Automatically Detecting Amusing Games in Wordle

2025-06-04

AI Summary

This study investigates the computability of humor in Wordle gameplay, aiming to automatically predict instances that Reddit users deem “funny.” Leveraging 80,000 user-reaction posts collected via Reddit web crawling, we engineer game-state features—including guess patterns, entropy, and failure types—and employ GPT-3.5 for few-shot annotation of humorous responses. A logistic regression model is trained to quantify feature importance. Our analysis provides the first empirical evidence that Wordle-related humor is weakly but statistically significantly predictable (p < 0.01). Key features—particularly guess patterns and entropy—exhibit robust associations with user amusement. The work contributes a reproducible, quantitative framework for modeling humor in gamified contexts, introducing novel, interpretable feature dimensions grounded in game mechanics and information theory.

Technology Category

• Humans and AI: Game Design — Procedural Content Generation & Storytelling
• Natural Language Processing: Lexical Semantics and Morphology
• Cognitive Modeling & Cognitive Systems: Computational Creativity

Application Category

• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Economics, Online Markets and Human Computation: Cost models of using LLMs in production systems
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

We explore automatically predicting which Wordle games Reddit users find amusing. We scrape approximately 80k reactions by Reddit users to Wordle games from Reddit, classify the reactions as expressing amusement or not using OpenAI's GPT-3.5 using few-shot prompting, and verify that GPT-3.5's labels roughly correspond to human labels. We then extract features from Wordle games that can predict user amusement. We demonstrate that the features indeed provide a (weak) signal that predicts user amusement as predicted by GPT-3.5. Our results indicate that user amusement at Wordle games can be predicted computationally to some extent. We explore which features of the game contribute to user amusement. We find that user amusement is predictable, indicating a measurable aspect of creativity infused into Wordle games through humor.

Problem

• Predicting amusing Wordle games from Reddit reactions
• Extracting features to forecast user amusement
• Measuring creativity in Wordle through humor detection

Innovation

• Scraping Reddit reactions to Wordle games
• Classifying amusement with GPT-3.5 few-shot prompting
• Extracting predictive features from Wordle games