@@danquaylesitsspeltpotatoe8307 I have gotten it in 3, it's unusual but you can get all 5 letters in the first couple words, then there is no need to use the others.
@@ArchimedesLP Very lucky using this method! So since new york post has purchased the word wordle (but not legally) and are not legal owners of the game (I wonder if they think they are) How many other sites offering the same game (that wont charge) will pop up! If the owners of the PATENTED idea of using colours to inform the play of the bulls and cows will sue new york post for breach of their patent? 😁
I disagree with wasting steps by not reacting to results of prior guesses. I use a starter word with 3 common consonants and 2 different vowels, like later or sonar, then I can usually get it in 4, and often in 3. One time in 2!
@@danquaylesitsspeltpotatoe8307 Lots of misinformation in your comment. The color coding was lifted from MasterMind anyway, and it was the New York Times, not Post. The Times said they will keep it free, but I feel certain they will monetize it somehow, or limit it to their subscribers. That will be a sad day.
I also haven't found a word that I couldn't solve with this strategy, but this just made the game trivial and unfun. The thrill of not being sure of your guess and sometimes getting it as early as on the second guess (happened once) is the real appeal of Wordle for me, not showing off your vocabulary.
I use the crane strat in the regular wordle but my sister introduced me to an offshoot called quordle in which you have 4 words to guess and have 9 tries to get it. Decided I’d use this strat for that and since it’s four words you’re guessing, maybe you have one you can guess right on turn 5 but from there is where you’ll probably have to be sharp
@@formerlyknownaseasrob - Lol... I've been ignoring this video for a few days. One, because I play on Hard Mode, so this doesn't work for me. And, two, because the point isn't to guess the word. The point is to guess the word in the lowest number of turns you can, and a 5 turn average is terrible. The only reason I finally watched it is because I discovered Quordle yesterday, and that really does seem to need a "hack" like this if you want to finish before the last guess.
The problem is, if you do it this way, it’s almost always going to take at least four guesses, and part of the fun is doing it in as few guesses as possible. Also, if you’re playing hard mode, this strategy doesn’t work. It’s also boring if you make it that easy.
Yes, if you're concerned about these things you can use this strategy as a backup though. For example, using EARNT CLIMB HOWFS PUDGY, if you hit a good amount of letters on word 1 or 2, you can just use a normal strategy from that point. But you still have the security of efficient words for the rare cases where you hit nothing.
I just started playing and love Wordle. Playing one word a day didn’t satisfy my interest . Thanks for making it fun by guessing more words per day and your hints.
This is really interesting to me. As you mentioned, using a standard set of opening words results in much faster solves at the cost of using more guesses. The best place to use this strategy is in head-to-head Wordle variants where you're simply trying to be the first to guess the word. In these situations, I preferred using TREAD SOILY BUNCH because you get common letters first (a, e, t, r, s, i, etc.) and it's much faster than typing four words. I did come across someone using BRINY STOCK GLADE WHUMP which I think is a stronger four-word opener than yours in terms of letter placement, but it doesn't frontload common letters. This becomes especially important when you're dealing with ~5 second head-to-head matchups where having an extra second or two to think -- particularly if your opener is locked in muscle memory -- can make the difference between victory and defeat. In short, four word openings are great but three words can be faster, especially if you put common letters first. Edit: funnily enough, I just raced someone using your opening and won. All they had time for was TUBES FLING which gave them two green letters, while I had already gotten all five greens. Granted, the word was TRIAD and I got a bit lucky, but I think this is a good case in point. They would have only gotten two greens and three yellows if they had finished all four guesses.
Interesting! Speed wordle as a competition is something I hadn't even really considered. By the way, I started using EARNT CLIMB HOWFS PUDGY for the exact reason you stated: better front-loading of common letters. There are many more factors to consider here. Someone else pointed out that it's better to put S first instead of last because wordle has a curated solution list of 5 letter words that largely doesn't contain plurals. But I find the "general" case of what would be a good strategy for working with an expanded solution list to be much more interesting.
