-
Follow our Instagram!
-
Welcome to Smogon! Take a moment to read the Introduction to Smogon for a run-down on everything Smogon, and make sure you take some time to read the global rules.
wordle? i dont even know her!
- Thread starter faint
- Start date
prepare for mind blown obsessive myzozoa post
ok I will tell u the algorithm I use. note u don't actually need an algorithm and I win 95%+ of hello wordls on max hard mode (did 16 in a row this morning, dm for screen shots of the series) just entering the previous games' answer as the first guess and going from there.
optimal vowel algorithm:
first word: peony. the reason ur first word is bad is because it doesn't tell u enough about certain letters. ex: if ur first word has a y in it and the y isn't the fifth letter, ur word is garbage. You should never put a Y into a first guess word except in the 5th slot. if the y in a word isnt in the 5th letter slot u want to know that asap not 2 guesses after u found out y was in the word. The other reason peony is good is cause it has 3 vowels and uses 2 of the more annoying and useless consonants right away (p and n).
after u put in peony u should improvise. in 'real' wordle u aren't allowed to not use clues youve acquired so if you hit on your first word u will have to adapt. no going from adieu->forty everytime or w.e your calculator tells u is optimal that is in fact not optimal at all.
if u miss every letter with peony you should go straight into 'trial' as your second guess ('triad' is worse cause u cant use the D later where as L is another garbage letter)
example: if u hit on the E of peony but it's not the 2nd letter you can to 'quiet' as guess number 2, eliminating 4 really important letters and finding the most likely location of the e. (and actually absurdle thinks that hitting on the unknown e is the worst outcome for the peony start, but you can still solve absurdle with the peony start w/o any calculators using the 'quiet' continuation:
)
Moving on, after 3 guesses you will usually have eliminated most word structures and will need to decide if and how to go about testing for double letters. In general, by 4th guess in hard mode, u should either be eliminating possible double letters or guessing directly at the word.
anyway this should get u started with making ur own anti-method for wordle. the 2 words peony and trial are perfect words, every letter in these words is in the correct place for maximizing information, i.e in the place the letter is likely to show up in a 5 letter word. u may think ur adieu+forty combo is better but it really isn't, how often is d gonna be the second letter in a word, ask urself. how often is u going to be the last letter? how often is r going to be the 3rd letter? this is why peony+trial is objectively vastly better than these supposedly 'mathematical' algorithms, the letters in this words are in their probabilistic locations relative to the locations within the set of 5 letter words.
Optimal 'Algorithm'
Ok so I've now taught u how to play wordle like a basic bitch. You can win 100% of games with these methods, however in all likelihood u will end up with a win rate closer to 95-97/100 games. How, you ask, do you get 100% accuracy?
To get 100% accuracy you need to completely discard the vowel based strategies we've already mentioned, like starting with adieu. Peony+Trial is definitely objectively the best way to begin a vowel based algorithm, someday your mathematics will demonstrate it, but for now just trust me. However, even this optimized method has to be completely set aside to get 100% accuracy. To get 100% accuracy we need to start with words that use as few vowels as possible, 4 consonants and 1 vowel words are the basis of this method that allows us to find the structure of the word in as few guesses possible, saving us guesses for situations when many possible letter combinations could fill in a word.
The main reason the vowel method has to be discarded is because vowels, for the 99.9 percentile lexicon wordlers that are capable of 100% accuracy over 100+ games, are basically entirely deducible from the consonant structure. Thus, the strategy for achieving 100% accuracy recognizes that finding the consonants is a much higher priority than vowels when facing extremely difficult word structures (i.e word structures that require trial and error to solve).
ok I will tell u the algorithm I use. note u don't actually need an algorithm and I win 95%+ of hello wordls on max hard mode (did 16 in a row this morning, dm for screen shots of the series) just entering the previous games' answer as the first guess and going from there.
optimal vowel algorithm:
first word: peony. the reason ur first word is bad is because it doesn't tell u enough about certain letters. ex: if ur first word has a y in it and the y isn't the fifth letter, ur word is garbage. You should never put a Y into a first guess word except in the 5th slot. if the y in a word isnt in the 5th letter slot u want to know that asap not 2 guesses after u found out y was in the word. The other reason peony is good is cause it has 3 vowels and uses 2 of the more annoying and useless consonants right away (p and n).
after u put in peony u should improvise. in 'real' wordle u aren't allowed to not use clues youve acquired so if you hit on your first word u will have to adapt. no going from adieu->forty everytime or w.e your calculator tells u is optimal that is in fact not optimal at all.
