Monday, August 24, 2009

Nigel Richards Article by Stefan Fatsis

NASPA Newsletter: Vol. 1, Issue 2

Sunday, August 16, 2009

When is a word a word?

Yes: negative studying is such a weakness of mine. And probably a big weakness of a lot of intermediate players. How many minutes do you think we all waste looking for sevens and eights that don't exist?

Sometimes I wish Zyzzyva would throw in a curveball such as AEILORT just to see if we're awake.

The problem with anagram quizzes is that we *know* there's a solution, so it's just a matter of finding it. Every week I get about 12-14 on David's "not-so-toughs", but I really think this flatters me. It's easy to find a seven from ABEEGPW when someone tells you there is one, but if that selection turned up in a real game, many of us probably wouldn't even look.

I wonder if anyone has any good tips for learning those non-words? Any foolproof nmemonics? Some that come to mind are:

IRELAND (every kid learns this one)
A SARNIE
RED LION
URINALE (I remember U & I LEARN + FAB MATHS)
SAURINE (should mean lizard-like but doesn't)
LUTRINE (should mean otter-like but doesn't)
RELATIO
AGONITE (and its plural)
TAENOID

I remember once seeing a list of the commonest non-word alphagrams, but this was pre-2007, and (come to think of it) probably TWL.

Is there anything out there you'd recommend?



Howard Warner

Monday, April 27, 2009

Statistics of word play by word length




>I'm looking for a list of word lengths with their percentages for
how much they make up the scores in games?

David,

I don't know it based on scoring, but I think playability is a much more reliable measure of a word's value to Scrabble players.  So these results were easily computed from John O'Laughlin's march 2006 playability list:

I sorted the groups of words by average playability per word.  This assumes the study time is proportional to the number of words alone (and not their length), and so is a rough guide of what order to learn these groups in (if playability is indeed a good measure of word importance, which I have found it to be - although I really suggest studying directly from the entire list).

The 2s make up 15.14% of the dictionary.  The 8s make up 24%.

It's interesting that playability suggests the JQXZ fours are more valuable per word than the 3s.  And that if you learn any of Mike Baron's bingo lists, you should start with the type III 8s.  Also, your time per word is better spent studying the first 1000 most probable 7s and 8s before starting on the 4s.  Quite different from Mike's recommended study order.  :)

I can easily run more stats for more word groups if anyone is interested.  Enjoy.

twos.input.txt
total playability: 365067920
percentage of dictionary: 0.151383539252
playability per word: 3614533.86139

JQXZ fours.input.txt
total playability: 82518801
percentage of dictionary: 0.0342182576607
playability per word: 306761.342007

threes.input.txt
total playability: 269987671
percentage of dictionary: 0.111956397567
playability per word: 265997.705419

JQXZ fives.input.txt
total playability: 85903362
percentage of dictionary: 0.035621741218
playability per word: 123247.291248

type III eights.input.txt
total playability: 12713430
percentage of dictionary: 0.00527190674391
playability per word: 84756.2

first 1000 probable eights.input.txt
total playability: 82582188
percentage of dictionary: 0.0342445424912
playability per word: 82582.188

first 1000 probable sevens.input.txt
total playability: 75138611
percentage of dictionary: 0.031157897598
playability per word: 75138.611

fours.input.txt
total playability: 268405002
percentage of dictionary: 0.111300108637
playability per word: 66601.737469

type I sevens.input.txt
total playability: 100290232
percentage of dictionary: 0.0415875771078
playability per word: 49599.5212661

type II eights.input.txt
total playability: 40653196
percentage of dictionary: 0.0168577526406
playability per word: 45070.0620843

type I eights.input.txt
total playability: 207199188
percentage of dictionary: 0.0859197554516
playability per word: 37032.9201072

highfives.input.txt
total playability: 96288222
percentage of dictionary: 0.0399280545786
playability per word: 36723.1967963

type II sevens.input.txt
total playability: 58078873
percentage of dictionary: 0.0240836974953
playability per word: 36619.7181589

type III sevens.input.txt
total playability: 94362310
percentage of dictionary: 0.039129432298
playability per word: 33833.7432772

fives.input.txt
total playability: 294498122
percentage of dictionary: 0.122120201664
playability per word: 32948.9955247

eights.input.txt
total playability: 580618144
percentage of dictionary: 0.240766237671
playability per word: 19506.085601

sevens.input.txt
total playability: 436826648
percentage of dictionary: 0.18113989313
playability per word: 18179.1438678

sixes.input.txt
total playability: 168069787
percentage of dictionary: 0.0696938783266
playability per word: 10645.4134153



------------------------------------

Yahoo! Groups Links

<*> To visit your group on the web, go to:
   http://groups.yahoo.com/group/crossword-games-pro/

<*> Your email settings:
   Individual Email | Traditional

<*> To change settings online go to:
   http://groups.yahoo.com/group/crossword-games-pro/join
   (Yahoo! ID required)

<*> To change settings via email:
   mailto:crossword-games-pro-digest@yahoogroups.com
   mailto:crossword-games-pro-fullfeatured@yahoogroups.com

<*> To unsubscribe from this group, send an email to:
   crossword-games-pro-unsubscribe@yahoogroups.com

<*> Your use of Yahoo! Groups is subject to:
   http://docs.yahoo.com/info/terms/


Tuesday, March 6, 2007

Some word list manipulation ideas

Euphoria speller program by Junko proved to be most effective and
speedy for comparing two lists and finding out words not in the second
list (in this case the sowpods word list).

BK Replacem is the best for tagging specific words with an XXX etc.
Create a search and replace import file with search word and
replacement word in each line. Replace with exact word option. Later
the tagged words can be bookmarked using Textpad and removed.

Anagrams database file can be converted to text by sqlite3 using the
following steps

sqlite3 anagrams.db
.output example.txt
.dump
.quit

command for recreating sqlite3 db file

cat anagrams.txt|sqlite3 example.db

Friday, February 9, 2007

Thursday, February 8, 2007