@@ArchimedesLP I've been playing race mode on Metzger Media's Wordle Archive (I don't think I'm allowed to put the link here). It's not super competitive, but the best strategies I've seen all involve guessing lots of letters and going from there. You can check it out if you'd like. As for letter placement, I have heard about the lack of plurals in Wordle's answer list but I agree that your words would be better with a more expanded set of answers. I think EARNT might be better substituted for another word that ends in A_E because it's more likely to turn up green, but unfortunately I don't see an easy way to do that with the words you have. The other letters are placed well, though. In particular, I like the DG_ ending and the placement of N and P.
After less than a minute of the video, I tried it out with those words and it worked every time. I wanna say I tried like 5 times and it worked every time!!
You’re just using 20 unique letters, it not even that bad. Even using 10 unique letter for the first 2 words can increase your chances of getting it in 3 guesses dramatically.
@@aurorazoe6011 - The point of the game isn't to simply get the answer. The point is to get the answer in the least number of guesses. 4 is average, 3 or less is good, 5 is bad. Using this method is going to get you an average close to or over 5. I guess you're free to enjoy being deliberately bad at the game, but I don't think that's what most people want.
@@alwaysdisputin9930 A hole in 1 in golf isn't 100 completed games with no triple bogies. Your first triple bogie doesn't stop you from future holes in 1. Hole in 1 is a per-hole term. Getting it in one guess is exactly a hole in 1.
@@SirRebrl You're taking the term too literally. "Hole in 1" is merely a metaphor for good play. Sasser is saying if you don't guess the word early you're subpar. However, his attitude is suboptimal & will lead to defeats E.g. say 3 guesses reveal 4 green: B L A _ E Sasser will go for the early victory & guess e.g. BLAME This is suboptimal because it could be BLADE, BLAZE or BLARE. He could lose the game going through each possibility in turn. Therefore it's better to guess MAZED & be guaranteed victory in round 5 or 6
@@alwaysdisputin9930 Hole in 1 is a very literal term, and "100 wins with no losses" is too drastically different in just the very basis of it for hole in 1 to be apt. That aside, Sasser isn't the one who suggested that _getting the answer in one guess_ should be "hole in 1" - AR Hemant is. And neither of them is saying "if you don't get it _early_ you're subpar" (nice edit there, just in time lol). Again literally, it's not subpar if you're at par, and from what I see, reasonably skilled players average on 4 guesses. Which defines par. If, in golf, you are still far from the hole after 3 shots and par is 4, you can try to hit the ball hard and straight to finish the par 4, but maybe you land in the sandtrap past the green. Or you can softly lob aiming for the near green to secure a better chance at a bogey. None of this affects the definition of par, or birdie, or bogey, or hole in 1.
I am older 78 and with your word list I won 87 times in row. so far . Such fun , very often the 3 letter words are tough but then I just test JKVQXZ words and the clue of the letter position and Then for an old guy then I search 5 letter word list goggle showed me .thanks so much it fun to do the puzzle every day. Thank you for the four words
I came up with the same strategy, and found it very effective - only my four words are BACON, FUDGE , WRIST and LYMPH. Interestingly, your strategy is superior to mine in the case of TAUNT, I’m left with three possibilities, TAUNT, VAUNT and JAUNT, but only two guesses, but you have T in the first position.
I think most(all?) of these 4-word sets have some 3 word weaknesses like your example. However, I think you can always get it with a clever test: In your example, guessing TROVE on 5 tells you which of the 3 words it is.
I've been using these words for a while now and I just wanted to add that because you know the 4th word is WORDY, you can sometimes guess the word in 4 guesses just by typing TUBES FLING CHAMP so for example take the word VIVID which you'll realize immediately only gives you one clue from TUBES FLING CHAMP but because you've ruled out 14 letters, by process of elimination you know the letters involved are WORDY QZXVJK you can't make a word from WORDYI so you'll look at the QZXVJK list and there are no other options but VIVID Another example: CREAM TUBES FLING CHAMP will give you clues that leave you with 3 possibilities. Or just one if you are a normal person: CREAM the other two options comae and cymae are not common words. But the point is you should be able to get a bunch of solves 4/6 just by thinking of how the hints fit in with the word WORDY
Interesting! To be honest I'm surprised there are so many situations where there's only one realistic option left after typing the first 3 words. Although it takes quite a bit more time and mental effort to do this.
