Selecting Seed Words for Wordle using Character Statistics
Nisansa de Silva
TL;DR
The paper tackles selecting effective Wordle starter words by leveraging character-frequency statistics over five-letter words to maximize coverage of distinct letters. It builds a frequency-based scoring pipeline: construct the all-word set $A$ from a seed list $G$, compute the character frequency map $F$, restrict to words with five unique characters forming $W$, assign word values $M(w)=\sum_{c\in w}F(c)$, and use greedy and non-overlapping filters $B$, $S$, $M'$ and processing function $P$ to produce an optimal triple. Key findings include a final triple ['raise','clout','nymph'] that achieves broad letter coverage while remaining Wordle-acceptable, with cross-validation from an auto-solver linking to the approach’s validity. The work highlights practical considerations such as Wordle’s accepted word lists and the impact of dataset choices on starting-word recommendations, offering a frequency-based, replicable method for starter-word design with potential extensions to dynamic lexicons. $M(w)=\sum_{c\in w}F(c)$, $F(c)$ being the character-frequency map, and $I_{w_1,w_2}=\exists c\in w_1\cap w_2$ are central to the approach.
Abstract
Wordle, a word guessing game rose to global popularity in the January of 2022. The goal of the game is to guess a five-letter English word within six tries. Each try provides the player with hints by means of colour changing tiles which inform whether or not a given character is part of the solution as well as, in cases where it is part of the solution, whether or not it is in the correct placement. Numerous attempts have been made to find the best starting word and best strategy to solve the daily wordle. This study uses character statistics of five-letter words to determine the best three starting words.