My latest strategy for playing Wordle has been refined over the last several months.

Key disclosures for my strategy:
– I play Wordle in the regular mode.
– I rely a lot on data that I have culled from more than a thousand Wordle games played thus far.
– I use word finder websites.

Since May 2^nd, 2024, I have started using CRANE as the seed word, the same as for the New York times WordleBot. The NYT WordleBot switched from STARE to CRANE earlier in the year. They claim it to be marginally better based on average guesses to solve a Wordle puzzle.

I select my second guess based upon the revelations from the seed word. The second guess is very much context dependent. I have frequency data at the letter level based upon the revelations from the seed world. This data forms the context for the second guess.

The following two examples illustrate the value of using context dependent second guess.
Example I
The seed word reveals A and N (with or without their position information).  In this case letter frequences (based on Wordles 1-1000) are as follows:
T/L (35%-30%), G/I/S/Y (29%-20%), O/M/U/K (19%-10%)

Example II
The seed word reveals R and E (with or without their position information).  In this case letter frequences (based on Wordles 1-1000) are as follows:
O (35%), T/I (25%-20%), P/G/S/H/U/VO/F/Y (19%-10%)

To pick the second guess, when there are anagrams of a word, it helps to select from the anagrams based upon the frequency of letters in positions 1-5. As an example, you may have three options SPLIT, SPILT and SLIPT to choose from.

The selection of second guess also depends upon whether the seed word reveals the exact position of a letter as opposed to just the presence of that letter in one of the 5 positions.  If a seed word reveals only the presence of a letter (not its position), it helps to include it in the second guess.  As an example, if the letters A and N have been revealed without their positions, a good second guess is TITAN.

This may sound quite complex and time consuming. Once you get used to the steps involved, it’s not any more complex. But it may still take more time than playing the game on an “ad hoc” / “gutfeel” basis. There is nothing wrong with this style of playing. The style of play is a very subjective and personal choice. Both styles are effective. One of my friends plays in hard mode and uses more gut feel based guesses. Our score averages over time (say 30 games) are about the same.

Usually after the second guess I have between one to five possibilities. I cross out the words that have already been used since I have that data. This usually leaves me between 1 and 3 possibilities.  Then I consult the Google word usage website (see below) to determine ranking of word usage. The normal choice would be to pick the one that has the highest rank. But it doesn’t always work out since the whole exercise is probability based. Based on my experience, it works out more often than not. If I feel strongly about a lower ranked word, I use it.

This has allowed me to solve most of the Wordles in three guesses with a few twos and fours occasionally.

You can find the words that have already been used by Wordle on the following website:

Past Wordle answers – every solution so far, by date | TechRadar

The Google word usage website URL is as follows
Google Ngram Viewer

Hope this helps. If you have any questions, please send them to me.

The averages of my score over the last 7, 10, 14 and 30 days are as follows: