WORDLE — The Mathematical Approach

Hearing several interesting responses about the word-guessing game Wordle, including a video recommended by our own D.J. Schreffler on solving Wordle using information theory , I revised my list of openers, to be a little more rigorous.

I am bemused that this math fellow is far more diligent and methodical than I could ever hope to be, using logarithmic equations and information theory to deduce his best guesses.

I would make a snarky comment about some people having too much time on their hands, but then again, I am writing a column about Wordle at the moment, so I cannot snark.

The Good Book should tell us to snark not, lest we be snarked. By the snark you snark on others, ye shall be snarked.

So, to eliminate as many high-probability words as possible, we attempt to eliminate the letters most frequently used in five letter words in their order pf frequency.

A E R O I S T L N   U Y C D H   M P B   G   K W F 

  • TARES or STARE
  • LINGO or LOGIN
  • DUCHY
  • ABAMP or POMBE 
  • WAKFS 

Leaves

V Z X J Q

The first two guesses eliminate the first nine most frequent letters at the outset plus G.

I do like that third guess eliminates all of the next five most frequent letters,  UYCDH, in one word. This eliminates all the vowels, including Y.

I do not like the use of “WAKFS” or “POMBE”  in the fourth and fifth guesses, as these are rare foreign words.

I am undecided between STARE and TARES as the first guess. Does it matter?  It is more likely to generate any additional information about position if the S is in the first position, not the last?

Also, no one who is not a Bible reader is likely to know what the word “tares” means. I notice the fellow who did the video mentioned above did not.