Defensive strategies for Appletters

Since Appletters involves building a zigzagging snake-like line of words, where new words can only be built off the beginning or end of the snake, the options for play can be sometimes highly constrained. Devious players can use this to their advantage.

For instance, you can position words such that you are able to form words from a position that others are unlikely to be able to. As in the example below, where the presence of the word SLIP forces new words starting from the V to be built only to the left, making the V tricky to deal with.

   SPORK
 V L
 I I
 MOP

Among the very few words that end with V are rev and shiv.

    SPORK
REV L
  I I
  MOP

Similarly, there are few words that start with X. Examples include xenon (the atomic element), xi (the Greek letter), xebec (a kind of three-masted Mediterranean ship), xylem (the part of a plant that transports water - like a plant's circulatory system), and xeric (super dry, desert-like).

I think that the only people who will be able to generate words that end in a J or Q are people who have deliberately memorized such words for such purpose as they all look like questionable Scrabble jibberish to me. By properly constructing the snake, one or both ends can almost be completely cut off using this tactic.

There's also what I call the Ouroboros stratagem.

Suppose you have manipulated the situation so the snake looks like this:

    SPORK
REV L   N
E I I   I
L MOP   V
I       E
C       S

A really fiendishly clever thing to do is to build a word that joins up the beginnings and end
of the snake, like so:

    SPORK
REV L   N
E I I   I
L MOP   V
I       E
CALENDARS

If it is allowed (and I can find nothing in the instructions that forbids it), finding a way to make the Appletters snake into a closed loop pretty definitively ends the game. In my book, whoever pulls off such a stunt should instantly win, as when you sink the eight-ball on the break in pool.

Scrabble reformers: Donald Sauter and the Dover Scrabble Club

In contrast with the Scrabble reform proposed by Peter Roizen (see the previous post on WildWords), Donald Sauter's complaints are more of the "95 Theses" number. When he says he wants to fix Scrabble, some of the complaints he lists are:

• Bluffing was never a part of the original game of Scrabble and was only introduced in 1976.
  
• The Official Scrabble Players Dictionary has been anointed as the definitive word list, in spite of the fact that it contains lots and lots of bogus words (no longer printed in any dictionary), suggesting that about 5% of all OSPD words may be in this category.
  
• As a corollary, conventional Scrabble encourages lots of little words to be played.
  
• The standard Scrabble tile distribution gets boring after a while.

His solution?

Scrabble II: Amongst the new rules:

• Play with 8 tiles on each player's rack instead of 7 and allow bonuses for 6-, 7-, and 8-tile plays.
  
• Use a real dictionary (but toss some of its more made-up, iffy, or downright ridiculous words).
  
• Allow words to stretch beyond the borders of the standard 15-by-15 board. (The extended board is such a cool idea!)
  
• Forbid plays where only a single two-letter word is formed.
  
• Mix three tile sets together and select a new, random tile distribution for each game.

Here are the full rules of Scrabble II, and a far more detailed description of the motivations behind and playing of Scrabble II. The latter is particularly fun to read because it is so spirited and analytical, and as a bonus it contains what amounts to a manifesto on two-letter words.

---

There are now tiny little ad hoc Bananagrams tournaments sprouting up in the U.S. without any organization or permanence to them. This is probably how Scrabble tournaments started up. Whether this will develop into anything like nationwide tournaments is yet to be seen. If it does, I would hope that some of the mistakes of the Scrabble tournament scene could be avoided. In particular, I would like to see competition divided into different groupings: one for the word-memorizers and one for the more casual player. Ideally, a newcomer to Bananagrams should be able to learn the rules of the game and start competing on the first day.

As Stefan Fatsis says in Word Freak (his book on the tournament Scrabble subculture), "Recruiting new players is Scrabble's toughest task." In addition to helping distinguish it from being just a "speed Scrabble" tournament, eliminating the requirement to learn a bunch of arbitrary letter sequences would make a Bananagrams tournament more inclusive and open and fun.

Synchronized Bananagramming and other ways to play Bananagrams

In thinking about games of Bananagrams I have played, I've realized that the amount of time it takes me to finish a grid can depend heavily on what letters I get and what order I get them in. Thus, the winner of any single game can depend heavily on the randomness of the drawn tiles.

Here is a way to take a massive amount of the randomness out of a Bananagrams game: Have someone who is not going to play, sort through all the letters and give each player an identical set of 21 (or however many) tiles to start off with. Also set up identical queues of tiles for each player to peel from. The outcome of such a match should be determined extremely strongly by the actual performances of the players.

