Frivolous Friday: Solving Candy Land

Candy Land box cover

One of my most vivid childhood memories is the day I solved Candy Land.

In game theory, a game is considered to be solved when the outcome can be correctly predicted from any position. To put it another way: Chess will be solved (if it ever is) when an optimal strategy has been found by which one player can always win (or both/all players can force a draw. In the computer world, solving games has long been an important benchmark, although “solving” a game is different from “being able to beat the best human players”: a computer first beat the world chess champion in 1997, but chess is not a solved game. (See the XKCD comic for a good, funny summary of which games have been solved, and which games are easy or hard for computers. Note: beer pong is on the list.)

I know what you’re thinking. “There’s no optimal strategy in Candy Land! The game is 100% luck! The outcome is entirely determined by the position of the shuffled cards, and there is literally no skill!” I know. I agree. Stay with me for a minute.

In my half-assed opinion than I came up with just this minute, an important stage in childhood development is learning the difference between things you can change and things you can’t. And for me at least, games had a lot to do with this. Figuring out that some games are 100% luck, some are 100% skill (or very close to it), and some have varying balances between the two, wound up being a useful metaphor for life. (My friends and I used to get very invested in the game War, for instance — a fact that now seems absurd, given that War is another 100% luck-based game.)

So I was playing alone one day, goofing around with the Candy Land set and playing against myself, and I had this revelation for the first time. Candy Land was entirely predetermined, with no skill involved. I was disappointed — I liked the game, what with all the colorful pictures of candy, and now it seemed like it was ruined. But at the same time, I was proud of myself for having figured this out. (It was similar to my pride in figuring out Santa Claus.)

So what did I do? What was my first reaction on learning this crucial truth about games and life?

I stacked the deck.

I thought, “Hm. If the game is totally decided based on how the cards are shuffled, I should be able to arrange the cards in a way that guarantees I’ll win.” I went through the deck, picked out a sequence of cards that would guarantee a win, and slipped them into the top of the deck, every other card.

So I stacked the deck, and called my mom in to play. She figured it out pretty quickly, though: I was smart enough to stack the deck, but not smart enough yet to be careful about making it a close game, and when I won in about ten moves, she gave me a look. I think she was disappointed that I had cheated, but also a little proud of my cleverness. I don’t think I ever played the game again after that: now that I knew, it seemed pointless.

I did something similar with counting-off games, although in that case I didn’t consider it cheating and still don’t. In my neighborhood, you decided who was It with various counting-off pre-games. “My mother and your mother were hanging up clothes. My mother socked your mother right in the nose. What color was the blood?” (Yes, kids are ghoulish.) If the count landed on you on the word “blood,” you said the name of a color: the count then continued by spelling out that word (“P-U-R-P-L-E”), and whoever it ended on was It. At some point in my career, I started thinking ahead of time at the beginning of the count: If it lands on me, how many people are in the circle, and how many letters are in the names of colors? What color should I give if I want to be It, or if I don’t? That wasn’t cheating: it was just solving the game. It was an algorithm any other player could use, and for all I know they did. It occurs to me now that this is almost the opposite of Candy Land: it’s entirely determined, not by luck, but by choice, where the counter starts the count and what color is given by the person they land on.

So when did you figure out that some games were 100% luck? Which game was it? And what did you do then? Did you keep playing the game? Did you tell your friends? Did you try to cheat?

Frivolous Fridays are the Orbit bloggers’ excuse to post about fun things we care about that may not have serious implications for atheism or social justice. Any day is a good day to write about whatever the heck we’re interested in (hey, we put “culture” in our tagline for a reason), but we sometimes have a hard time giving ourselves permission to do that. This is our way of encouraging each other to take a break from serious topics and have some fun. Check out what some of the other Orbiters are doing!

{advertisement}
Frivolous Friday: Solving Candy Land
{advertisement}
The Orbit is still fighting a SLAPP suit! Help defend freedom of speech, click here to find out more and donate!

7 thoughts on “Frivolous Friday: Solving Candy Land

  1. 2

    I believe modern (2000-teens) Candy Land uses a spinner rather than a deck of cards to drive play. (If memory serves, the current game board is simpler than the classic one as well, but I don’t think that makes a meaningful game-theory difference.) Anyway, a this-generation Greta will have to find a different way to chea… er, solve it.

  2. 3

    Re. when I first realized that some games were purely luck-based:

    That was probably War, sometime around 1st or 2nd grade.

    Re. the XKCD Game AI comic that you linked:

    That’s now a bit outdated. The AlphaGo self-learning program developed by Google DeepMind in London beat Lee Sedol, the 2nd-ranked human Go player, in four games out of five in Seoul in March 2016: https://deepmind.com/alpha-go .

    So Randall Monroe was correct to list Go under “focused R&D could change this”, although the techniques AlphaGo uses have much more general applications than just figuring out where to place the stones on the board.

    But it would be extremely difficult to solve Go, as opposed to just being more skilled than human players.

  3. 4

    At least the original version of Candyland was also somewhat infamous among gamers for the fact that, despite having fairly simple rules, the rules were nonetheless ambiguous and poorly written. Specifically, it wasn’t clear exactly how you won. Did you have to draw a purple card to end up on the last space before the Candy Castle precisely? Or were you supposed to treat the Candy Castle itself as its own space, possibly multi-colored? The latter would make for a shorter game, always a good idea.

    It’s still a great design for teaching the concepts of taking turns and accepting setbacks gracefully, without the need for either words or counting among kids. And I always had fun reading the box text in the voices of the different characters. Lord Licorice was my favorite.

  4. 5

    Back before Standardized Testing became the One True God in public education, I used to assign a project in my eighth grade math classes. Students worked in small groups to develop a game. Their game had to meet three criteria:

    1) The only role a random element can play was deciding order of play.
    2) The game must end with a definitive winner. (No loops.)
    3) The rules must be clear and consistent.

    I got an amazing variety of games, including three-player checkers on a hexagonal board to a three-dimensional version of tick-tack-toe. Creating a new game based on skill alone is a bigger task than you’d think.

  5. 6

    “Eeny-meeny-miny-mo” was the count-off rhyme of choice in my childhood, but it was so absurdly easy to game that, as I recall, none of us even tried to pretend we weren’t. You could stop the count on the end of the rhyming part (“…if he hollers, let him go / Eeny-meeny-miny-mo”), or you could add the tag “My mother told me to choose the very best one, and you are It”–except that you could end that tag with “you are It” or “you are NOT It,” as you preferred, or you could follow it up with “…and you are It–no, I changed my mind, you are.” Whether it would be a worthwhile endeavor to call out the person who was gaming the count mostly had to do with the relative social status of the caller-out and the called-out, which was also a valuable lesson, though not necessarily a fun one.

Comments are closed.