Error Bars for Distributions of Numbers of Events

TL;DR

The paper critiques current error bar practices for event distributions and proposes a clearer presentation method to better show agreement between expected and observed data.

Contribution

It introduces a new style of error bar visualization that improves clarity in representing the agreement between expectations and observations.

Findings

New error bar style enhances interpretability.

Proposed method reduces misinterpretation of data.

Improves communication of statistical agreement.

Abstract

The common practice for displaying error bars on distributions of numbers of events is confusing and can lead to incorrect conclusions. A proposal is made for a different style of presentation that more directly indicates the level of agreement between expectations and observations.