A Wordle Shannon entropy strategy developed by researchers at Binghamton University and the State University of New York solves the puzzle 99% of the time, outperforming the common letter-frequency approach by nine percentage points.
The research, published in the Northeast Journal of Complex Systems, applies information theory to one of the world’s most-played daily word games. In The New York Times game, players have six attempts to identify a secret five-letter word. Each guess earns colour-coded feedback: grey for a letter absent from the word, yellow for a letter present but misplaced, and green for a letter in exactly the right position. The challenge is deciding which words to guess when the board is still mostly blank.
How the Wordle Shannon Entropy Strategy Works
Shannon entropy, as the Times of India explains, is a mathematical concept developed by Claude Shannon in the late 1940s to measure how much uncertainty exists in a system. The Binghamton team adapted it from that static scientific context into a dynamic decision-making tool, selecting each new guess based on how much it is likely to reduce the pool of remaining candidate words.
Rather than asking ‘which word is most likely to be the answer?’, the method asks ‘which word will give me the most useful information, whatever colour pattern comes back?’ The distinction matters. A guess that looks unlikely to be the secret word can still slash the number of remaining possibilities far more efficiently than a guess that simply happens to share letters with common English words.
‘Let’s say you’re at a certain guess. The previous guesses will eliminate a whole bunch of options, and based on the remaining options, guessing some words will send you into a trajectory where information gain is speedier,’ said Assistant Professor Congyu ‘Peter’ Wu, who led the research team.
Doctoral student Donald Stephens put the underlying logic plainly: ‘A guess doesn’t have to be the most likely answer; it simply has to be informative. By applying Shannon entropy, the objective shifts to maximising the expected reduction in uncertainty rather than the probability of being right. In practice, this approach can lead to solving the puzzle in fewer guesses.’
Results Against a Traditional Approach
The team tested their method against a more conventional strategy built around guessing words containing common letters such as ‘A’, ‘E’, and ‘R’. In simulations, the entropy-based approach solved 99% of Wordle puzzles, while the traditional method solved just 90%. That gap of nine percentage points may sound modest, but across the millions of daily players the game attracts, it represents a very large number of failed boards.
The full author list, recorded in the Binghamton University Institutional Repository, is Talal Aladaileh, Donald Stephens, Mallak Alqaisi, and Congyu Wu. Their paper sets out the mathematics in detail, with charts, across its published pages in the Northeast Journal of Complex Systems.
There is a practical catch worth naming. Using the method in real time is not as simple as remembering a good opening word. As the team explained, a player would need to run a script or programme alongside the game, entering the colour-coded feedback after each guess and letting the programme output the optimal next word. For most casual players, that is a meaningful barrier. The research is, at its core, a demonstration of how far information theory can push performance on a constrained combinatorial problem, rather than a tool that slots easily into a morning routine.
Wu described the creative value in those terms: ‘It transformed a static measurement (Shannon entropy) in a scientific domain into a dynamic solution that helps accomplish a popular task better, which showcases the team’s deep understanding of class material and their talent as engineers.’
The paper is available in full through the Binghamton University Institutional Repository, where the methodology and simulation data are set out for anyone wanting to look at the detail behind the 99% figure.
