Wordle Lovers - Play the NYT Daily Game

[Herman Trivilino]

Wordle 861 4/6*

 

[fresh_42]

Wordle 862 3/6

 

[Orodruin]

I think it is interesting to find out how other people do these puzzles.

I have two starting words that I will always first and second unless the first gives a lot of information. They cover all vowels and a number of important consonants. I will make my guesses based on the result of that keeping mindful of when there is a necessity to make a guess that cannot be the solution to gain extra information to guarantee success, which I value higher than solving in fewer guesses.

Barrring any misshaps I will reach my longest streak a week from now (that one was broken by a stupid typo in the final guess )

Also

Wordle 862 4/6

 

[OmCheeto]

Wordle 862 4/6

 

[Herman Trivilino]

Wordle 862 2/6*



Could this be the end of my slump?

 

[Herman Trivilino]

It should be noted that it is not necessarily stupid to omit lettters you know are in the word. Many times it can give more information, allowing you to guarantee a solve where strictly keeping all known letters would not. This is why I don’t like ”hard” mode.

I have a friend who uses that same strategy. I prefer hard mode though. Perhaps it's because I never had an interest in developing my vocabulary during my formative years, so I'm not very good at guessing words or in general playing word games. Wordle is the only word game I ever liked.

 

[kuruman]

Wordle 862 3/6

I think it is interesting to find out how other people do these puzzles.

I am following a strategy of word elimination which requires having a list of all candidate words on a spreadsheet. My seed is always the same and chosen such that the number of words that have no letters in common with it is minimized. My seed has an exclusion percentage or 91%. This means that only 9% of the words on the list share zero words with the seed and remain viable. The idea is to maximize the probability of catching letters that are shared with the target word because 91% have at least one shared letter. Having zero shared letters, as was the case today, is no setback because the target word belongs to the limited list of 9%.

My second choice is contingent upon the response to the seed. I make a list of the words remaining viable but instead of trying to guess one, I look at the frequency distribution of the letters in the list. The idea is to form a 5-letter word high frequency letters. This step is intended to increase the probability of catching extra letters and of eliminating words that do not belong.

The third choice is contingent upon what I learned from the second attempt. I look at the viable candidates and see if I can form another word with high frequency letters. If the candidates are few in number, 5 or less, I see if I can find a word that eliminates all but one. A third step was not needed today.

This method has served me well in that I've had 100 Win % since I started. My streak was broken on occasion when I went on vacation where I had no internet access.

 

[gmax137]

Wordle 862 4/6

 

[sbrothy]

Wordle 862 3/6*





My "strategy" can be likened to how Dirk Gently from Douglas Adams' "whodunnit" stories goes about his "detective-work". It's a combination of trial&error, blind luck and a smidge of deduction. :)

 

[Orodruin]

The idea is to maximize the probability of catching letters that are shared with the target word because 91% have at least one shared letter. Having zero shared letters, as was the case today, is no setback because the target word belongs to the limited list of 9%.

I assume that is still the largest class of possibilities after the first guess? What is the distribution of classes giving shared results for your seed?

 

[kuruman]

What is the distribution of classes giving shared results for your seed?

That I haven't looked at, if I understand what you're asking.

 

[Orodruin]

That I haven't looked at, if I understand what you're asking.

To clarify: I am asking about the distribution of the sizes that the guess divides the words into, i.e., how large the groups are that remain given any particular outcome of the guess. For example how many of the words survive on the first guess. One might expect to be the largest group, but it is not necessarily a given.

 

[fresh_42]

This one was mean.

Wordle 863 4/6

 

[OmCheeto]

Wordle 863 4/6

 

[Herman Trivilino]

Wordle 863 4/6*

 

[sbrothy]

Wordle 863 5/6*





I hate when this happens.

 

[fresh_42]

Wordle 863 5/6*





I hate when this happens.

See mine, I had the same issue with the last letter.

 

[sbrothy]

See mine, I had the same issue with the last letter.

Heh yeah, but at least you guessed correctly the first time. :)

Guess it's also easier at the end.

 

[fresh_42]

Heh yeah, but at least you guessed correctly the first time. :)

Guess it's also easier at the end.

It may be easier, but it annoys to have chosen the wrong of the two.

 

[sbrothy]

It may be easier, but it annoys to have chosen the wrong of the two.

Indeed.

 

[gmax137]

Wordle 863 3/6

 

[kuruman]

Wordle 863 3/6

To clarify: I am asking about the distribution of the sizes that the guess divides the words into, i.e., how large the groups are that remain given any particular outcome of the guess. For example how many of the words survive on the first guess. One might expect to be the largest group, but it is not necessarily a given.

Thanks for the clarification, you verified my understanding of what you were asking. No, I don't look at the distribution of sizes because I consider it irrelevant to my method. Let my describe how I proceeded with today's puzzle.

The seed gave me this response . How many words have this pattern is irrelevant. Of relevance is which words have this pattern. Let me illustrate with a hypothetical example that matches today's response to my seed but is not related to today's puzzle.
Target word = AFTER
Seed word = SATIN
First response

(colors from my spreadsheet simulator)
At this point the words that fit this pattern are
ACTOR
AFTER
ALTAR
ALTER
APTLY
EXTRA
FETAL
METAL
OCTAL
PETAL
TOTAL
ULTRA
and the frequency distribution of the letters in them is

Now I look for a word that contains as many of the high frequency letters as possible. I discard T because I already know where it is (I ignore the duplication in TOTAL.) I choose RELAY for the second try.

The second response is

At this point the words that fit this pattern are
AFTER
ALTER
EXTRA
Entering as third guess one of the words that start with A, gives me probability that I will get the solution in 3 is 1/3 while the probability that will get the solution in at most 4 is 1.

 

[fresh_42]

Wordle 864 4/6

 

[OmCheeto]

Wordle 864 3/6

 

[sbrothy]

Wordle 864 5/6*

 

[Herman Trivilino]

Wordle 864 3/6*

 

[Herman Trivilino]

I am following a strategy of word elimination which requires having a list of all candidate words on a spreadsheet. My seed is always the same and chosen

Clever. It removes most of the frustration of playing, but it seems it would also remove most of the fun!

 

[sbrothy]

Clever. It removes most of the frustration of playing, but it seems it would also remove most of the fun!

I tend to agree. It seems to me be a little too geeky an approach. But you do you :)

 

[kuruman]

Wordle 864 3/6

 

[gmax137]

Wordle 864 3/6