I have been using a similar strategy. I start with the following three words: LOVES, GAUZY, and BRINK. These words have 15 different letters and all the vowels. If needed, my next word is CHxMP. You knock out an additional 4 letters and leave a spot for a vowel. The idea is use a different vowel (A, I, O, or U) depending on what you need. If needed, I use a fifth word SWIFT to knock out another 3 letters. I like your words and I’ll definitely give it a try. Thanks!
Thanks! Yeah, with my strategy you have to be very specific about not testing letters like V, Z and K because with so many guesses used on scouting, you need to make sure to collect a lot of information. Even though words like KAYAK and PIZZA can be a bit tough to figure out, I haven't found a situation where it's not possible to deduce the presence of these letters.
I do the same. MILKY, HEART, SOUND which covers all the vowels + Y (same as yours) and covers the "combo consonants" (L, N, H). Your list has V, G, Z, B my list trades those for M, T, D, H. I think the "Best" list is subjective. The word list should contain the letters (and preferably the correct position) that a person is likely to forget to properly use in regard to spelling rules. So it would make sense that different folks develop different preferred lists. I did not have as much luck with a 4 word list, but I will try the words listed in this video and see how I do.
@@markschnittker9668 Sorry, I just found out your comment was auto-hidden by youtube and have restored it. Good thoughts on the letters used and especially positions occupied. I have started using EARNT CLIMB HOWFS PUDGY which does a pretty good job of putting common letters first, so I can switch strategies early if i get a lot of letters.
I always start with RAISE. If that doesn't give me 2 vowels, I go to POUTY. I wouldn't want to waste 2 more guesses using other standard words. I've also found that using a random starting word works well. I'm a 98% winner!
One minor tweak to your four words: When a five-letter word has the letter B, the B is most often the first letter. So, change TUBES and FLING to BITES and FLUNG. Also, there are always going to be tough words regardless of which four words you start with. But I'd imagine the plan for that possibility is to anticipate the likely problematic words. For example, if the only clues you get from your four words are a correctly placed Y and a misplaced A, you appear to pull the answer out of your butt and enter JAZZY next. 😉
Yes, there are some improvements to be made on the exact word choice. Another thing I've done is put the most common letters earlier, so that sometimes I only need to enter 2 or 3 words to get all 5 letters. For example, EARNT CLIMB HOWFS PUDGY. As to the words that appear tough, they are actually surprisingly easy to find! Other examples include KAYAK, PIZZA, and VIVID. In the example you cited, it' particularly easy because the placement of the Y is known. It looks fishy because there is so little "information" known but the fact is that every other letter has to be a repeat(A or Y) or QZXVJK. I believe there's only one word that fits these clues. By the way, I put another example of this in the thumbnail.
Checked this with my code (you can see it on my github page- user is pulsie and repo name is 4-word-wordle-algo-prover, cant post link because spam filter), 18 words where there are 3 or more options afterthese 4 words- still pretty good but slightly worse (could be easier for a human to find words afterwards tho!) - one unique ex is taunt/jaunt/vaunt As for earnt climb howfs pudgy, it's far worse- having 41 failures (the highest i've found so far by a lot), but I figure it might pay off with faster sols on average [and its still 41 out of 2k possibilities which is still absurd)
@@ashlynwoods8464 I replied in more detail to your top-level comment, but this is a much better example of failure than the other one you gave, because entering one of taunt/jaunt/vaunt will never give you clues to distinguish between the other two. Even in this case it's still solvable with perfect play, 5th guess is any word starting with T that contains either J or V(e.g. TROVE), this will always give enough clues to distinguish between the 3.
Interesting idea, a very stable strategy! When I tried coming up my own four words, assisted by statistics, I ended up with: roate, linds, chump, gleby.
Nice! I wonder about having the e in the last word though. I came up with earnt, climb, howfs, pudgy. I'm pretty happy with this although I wish the s was in word 1 or 2.
@@ArchimedesLP yeah, the repeated characters are weird. There are even more of them in my German Wordle word sequence. But the words are picked such that they split up the possible solutions most evenly. In other words, such that the expected remaining words are fewest
I tried this method after learning the four words and I've knocked out like 10 words in a row immediately following the starters, wow this method works well.
Glad you liked it! There are plenty of other word sets to try, I started using EARNT CLIMB HOWFS PUDGY which does a better job of front-loading common letters.