if u miss every letter with peony you should go straight into 'trial' as your second guess ('triad' is worse cause u cant use the D later where as L is another garbage letter)
example: if u hit on the E of peony but it's not the 2nd letter you can to 'quiet' as guess number 2, eliminating 4 really important letters and finding the most likely location of the e. (and actually absurdle thinks that hitting on the unknown e is the worst outcome for the peony start, but you can still solve absurdle with the peony start w/o any calculators using the 'quiet' continuation:
Moving on, after 3 guesses you will usually have eliminated most word structures and will need to decide if and how to go about testing for double letters. In general, by 4th guess in hard mode, u should either be eliminating possible double letters or guessing directly at the word.
anyway this should get u started with making ur own anti-method for wordle. the 2 words peony and trial are perfect words, every letter in these words is in the correct place for maximizing information, i.e in the place the letter is likely to show up in a 5 letter word. u may think ur adieu+forty combo is better but it really isn't, how often is d gonna be the second letter in a word, ask urself. how often is u going to be the last letter? how often is r going to be the 3rd letter? this is why peony+trial is objectively vastly better than these supposedly 'mathematical' algorithms, the letters in this words are in their probabilistic locations relative to the locations within the set of 5 letter words.
Optimal 'Algorithm'
Ok so I've now taught u how to play wordle like a basic bitch. You can win 100% of games with these methods, however in all likelihood u will end up with a win rate closer to 95-97/100 games. How, you ask, do you get 100% accuracy?
To get 100% accuracy you need to completely discard the vowel based strategies we've already mentioned, like starting with adieu. Peony+Trial is definitely objectively the best way to begin a vowel based algorithm, someday your mathematics will demonstrate it, but for now just trust me. However, even this optimized method has to be completely set aside to get 100% accuracy. To get 100% accuracy we need to start with words that use as few vowels as possible, 4 consonants and 1 vowel words are the basis of this method that allows us to find the structure of the word in as few guesses possible, saving us guesses for situations when many possible letter combinations could fill in a word.
The main reason the vowel method has to be discarded is because vowels, for the 99.9 percentile lexicon wordlers that are capable of 100% accuracy over 100+ games, are basically entirely deducible from the consonant structure. Thus, the strategy for achieving 100% accuracy recognizes that finding the consonants is a much higher priority than vowels when facing extremely difficult word structures (i.e word structures that require trial and error to solve).
Last edited:
that's awesome but i have done lots of research on my own and the optimal starting word is actually "boobs" 
yeah i understand current wordle scientists still think 'doggo' is a top 10 starting word, but I'm trying to advance the meta over here.that's awesome but i have done lots of research on my own and the optimal starting word is actually "boobs"
I think that Mew isn't the best start, since most mons have less than 600 BST. This means that 100 is on the high end in the grand scheme of things, when you probably want a number that can cut the possibilities in half. Glalie and phione seem better with their 80s. Ditto's 48s are probably too far in the other direction, however.
let this thtread die clearly nobody plays woeedle anymore
this blatant attempt to antagonize me has been noted
owned
glad this thread is alive again
the most fun i've had in wordle recently is with scams. a scam is when you make a suboptimal guess but it pays off due to sheer luck and it ends up being more efficient than the optimal strategy.
suboptimal guesses are fun because if they don't pay off, you need to git gud in order to be consistent, and if they do pay off you get to laugh like a maniac. it's even more fun with friends. one time my friend challenged me to beat him in squirdle daily (he got it in 3 guesses) and i decided to start with deino, i learned very quickly that the secret pokemon was guzzlord. so many great memories.
in regular wordle i was starting with the word "pixie" for a while hoping to get some fun scams but it never really paid off for me unfortunately. i think constantly starting with the same word/system is against the spirit of scamming anyway, it's all about getting a weird slice of information that you're unfamiliar with, and running with that.
the most fun i've had in wordle recently is with scams. a scam is when you make a suboptimal guess but it pays off due to sheer luck and it ends up being more efficient than the optimal strategy.
suboptimal guesses are fun because if they don't pay off, you need to git gud in order to be consistent, and if they do pay off you get to laugh like a maniac. it's even more fun with friends. one time my friend challenged me to beat him in squirdle daily (he got it in 3 guesses) and i decided to start with deino, i learned very quickly that the secret pokemon was guzzlord. so many great memories.
in regular wordle i was starting with the word "pixie" for a while hoping to get some fun scams but it never really paid off for me unfortunately. i think constantly starting with the same word/system is against the spirit of scamming anyway, it's all about getting a weird slice of information that you're unfamiliar with, and running with that.