Do I need to make any tweaks for it to be worthwhile? Please let me know in the comments. Thanks!


  1. Balderdash games do well at minicons, and the Press Crosswords clues and entries would be a great creative inspiration for humorously bad decoys. I also like the “like” points idea. Even though, as you point out, crossword puzzle experience is not necessary to play the game, I would recommend that you try to seed each team with at least one person with some crossword interest: While the clues are bad they generally adhere to a basic editorial style and a savvy player can offer guidance in this regard to teammates who do not regularly solve American crossword puzzles. Also, you may encounter a challenge in dividing the group into a manageable number of teams yet not have teams be so large as to hinder all players from being able to make a contribution. This probably won’t be a problem with normal minicon attendance but if large numbers show up then you might consider a contingency plan, e.g. dividing the group into ten teams instead of five and having one set of five teams vote on the other sets’ decoys in alternate rounds.

  2. Thanks for the comments, T. I’m not sure about the “basic editorial style” — I’ve seen plenty of clues that break Pat Merrell’s clue rules (as well as other rules like “spell the words correctly”) but I think some intelligent pairing might be helpful. The point about large numbers is also a good one — I may end up just breaking up the crowd into smaller groups and not having an emcee if that happens.

    I’m also worried about the speed of the game. I don’t want it to drag on forever. It might be worth play-testing it with another group before launching it for real.

Leave a Reply

Required fields are marked *.