Everyone’s very critical of this but I quite enjoy it for repeatable wordle clones. It’s cool enough imo just racing through every time getting the word in a minute or less
I have solved Wordle using 2, 3, and 4 lines with this strategy, trying the words in a different order. It's simple: as soon as you have all five letters, either in green or yellow, you stop using the initial four words and you start guessing in the usual way.
I realized pretty quickly my favorite way to play was to basically do this, I have 4 words I use as you do: Comet, Flash, Brink, Pudgy, (using K instead of W) but if I'm not in a rush I'll do the first two words and see if I can solve it and if I easily think of more than 1 possibility or have uncovered very little I'll add the next word and repeat step...example: one day the word was Chest, and I had 2 green and 3 yellow squares by guess two.
This strategy has a couple flaws though, it only gives you two guesses to figure out the order of the letters you've uncovered. Also, it makes it near impossible to complete in less than 4 attempts. Even just using a single good starting word, I've averaged 3 guesses and have yet to fail a single day
I like your words. I do similar and even though I don't get 20 unique letters in my first 4 guesses, I think I have better coverage of mot common letters in my first couple words so I probably have a higher chance of getting the answer in 3 or 4 guesses. You suggested L and H for the second spot and my experience shows that vowels are often in 2nd position. Also, in terms of S at the end, the site I use doesn't use plurals so I switched my S elsewhere. I agree with most of the rest of what you wrote and have big respect. Thank you.
Thanks! I have since switched to EARNT CLIMB HOWFS PUDGY which has better front-loading of common letters but is still probably far from optimal in terms of placement(particularly the S for the reason you stated.)
@@ArchimedesLP thanks for responding. Here are my words.... SPOILER ALERT ... ... ... ... ... ... ... ... ... STARE FLICK MOUND (can reverse 2nd and 3rd since 3rd has more common letters, but for some reason I like having the vowels in AEIOU order). And then, depending on what I learn, I may already have enough info to make an educated guess. Or if I don't, then my options for the 4th word are usually HAPPY or HIPPY or BUGGY or BOUGH (aside from trying to get those next 3 most common consonants, if the vowels matched but wrong position then I'm testing if they might go in the 2nd position). The site I use has stats and I've had: 2 tries: 1% 3 tries: 8% 4 tries: 39% 5 tries: 36% 6 tries: 17% and that includes a lot of not serious attempts and silly mistakes (like typing in 3rd word too quickly without noticing first 2 words caught all 5 letters). PS - I don't like HOWFS since it's not a common word.
flame brick podgy and shunt is also really good. Edit: and now that i think about it, its probably better because it covers the more common letters, making it more likely to give results.
I prefer to test W over K because I feel like it's easier to "detect" the presence of K without testing it. But I can definitely see it going either way!
Also flame shunt brick podgy has better placement and is decent at frontloading common letters Plus using flame shunt brick podgy on speedle took me 103 seconds while tubes fling champ wordy took 178
Well congrats on this video blowing up lol, a lot of people use this set now in fact, somebody with crazy typing speed managed to use this set and get under 19 seconds (for 10 words) which is pretty awesome as a wordle setmaker though i cant help but call it unoptimal because of letter placements (the set I use for running is also unoptimal but I'm learning a better one)
Yeah, it's way less than optimal. I quickly threw this together just when wordle was starting to become popular back in January. This is literally the first set of 4 words I came up with back then. It's been really interesting seeing my wordset pop up in other places. Another one that a ton of people seem to be using is FLAME BRICK SHUNT PODGY, I wonder where this one originated? And then there is the 5-word set popularized by DisguisedToast(I don't know if it's his creation though.) Not too long after making this video I started using EARNT CLIMB HOWFS PUDGY which does a better job of front-loading the more common letters but it's still far from optimal, especially since having S in the last spot isn't useful for wordle proper.
@@ArchimedesLP Yeah S in last spot and E in first spot really isn't great. But some of the other positions are actually really good! (A, R, N, T, C, L, I, D, G, Y, and probably O, U, P). FLAME SHUNT BRICK PODGY is basically the most popular speedrunning set out there, and it is a set optimized for the maximum amount of greens instead of guaranteed solves. SPLAT MIRED GONCH is what I consider the best (it's 3 words, which is faster but harder than 4). And there are a few 5 word sets designed to get 24/25 unique letters, but they all suck and are almost at the same level as the best 4 word sets lol. On hellowordl/speedle there's a way to always guarantee you solve it with 5 words, but for the nyt list versions you still have 16 words at most that you can't distinguish. VERGE SPILL CHOTT FAWNY DUMBO (yeah it's stupid looking but it's really good somehow. Also there's no K lol). This could change though, it's the best set we know of and there could be better out there. Compare that 16 to GLENT BRICK JUMPY VOZHD WAQFS which has 25 unique letters and has around 80 words it can't distinguish.
