Guide

THE TOOLBOX / ACCESSIBILITY & DESIGN SYSTEMS

WCAG 3 Contrast Checker (APCA) —
Perceptual Lc Contrast, Polarity Testing & WCAG 2.1 Comparator.

Test foreground and background color pairs with both the APCA (WCAG 3) perceptual lightness contrast (Lc) algorithm and classic WCAG 2.1 AA/AAA ratios on a live typography preview.

Why Polarity Matters in WCAG 3:

Unlike WCAG 2.1, APCA is polarity-sensitive: dark text on a light background yields a positive Lc, while light text on a dark background yields a negative Lc to reflect real human eye contrast perception.

APCA (WCAG 3) Contrast
Lc 104.5
Lc 90+ · Preferred Body Text
WCAG 2.1 Ratio
17.74:1
AA: PASS · AAA: PASS
Large Headline Sample (24px Bold)

Body text at 16px regular weight. APCA evaluates both font size and font weight alongside background polarity to measure true perceptual readability.

12px UI Caption / Metadata · Requires Lc 75+ for crisp legibility.

APCA THRESHOLD GUIDE

APCA Lc Targets by Typography Role.

APCA |Lc|Recommended Use
Lc 90+Preferred for fluent long-form body copy and small UI text
Lc 75+Minimum for standard body text (16px regular)
Lc 60+Content text (18px+ bold or 24px regular) & buttons
Lc 45+Large display headlines (36px+) & icons

Frequently Asked Questions

What is APCA (WCAG 3) contrast and how is it different from WCAG 2.1?

APCA (Accessible Perceptual Contrast Algorithm) is the candidate contrast model for WCAG 3. Instead of WCAG 2.1’s simple luminance ratio (4.5:1), APCA outputs a lightness contrast value (Lc from −108 to +106) that accounts for human eye adaptation, dark-mode vs. light-mode polarity, and font weight/size.

What APCA Lc score do I need for body text vs. large headings?

As a practical guide: aim for Lc 75–90+ for standard 14px–16px body copy, Lc 60+ for larger content text (18px+ bold or 24px regular), and Lc 45+ for large display headlines (36px+).

Why does white text on orange or cobalt look readable even when WCAG 2.1 says it fails?

WCAG 2.1’s math over-estimates contrast on very dark colors and under-estimates white text on mid-tone saturated buttons like orange or blue. APCA aligns much closer to actual human perception—test it with the "Orange Button Test" preset above.