Wordle is a popular puzzle currently offered daily by the New York Times. Players try to solve the puzzle by guessing a five-letter word in six tries or less, receiving feedback with every guess. Wordle continues to grow in popularity and versions of the game are now available in over 60 languages.

In this paper, the scientific computing method is used to establish a wordle decision model by combining the statistical data of authoritative English corpus and a wordle solution model based on heuristic algorithm. Based on this, we propose the HSW model to fit and analyze the real game situation of wordle players, so as to make suggestions for designers of wordle.

Based on the previous social research reports and the principle of normal distribution, we summarize and propose a VP model that can describe the vocabulary size of the population. Based on the wordle game rules, we pioneered an heuristic solution strategy with multiple controllable variables. Combined with the previous VP model, we optimized the algorithm used and built a large-scale simulation wordle game model. Through continuous parameter adjustment,we obtained a HSW model that fits the real wordle game situation. Based on the HSW model, we discuss its structural characteristics to explain the different reasons for the results of wordle games in reality. In order to describe the difficulty of a given solution word in wordle, we establish a difficulty evaluation model based on the average number of guess rounds. Finally, we compare the attributes of possible words with fixed irrelevant variables to determine the influence of specific attributes on word difficulty.