Thanks! I switched to EARNT CLIMB HOWFS PUDGY to put more commonly used letters first, this gives a greater chance to get enough letters early to use a different strategy, but I really like having this as a backup in case the first 1 or 2 guesses give nothing.
I always start with ADIEU. It frees up room to have fun, while also giving you lots of information. Even if there's nothing from ADIEU, you know the only vowel it can have is O.
I use eg SHORT first then you can choose ADIEU pragmatically based on the results ... you will generally need a word with O and common consonants straight after ADIEU ... but you cannot read a word of only vowels whereas you can re-arange 3 consonants to mouth five letter words ... so best try for the common consonants first. I sometimes try BROTH first if we have had some Ss in the last few answer.
I usually start with AUDIO and guess from there. I didn't like wordle when I first tried it, but find myself playing every day. It's hard enough to challenge yet easy enough to solve, and since there's only one a day, you cant waste your time playing over and over. I did write down your 4 words though! Just in case I get stuck one day
Keep in mind it's too late once you get stuck. This strategy only works if you do it from the beginning. Although you could do a synthesis... start with EARNT and if you get some letters just play normally. If you get nothing follow up with CLIMB HOWFS PUDGY. Good luck!
Don’t know why people are getting upset at this vid. You’re just giving them advice on what to do if they WANT to. Nobody is making them actually do these things. If they find it to easy, they don’t have to do it. Good vid.
My standard plan is... 1. irate 2. sound 3. lymph Or... 1. irate 2. longs 3. duchy I find it entertaining to play a hybrid style, meaning that I deviate from this plan, by looking for particular letters I want to target. This allows me to get many words in 3 and 4 tries, and I think in a couple of hundred attempts, I have only failed once, and that was entirely me having a brain cramp and no fault of my method. Anyway, I find these word games fun, and you can play it different ways.
Nice! It worked on my first try 😁 update; it's great, I tried it several times now and I've been able to guess most at the fifth try as expected but sometimes at the third. Changing the words around might not be the best strategy and wordy should still be placed last. The only problem I see is that now that I know this, why struggle to guess in the original way by just plotting possible solutions. I now see what they say, it's spoiled the fun of Guessing! 😁
Now that the game is too easy, you can try hard mode :) It forces you to use words that conform to the clues you've received so far(which usually invalidates this strategy.)
This will be extremely helpful for solving the multiple word variants such as Quordle and Octordle. As others commented, I'd miss the thrill of a solution in three or even two tries.
I have played this game multiple times and multiple ways. I will add to your word choices by coaling that I cam by your theory in an unusual way. I used what I have called a “geographic” method as I eliminated based on their placement on the keyboard. It allows me to really focus on the other parts of the keyboard and minimize looking around. Originally, I had used Earth and audio to cover the vowels but shifted to the geographic word SWEAR , it’s almost the same as earth but it removes the top left corner except Q. Then I go to the other corner and would do POINT but found that my next words geographically were harder. So I switched to POUND…the D at the end was attached to the left corner and now I had only to find a word for C B M N L I. Again originally, I went with LIMBY switched to CLIMB , and then I would throw in FIGHT even though I was reusing the I. The irony is that as a result, and this may be the biggest benefit is that I have memorized the keyboard. And have become a touch typist….ha, eh? Funniest side effect so far. I have also imagined that this may become a competition so sometimes I have sacrificed the virtue of playing for shortest number of rows to fastest. Thinking I could probably never not get it 4-5 rows but a breakneck speed. Not hard when essentially you get to eliminate so may letters. I also enjoy the newer knowledge of seeing where and perhaps why certain letters reside where they do in words. Knowing some of the etymology has helped but the rarity of certain letter combinations (g before m) (P before N) is always quite interesting. Thanks for your take Rob
I got "Civic", which didn't give me a lot of letters to work with from the first four guesses, but I had narrowed it down enough that I could guess it on the fifth. Interesting strategy.
