This study presents a comparative analysis of bar chart configurations and their effectiveness in highlighting minimum or near-minimum values across diverse dataset characteristics and communication and interaction objectives. Through the design and deployment of a web-based experimental application, over 100 users participated in visualization tasks under controlled variable conditions. The resulting data were analyzed using association rule mining techniques to identify patterns that link specific visual attributes (such as axis orientation, spacing, color schemes, or labeling) with user response times and task performance when detecting data minimum. The findings provide empirical evidence on how chart features should be adapted according to data volume, variability, and the communicative goal of emphasizing the lowest values. A set of practical recommendations is offered to guide the design of bar charts that optimize interpretability and reduce cognitive load in tasks focused on identifying low-ranking data points.