Complete Statistical Evidence Calculator

Calculate p-fr-nb triplets for 2×2 independent binary outcome trials

2×2 Table Calculator

Enter your trial data to calculate complete statistical evidence (significance, fragility, and robustness).

	Events	Non-events
Group A (Treatment)
Group B (Control)

Example: Treatment vs Control

A clinical trial comparing a new treatment to standard care:

Group A (Treatment): 15 events, 85 non-events
Group B (Control): 25 events, 75 non-events

Enter: a=15, b=85, c=25, d=75

Why Complete Evidence?

Traditional reporting provides only p-values and confidence intervals — this is partial evidence. Complete statistical evidence requires three orthogonal dimensions:

Significance (p): Compatibility with null hypothesis
Fragility (fr): Classification stability (MFQ)
Robustness (nb): Distance from neutrality (NDI, RQ)

Only when all three dimensions align do you have truly convincing, replication-ready evidence.

Read Full Documentation

Note: The FI used here is a modified version that allows bidirectional moves (event to/from non-event) and defaults to the arm with the fewest subjects when the number of events is tied. Toggles are allowed in one treatment arm only. For the original Walsh 2014 FI definition, see the Walsh 2014 FI Calculator.

Citation: Heston TF. Fragility metrics toolkit v6.0.0. Zenodo. 2026. DOI: 10.5281/zenodo.17254763