Admittedly, it takes a lot of the drama out of the game and makes it a very different exercise. Played in a less competitive fashion, this would be a really strange and unique game: Imagine the reactions as all players peel a Q at the same time. It would be almost like everyone was sharing the same experience. Comparing grids at the end would also be interesting as it would better highlight the different approaches and tricks of each player.

This idea suggests to me a puzzle: Is there a way that two players could agree on an algorithm ahead of time which would allow them to make the same (or nearly the same) word grid? In principle, of course, you could just give them the same instructions as a Bananagrams-playing computer program, so the better question is whether there is a relatively simple way of doing it?

One more twist: It would be interesting to try to play a game where the objective was for players to make maximally similar grids. Without looking at the other player's grid, it would have a sort of mind-reading, Taboo-ish potential. If played where the players could look at the other players' grids (and where we would go back to a standard random drawing of tiles), it would add a sort of frantic element to gameplay.

If you try any of these Bananagrams variations or have ideas about solving the puzzle, feel free to contact me. I'd love to hear your game reports or other feedback.

Dixit Review

My new favorite game is Dixit. It is this cool game, originally from France, in which everyone has a hand of cards with fanciful illustrations on them, and one player has to say something (a word, a phrase, a sentence, a sound effect) and place one of their cards face down. Everyone else adds a card to try to fool the other players into picking their card. I find that the trick is to pick a card that is close to the theme, but not too obvious about it. The "storyteller" player wants to get neither all of the players to choose his card nor none of the players. Something in between earns him three points. Players who pick his card earn points. Players whose card is picked by others earn points (which allow the players to move their rabbit pieces around the board). The first to the finish wins.

Gameplay resembles a combination of Apples to Apples and Balderdash, but, in my opinion, Dixit is a much better game. It's a fun and unique challenge. The artwork is gorgeous. Everyone I've played with seems to love this game. Also, you don't just have to take my word for it. Dixit was recently awarded the coveted Spiel des Jahres award, so the Germans like it too.

If you want more Dixit cards (as I expect you eventually will), you can get them through an expansion pack: Dixit 2.

Stupid Bananagrams tricks: Word grid with mirror symmetry

I thought it would be an interesting challenge to try to make a perfectly symmetric Bananagrams word grid using all 144 Bananagrams tiles. I decided to use mirror symmetry, so a word spelled out horizontally on the left hand side would have to be reversed on the right hand side and yet still spell a real word. Can it be done? Let's find out.

First, I consulted my letter distribution for Bananagrams to see which letters occurred an odd number of times in a Banangrams set. For instance, there are three W tiles, so I knew that at least one (and possibly all three) would have to occur on the center line of the grid to maintain mirror symmetry.

I then tried to build those letters into vertical words. Next I built some horizontal palindromes off of those center-line words. I got some good palindromic words from Wiktionary.

Finally, I worked on linking everything up. This step took the longest time. Fortunately, I found a list of words, that when reversed, yield different English words (like "tuba" and "abut"). These are called "semordnilaps". I particularly liked the ERGO/OGRE pair, so I managed to work that one into my grid. After a lot of fiddling around, I wound up with the grid shown below.

C H GAG R M OH E HO I MADAM I N A A N KNOW WONK E B E U EYE U R R R R OGRE ERGO BUN NUB O O X J P J X RECAPS A SPACER S I L V L I S L A E A L T T Q DEIFIED Q U A L A U I Z O Z I TIDE EWE EDIT V O O V Y TIN NIT Y F T F STATS D

Voila! The world's first palindromic Bananagrams grid!

Banana Flambe - A tribute to Abe

Among the tributes to the memory of Bananagrams creator Abraham Nathanson is this wonderful story about how he was once approached to collaborate in a game-designing competition as part of Providence's first Maker Faire. The story gives a sense of who Abe was and what the community of Providence is like. Enjoy!

Spanish Bananagrams tiles spotted in the wild!

Since the introduction of the Spanish version of Bananagrams, I've been bugged by the fact that none of the online photos showed what the extra Spanish tiles looked like. And then I found this creative attempt to make sentences using all the tiles in a Bananagrams set. As the set on hand happened to be Spanish Bananagrams, three of the four special Spanish Bananagrams tiles are visible in the picture: LL, RR, and CH:

There you have it folks: actual proof that Spanish Bananagrams is not merely a figment of my imagination. (Though I admit that the image of the CH tile is almost as fuzzy as a Bigfoot photo.)