I’m on a 10 day streak with 3 words You don’t need a method, you need logic, I love using crane and then would, it makes the third pretty easy to tell based on groups of consonants that are left
I get the strategy. And thanks for sharing it! However, then you have no guess work with every try. So if the ES comes up as yellow in the first, and then you type Fling, then it is purely strategy and you only leave the guess till the end. I love it when I get it on a second try. I love it when I get loads of letters on the first try. This is ok if you want to get it on the fifth try. But I am going for second or third try as much as I can!
Yeah, this is more of a backup strategy. It will take fewer guesses if you incorporate information, especially if you get a lot of clues in the first 1-2 words.
This seems to work a good percentage of the time. If I don't have most of the letters by then, adding a word with the few remaining letters tends to work. Thanks.
Thanks, glad you like it! Someone in the comments ran a simulation and it works 100% of the time although sometimes you need a clever 5th guess to distinguish between 3 words.
When I do speedle (wordle speedruns), I use CHAVE, STROW, FLING, BUMPY. It's interesting how fling is used in both of these. In normal wordles I use TASER, and then try out some other words like CLOUD, MOULD, PINKY, GIMPY, BUMPY, PINCH
I think the biggest advantage of your strategy isn't just the information you get, but that people tend to be able to guess words that start with a letter. It's much more difficult to come up with words that have certain letters in the middle. With your strategy, you're almost guaranteed to know what the candidates for the first letter are.
Given that most strategies aim to solve in 3-4 guesses, aiming for 5-6 seems like a bad idea. A better variation seems to me to use the most common 15 letters and go from there: snore-tidal; antre-solid; train-soled; trail-sonde; stone-laird all chain with chump, or there is the more popular crane-solid-thump. If you are going for a 20 letter strategy, may I suggest snore-clump-fight-bawdy as an idea for at least having a shot at a few extra faster guesses along the way.
Try using ADIEU as your first word. It tells you all the vowels there are in the word in the first try. If the word doesn’t contain AEI or U, then it has to be O. I noticed that since NYT took over, they have changed the algorithm to avoid words with the most frequently used letters and chose more words with repeated letters. So I had to adjust my strategy accordingly. I am still averaging 3.4 tries.
I talked about this a bit in the video. The problem with this is that you then can't test the remaining consonants without "wasting" guesses on vowels that you already know about. I'm not saying it's a bad strategy, just it's kind of incompatible with what I'm trying to do here(test 20 unique letters in 4 guesses.)
@@ArchimedesLP Statistically, it is a lot easier to nail down one or two letters amongst 5 vowels than 3 or 4 consonants out of the remaining letters in the alphabet. Once you nailed down the vowels, you have significantly reduced the number of words that you have to choose from, especially if it is a word that starts or ends with a vowel. Computers follows a fix set of rules. It is easier to beat it once you figured out the algorithm.
@@gj8550: I agree. I often start with "atone" and "juicy," and then I've nailed down all the vowels plus the two most common consonants, in only two tries instead of four. If I can get 2-3 hits on "atone" I don't bother with "juicy," and then the wordle can be solvable in 2-3 lines.
Quite ingenious! If this were like those chess matches where time mattered, this would be the supreme strategy. I'd like to see a game where two players play for a total of 3 minutes. The clocks starts and the goal is to win as many games as possible. So you just type "tubes champ fling wordy" as fast as you can and if you don't instantly see the answer you just type xxxxx xxxxx and start again. Not as relaxing as regular wordle, but it would be cool to see people who'd gotten really good at it whip off 30 winning games in 3 minutes or whatever.
Speed wordle has evolved quite a bit since I made this video. Another commenter let me know that someone got 10 words in under 19 seconds with my word list. I'm not sure of the exact rules but I don't believe you are allowed to "skip" words like you described, you have to get every one.
A quick change I came up with after watching this was changing WORDY to DORKY as K is more used than W. This makes the remaining letters the least used in 5 letter words = W V Z J X Q. My current strategy uses SALET, CRONY, HUMID since these are not only common letters but most are in their best or near best location.
If you only want to solve it in 5 or 6 moves it's fine. But I like solving it in 3 or 4 goes and this isn't possible with this strategy. It takes the fun out of it.
