Deep Learning models have shown remarkable performance across multiple domains, yet their lack of interpretability remains a significant challenge. In this work, we propose the Explainability-Performance Coefficient (EPC), a novel metric that quantifies the trade-off between identifying the most influential input features and preserving model performance, where a higher EPC implies an improved balance between these two critical aspects. Our results show that model-specific explainability techniques yield higher EPC values than other methods such as feature selection. Furthermore, when combined with model-based approaches, regularization significantly improves explainability by effectively reducing the number of relevant features without compromising performance.