Interesting. I chose W over K more because of the perceived "obviousness" of various missing letters. Like, it's true that K is fairly common but in many cases it shows up in conjunction with C so that 'clue' will already be present.
The one I've seen that most speed runners use is BRICK SHUNT FLAME PODGY which at the cost of changing to k to a w puts letters in more likely positions (like e at the end, s/sh at the beginning, b at the start, and u and o in the middle) but I don't think there would be a huge difference between the two strategies.
When I made this I just used the first words I came up with. Also at the time I didn't realize that the wordle solution list doesn't use plurals, so I thought it was good to have the S at the end. The list you provided should give more green hits, which would tend to give a significant speed boost.
Not sure if this has been pointed out already, but I think it could help improve your strategy: At 15:36 you say you haven’t found any vowels that work yet. This isn’t really true. You eliminated A, E, and I from the answer, three out of five vowels. This means the word would have to use O and/or U. This leaves you with a good amount of information, especially since you know it ends with D and contains L. I’m not sure how many words would fit those requirements, but I think WOULD is probably one of the first ones you would think to guess; it’s a pretty common word. Hope this is helpful for situations like this (if you aren’t using your 5+ guess strategy)
Hmm, true. This was kind of an awkward section in the video because it was more about talking about the speed differences than me trying to solve it. I have done some other recordings of puzzle games so I'm pretty familiar with the problems of multitasking gameplay with commentary, maybe I should have collected some footage and then recorded commentary over it for this part.
I came up with a similar strategy (but mathematically provable) for the old board game Mastermind (four colored pegs and the code maker has to give you similar hints). With 6 predefined guesses, I can easily solve any pattern on the 7th guess, but when I play for fun, I often win a round of Mastermind in 4 or 5 guesses (most versions of the game give you 10 or 11 tries).
Since in The NY Times wordle plurals are not used, i suggest changing the first word to REDUB and the fourth word to TOWSY. In this way the S has a fighting chance to come up green.
Concerning this list, I would personally lean towards beast and chump rather than tubes and champ, but that is because wordle (traditional) and some of its variants do not allow for plurals which shifts the commonality of s being in the last slot.
"I haven't yet found a word I can't solve with this strategy." I just put this to the test. On guess four I had all of the letters I needed. Those letters were T H I G S. But I couldn't arrange them. The word was "sight." I put in the word "thigs" first, not knowing that it is a real word. It left me with just one guess which I gave up on. It was SIGHT!!!!
Nice! I tried to avoid using "awkward" letters like J and V though because I thought it would be easier to spot them in words where I didn't get much info.
so basically you're hacking wordle with brute force, nice. Clearly the strategy works, but i prefer solving it before the fourth try. Anyway, nice video
Thanks! Yeah, if you're trying to optimize for early solves you might be able to come up with a four-word list that you would only use in the extreme case that you don't get any clues from your first 2 words. It's tough to pick starter words that balance the needs of this strategy with the needs of an "optimal" strategy though.
My first three words are least, mound, and picky. If I can’t guess from that, I use either burgh or brogh. I’ve missed only 4 times out of 182 games. My current streak is 55 games; my longest streak is 65 games.
A sure way to get there in 6 but out of 21 days I have got the amswer twice in 2’ nine times in 3’ five times in both 4 and 5. In no case has it taken 6. This by picking my first word carefully then based on that the second word becomes even more important.
I think this strategy would be really good for a variant I've played called Polydle. You have to solve multiple words simultaneously, so after these 4 guesses, you'll be able to snipe pretty much all the words!
This is the exact strategy I devised only my words are INERT COALS DUMPY FIGHT. I came up with them by ranking the letters in order of frequency and figuring out which words used those from most to least. BEDEW is a backup. Lots of people say this is wrong because it isn’t fun or it isn’t the point of the game. What they mean is it isn’t fun FOR THEM. How childish. There are many ways to play, many “points” to the game. No one is forcing you to play this way. Why must you impose your preferences on others?
I think duplicating both I and T hurts you too much in the long run. Each additional untested letter massively increases the space of words you need to draw from in those cases where you only get 2-3 clues. Agreed on the negative stuff, but you can always rely on people to tell you exactly how they feel in youtube comments :) Strategies like these have proven useful in speedrunning and multi-wordle clones.
